WIP Data area resize tool

CMakeLists.txt

@@ -2,6 +2,6 @@ cmake_minimum_required(VERSION 2.8)
 
 project(vitastor)
 
-set(VERSION "0.8.3")
+set(VERSION "0.7.1")
 
 add_subdirectory(src)

README-ru.md

@@ -58,7 +58,6 @@ Vitastor поддерживает QEMU-драйвер, протоколы NBD и
   - [Метаданные образов в etcd](docs/config/inode.ru.md)
 - Использование
   - [vitastor-cli](docs/usage/cli.ru.md) (консольный интерфейс)
-  - [vitastor-disk](docs/usage/disk.ru.md) (управление дисками)
   - [fio](docs/usage/fio.ru.md) для тестов производительности
   - [NBD](docs/usage/nbd.ru.md) для монтирования ядром
   - [QEMU и qemu-img](docs/usage/qemu.ru.md)

README.md

@@ -58,7 +58,6 @@ Read more details below in the documentation.
   - [Image metadata in etcd](docs/config/inode.en.md)
 - Usage
   - [vitastor-cli](docs/usage/cli.en.md) (command-line interface)
-  - [vitastor-disk](docs/usage/disk.en.md) (disk management tool)
   - [fio](docs/usage/fio.en.md) for benchmarks
   - [NBD](docs/usage/nbd.en.md) for kernel mounts
   - [QEMU and qemu-img](docs/usage/qemu.en.md)

csi/Dockerfile

@@ -18,19 +18,15 @@ ENV CSI_ENDPOINT=""
 
 RUN apt-get update && \
     apt-get install -y wget && \
+    wget -q -O /etc/apt/trusted.gpg.d/vitastor.gpg https://vitastor.io/debian/pubkey.gpg && \
+    (echo deb http://vitastor.io/debian buster main > /etc/apt/sources.list.d/vitastor.list) && \
     (echo deb http://deb.debian.org/debian buster-backports main > /etc/apt/sources.list.d/backports.list) && \
     (echo "APT::Install-Recommends false;" > /etc/apt/apt.conf) && \
     apt-get update && \
-    apt-get install -y e2fsprogs xfsprogs kmod && \
+    apt-get install -y e2fsprogs xfsprogs vitastor kmod && \
     apt-get clean && \
     (echo options nbd nbds_max=128 > /etc/modprobe.d/nbd.conf)
 
 COPY --from=build /app/vitastor-csi /bin/
 
-RUN (echo deb http://vitastor.io/debian buster main > /etc/apt/sources.list.d/vitastor.list) && \
-    wget -q -O /etc/apt/trusted.gpg.d/vitastor.gpg https://vitastor.io/debian/pubkey.gpg && \
-    apt-get update && \
-    apt-get install -y vitastor-client && \
-    apt-get clean
-
 ENTRYPOINT ["/bin/vitastor-csi"]

csi/Makefile

@@ -1,4 +1,4 @@
-VERSION ?= v0.8.3
+VERSION ?= v0.7.1
 
 all: build push
 

csi/deploy/004-csi-nodeplugin.yaml

@@ -49,7 +49,7 @@ spec:
             capabilities:
               add: ["SYS_ADMIN"]
             allowPrivilegeEscalation: true
-          image: vitalif/vitastor-csi:v0.8.3
+          image: vitalif/vitastor-csi:v0.7.1
           args:
             - "--node=$(NODE_ID)"
             - "--endpoint=$(CSI_ENDPOINT)"
@@ -102,7 +102,7 @@ spec:
             - "--health-port=9898"
           env:
             - name: CSI_ENDPOINT
-              value: unix:///csi/csi.sock
+              value: unix://csi/csi.sock
           volumeMounts:
           - mountPath: /csi
             name: socket-dir

csi/deploy/007-csi-provisioner.yaml

@@ -116,7 +116,7 @@ spec:
             privileged: true
             capabilities:
               add: ["SYS_ADMIN"]
-          image: vitalif/vitastor-csi:v0.8.3
+          image: vitalif/vitastor-csi:v0.7.1
           args:
             - "--node=$(NODE_ID)"
             - "--endpoint=$(CSI_ENDPOINT)"

csi/src/config.go

@@ -5,7 +5,7 @@ package vitastor
 
 const (
     vitastorCSIDriverName    = "csi.vitastor.io"
-    vitastorCSIDriverVersion = "0.8.3"
+    vitastorCSIDriverVersion = "0.7.1"
 )
 
 // Config struct fills the parameters of request or user input

debian/changelog

@@ -1,10 +1,10 @@
-vitastor (0.8.3-1) unstable; urgency=medium
+vitastor (0.7.1-1) unstable; urgency=medium
 
   * Bugfixes
 
  -- Vitaliy Filippov <vitalif@yourcmc.ru>  Fri, 03 Jun 2022 02:09:44 +0300
 
-vitastor (0.8.3-1) unstable; urgency=medium
+vitastor (0.7.1-1) unstable; urgency=medium
 
   * Implement NFS proxy
   * Add documentation

debian/control

@@ -18,7 +18,7 @@ Description: Vitastor, a fast software-defined clustered block storage
 
 Package: vitastor-osd
 Architecture: amd64
-Depends: ${shlibs:Depends}, ${misc:Depends}, vitastor-client (= ${binary:Version}), fdisk, util-linux, parted
+Depends: ${shlibs:Depends}, ${misc:Depends}, vitastor-client (= ${binary:Version})
 Description: Vitastor, a fast software-defined clustered block storage - object storage daemon
  Vitastor object storage daemon, i.e. server program that stores data.
 

debian/pve-storage-vitastor.install

@@ -1 +1 @@
-patches/VitastorPlugin.pm usr/share/perl5/PVE/Storage/Custom/
+patches/PVE_VitastorPlugin.pm usr/share/perl5/PVE/Storage/Custom/VitastorPlugin.pm

debian/vitastor-client.install

@@ -4,3 +4,4 @@ usr/bin/vitastor-rm
 usr/bin/vitastor-nbd
 usr/bin/vitastor-nfs
 usr/lib/*/libvitastor*.so*
+mon/make-osd.sh /usr/lib/vitastor

debian/vitastor-mon.install

@@ -1,2 +1 @@
 mon usr/lib/vitastor
-mon/vitastor-mon.service /lib/systemd/system

debian/vitastor-mon.postinst

@@ -1,9 +0,0 @@
-#!/bin/sh
-
-set -e
-
-if [ "$1" = "configure" ]; then
-	addgroup --system --quiet vitastor
-	adduser --system --quiet --ingroup vitastor --no-create-home --home /nonexistent vitastor
-	mkdir -p /etc/vitastor
-fi

debian/vitastor-osd.install

@@ -1,6 +1,3 @@
 usr/bin/vitastor-osd
 usr/bin/vitastor-disk
 usr/bin/vitastor-dump-journal
-mon/vitastor-osd@.service /lib/systemd/system
-mon/vitastor.target /lib/systemd/system
-mon/90-vitastor.rules /lib/udev/rules.d

debian/vitastor-osd.postinst

@@ -1,10 +0,0 @@
-#!/bin/sh
-
-set -e
-
-if [ "$1" = "configure" ]; then
-	addgroup --system --quiet vitastor
-	adduser --system --quiet --ingroup vitastor --no-create-home --home /nonexistent vitastor
-	install -o vitastor -g vitastor -d /var/log/vitastor
-	mkdir -p /etc/vitastor
-fi

debian/vitastor.Dockerfile

@@ -34,8 +34,8 @@ RUN set -e -x; \
     mkdir -p /root/packages/vitastor-$REL; \
     rm -rf /root/packages/vitastor-$REL/*; \
     cd /root/packages/vitastor-$REL; \
-    cp -r /root/vitastor vitastor-0.8.3; \
-    cd vitastor-0.8.3; \
+    cp -r /root/vitastor vitastor-0.7.1; \
+    cd vitastor-0.7.1; \
     ln -s /root/fio-build/fio-*/ ./fio; \
     FIO=$(head -n1 fio/debian/changelog | perl -pe 's/^.*\((.*?)\).*$/$1/'); \
     ls /usr/include/linux/raw.h || cp ./debian/raw.h /usr/include/linux/raw.h; \
@@ -48,8 +48,8 @@ RUN set -e -x; \
     rm -rf a b; \
     echo "dep:fio=$FIO" > debian/fio_version; \
     cd /root/packages/vitastor-$REL; \
-    tar --sort=name --mtime='2020-01-01' --owner=0 --group=0 --exclude=debian -cJf vitastor_0.8.3.orig.tar.xz vitastor-0.8.3; \
-    cd vitastor-0.8.3; \
+    tar --sort=name --mtime='2020-01-01' --owner=0 --group=0 --exclude=debian -cJf vitastor_0.7.1.orig.tar.xz vitastor-0.7.1; \
+    cd vitastor-0.7.1; \
     V=$(head -n1 debian/changelog | perl -pe 's/^.*\((.*?)\).*$/$1/'); \
     DEBFULLNAME="Vitaliy Filippov <vitalif@yourcmc.ru>" dch -D $REL -v "$V""$REL" "Rebuild for $REL"; \
     DEB_BUILD_OPTIONS=nocheck dpkg-buildpackage --jobs=auto -sa; \

docs/config/layout-cluster.en.md

@@ -9,34 +9,34 @@
 These parameters apply to clients and OSDs, are fixed at the moment of OSD drive
 initialization and can't be changed after it without losing data.
 
-OSDs with different values of these parameters (for example, SSD and SSD+HDD
-OSDs) can coexist in one Vitastor cluster within different pools. Each pool can
-only include OSDs with identical settings of these parameters.
-
-These parameters, when set to a non-default value, must also be specified in
-etcd for clients to be aware of their values, either in /vitastor/config/global
-or in pool configuration. Pool configuration overrides the global setting.
-If the value for a pool in etcd doesn't match on-disk OSD configuration, the
-OSD will refuse to start PGs of that pool.
-
 - [block_size](#block_size)
 - [bitmap_granularity](#bitmap_granularity)
 - [immediate_commit](#immediate_commit)
+- [client_dirty_limit](#client_dirty_limit)
 
 ## block_size
 
 - Type: integer
 - Default: 131072
 
-Size of objects (data blocks) into which all physical and virtual drives
-(within a pool) are subdivided in Vitastor. One of current main settings
-in Vitastor, affects memory usage, write amplification and I/O load
-distribution effectiveness.
+Size of objects (data blocks) into which all physical and virtual drives are
+subdivided in Vitastor. One of current main settings in Vitastor, affects
+memory usage, write amplification and I/O load distribution effectiveness.
 
 Recommended default block size is 128 KB for SSD and 4 MB for HDD. In fact,
 it's possible to use 4 MB for SSD too - it will lower memory usage, but
 may increase average WA and reduce linear performance.
 
+OSDs with different block sizes (for example, SSD and SSD+HDD OSDs) can
+currently coexist in one etcd instance only within separate Vitastor
+clusters with different etcd_prefix'es.
+
+Also block size can't be changed after OSD initialization without losing
+data.
+
+You must always specify block_size in etcd in /vitastor/config/global if
+you change it so all clients can know about it.
+
 OSD memory usage is roughly (SIZE / BLOCK * 68 bytes) which is roughly
 544 MB per 1 TB of used disk space with the default 128 KB block size.
 
@@ -50,7 +50,12 @@ of disk_alignment. It's called bitmap granularity because Vitastor tracks
 an allocation bitmap for each object containing 2 bits per each
 (bitmap_granularity) bytes.
 
-Can't be smaller than the OSD data device sector.
+This parameter can't be changed after OSD initialization without losing
+data. Also it's fixed for the whole Vitastor cluster i.e. two different
+values can't be used in a single Vitastor cluster.
+
+Clients MUST be aware of this parameter value, so put it into etcd key
+/vitastor/config/global if you change it for any reason.
 
 ## immediate_commit
 
@@ -94,12 +99,26 @@ unsafe to change by hand). The same may apply to newer HDDs with internal
 SSD cache or "media-cache" - for example, a lot of Seagate EXOS drives have
 it (they have internal SSD cache even though it's not stated in datasheets).
 
-Setting this parameter to "all" or "small" in OSD parameters requires enabling
-disable_journal_fsync and disable_meta_fsync, setting it to "all" also requires
-enabling disable_data_fsync.
+This parameter must be set both in etcd in /vitastor/config/global and in
+OSD command line or configuration. Setting it to "all" or "small" requires
+enabling disable_journal_fsync and disable_meta_fsync, setting it to "all"
+also requires enabling disable_data_fsync.
 
 TLDR: For optimal performance, set immediate_commit to "all" if you only use
 SSDs with supercapacitor-based power loss protection (nonvolatile
 write-through cache) for both data and journals in the whole Vitastor
 cluster. Set it to "small" if you only use such SSDs for journals. Leave
 empty if your drives have write-back cache.
+
+## client_dirty_limit
+
+- Type: integer
+- Default: 33554432
+
+Without immediate_commit=all this parameter sets the limit of "dirty"
+(not committed by fsync) data allowed by the client before forcing an
+additional fsync and committing the data. Also note that the client always
+holds a copy of uncommitted data in memory so this setting also affects
+RAM usage of clients.
+
+This parameter doesn't affect OSDs themselves.

docs/config/layout-cluster.ru.md

@@ -9,19 +9,10 @@
 Данные параметры используются клиентами и OSD, задаются в момент инициализации
 диска OSD и не могут быть изменены после этого без потери данных.
 
-OSD с разными значениями данных параметров (например, SSD и гибридные SSD+HDD
-OSD) могут сосуществовать в одном кластере Vitastor в разных пулах. Один пул
-может включать только OSD с одинаковыми настройками этих параметров.
-
-Данные параметры, отличаясь от значения по умолчанию, должны также быть заданы
-в etcd, чтобы клиенты могли узнать их значение, либо в глобальной конфигурации
-/vitastor/config/global, либо в настройках пулов. Настройки пула переопределяют
-глобальное значение. Если значение в настройках пула не будет соответствовать
-конфигурации OSD, OSD откажется запускать PG данного пула.
-
 - [block_size](#block_size)
 - [bitmap_granularity](#bitmap_granularity)
 - [immediate_commit](#immediate_commit)
+- [client_dirty_limit](#client_dirty_limit)
 
 ## block_size
 
@@ -29,15 +20,24 @@ OSD) могут сосуществовать в одном кластере Vita
 - Значение по умолчанию: 131072
 
 Размер объектов (блоков данных), на которые делятся физические и виртуальные
-диски в Vitastor (в рамках каждого пула). Одна из ключевых на данный момент
-настроек, влияет на потребление памяти, объём избыточной записи (write
-amplification) и эффективность распределения нагрузки по OSD.
+диски в Vitastor. Одна из ключевых на данный момент настроек, влияет на
+потребление памяти, объём избыточной записи (write amplification) и
+эффективность распределения нагрузки по OSD.
 
 Рекомендуемые по умолчанию размеры блока - 128 килобайт для SSD и 4
 мегабайта для HDD. В принципе, для SSD можно тоже использовать 4 мегабайта,
 это понизит использование памяти, но ухудшит распределение нагрузки и в
 среднем увеличит WA.
 
+OSD с разными размерами блока (например, SSD и SSD+HDD OSD) на данный
+момент могут сосуществовать в рамках одного etcd только в виде двух независимых
+кластеров Vitastor с разными etcd_prefix.
+
+Также размер блока нельзя менять после инициализации OSD без потери данных.
+
+Если вы меняете размер блока, обязательно прописывайте его в etcd в
+/vitastor/config/global, дабы все клиенты его знали.
+
 Потребление памяти OSD составляет примерно (РАЗМЕР / БЛОК * 68 байт),
 т.е. примерно 544 МБ памяти на 1 ТБ занятого места на диске при
 стандартном 128 КБ блоке.
@@ -52,7 +52,13 @@ amplification) и эффективность распределения нагр
 потому, что Vitastor хранит битовую карту для каждого объекта, содержащую
 по 2 бита на каждые (bitmap_granularity) байт.
 
-Не может быть меньше размера сектора дисков данных OSD.
+Данный параметр нельзя менять после инициализации OSD без потери данных.
+Также он фиксирован для всего кластера Vitastor, т.е. разные значения
+не могут сосуществовать в одном кластере.
+
+Клиенты ДОЛЖНЫ знать правильное значение этого параметра, так что если вы
+его меняете, обязательно прописывайте изменённое значение в etcd в ключ
+/vitastor/config/global.
 
 ## immediate_commit
 
@@ -102,7 +108,8 @@ HDD-дисках с внутренним SSD или "медиа" кэшем - н
 многих дисках Seagate EXOS (у них есть внутренний SSD-кэш, хотя это и не
 указано в спецификациях).
 
-Указание "all" или "small" в настройках / командной строке OSD требует
+Данный параметр нужно указывать и в etcd в /vitastor/config/global, и в
+командной строке или конфигурации OSD. Значения "all" и "small" требуют
 включения disable_journal_fsync и disable_meta_fsync, значение "all" также
 требует включения disable_data_fsync.
 
@@ -112,3 +119,16 @@ immediate_commit в значение "all", если вы используете
 такие SSD для всех журналов, но не для данных - можете установить параметр
 в "small". Если и какие-то из дисков журналов имеют волатильный кэш записи -
 оставьте параметр пустым.
+
+## client_dirty_limit
+
+- Тип: целое число
+- Значение по умолчанию: 33554432
+
+При работе без immediate_commit=all - это лимит объёма "грязных" (не
+зафиксированных fsync-ом) данных, при достижении которого клиент будет
+принудительно вызывать fsync и фиксировать данные. Также стоит иметь в виду,
+что в этом случае до момента fsync клиент хранит копию незафиксированных
+данных в памяти, то есть, настройка влияет на потребление памяти клиентами.
+
+Параметр не влияет на сами OSD.

docs/config/network.en.md

@@ -29,7 +29,6 @@ between clients, OSDs and etcd.
 - [etcd_slow_timeout](#etcd_slow_timeout)
 - [etcd_keepalive_timeout](#etcd_keepalive_timeout)
 - [etcd_ws_keepalive_timeout](#etcd_ws_keepalive_timeout)
-- [client_dirty_limit](#client_dirty_limit)
 
 ## tcp_header_buffer_size
 
@@ -213,16 +212,3 @@ etcd_report_interval to guarantee that keepalive actually works.
 
 etcd websocket ping interval required to keep the connection alive and
 detect disconnections quickly.
-
-## client_dirty_limit
-
-- Type: integer
-- Default: 33554432
-
-Without immediate_commit=all this parameter sets the limit of "dirty"
-(not committed by fsync) data allowed by the client before forcing an
-additional fsync and committing the data. Also note that the client always
-holds a copy of uncommitted data in memory so this setting also affects
-RAM usage of clients.
-
-This parameter doesn't affect OSDs themselves.

docs/config/network.ru.md

@@ -29,7 +29,6 @@
 - [etcd_slow_timeout](#etcd_slow_timeout)
 - [etcd_keepalive_timeout](#etcd_keepalive_timeout)
 - [etcd_ws_keepalive_timeout](#etcd_ws_keepalive_timeout)
-- [client_dirty_limit](#client_dirty_limit)
 
 ## tcp_header_buffer_size
 
@@ -223,16 +222,3 @@ etcd_report_interval, чтобы keepalive гарантированно рабо
 - Значение по умолчанию: 30
 
 Интервал проверки живости вебсокет-подключений к etcd.
-
-## client_dirty_limit
-
-- Тип: целое число
-- Значение по умолчанию: 33554432
-
-При работе без immediate_commit=all - это лимит объёма "грязных" (не
-зафиксированных fsync-ом) данных, при достижении которого клиент будет
-принудительно вызывать fsync и фиксировать данные. Также стоит иметь в виду,
-что в этом случае до момента fsync клиент хранит копию незафиксированных
-данных в памяти, то есть, настройка влияет на потребление памяти клиентами.
-
-Параметр не влияет на сами OSD.

docs/config/osd.en.md

@@ -17,7 +17,6 @@ initialization and can be changed with an OSD restart.
 - [autosync_interval](#autosync_interval)
 - [autosync_writes](#autosync_writes)
 - [recovery_queue_depth](#recovery_queue_depth)
-- [recovery_pg_switch](#recovery_pg_switch)
 - [recovery_sync_batch](#recovery_sync_batch)
 - [readonly](#readonly)
 - [no_recovery](#no_recovery)
@@ -116,16 +115,6 @@ Maximum recovery operations per one primary OSD at any given moment of time.
 Currently it's the only parameter available to tune the speed or recovery
 and rebalancing, but it's planned to implement more.
 
-## recovery_pg_switch
-
-- Type: integer
-- Default: 128
-
-Number of recovery operations before switching to recovery of the next PG.
-The idea is to mix all PGs during recovery for more even space and load
-distribution but still benefit from recovery queue depth greater than 1.
-Degraded PGs are anyway scanned first.
-
 ## recovery_sync_batch
 
 - Type: integer

docs/config/osd.ru.md

@@ -18,7 +18,6 @@
 - [autosync_interval](#autosync_interval)
 - [autosync_writes](#autosync_writes)
 - [recovery_queue_depth](#recovery_queue_depth)
-- [recovery_pg_switch](#recovery_pg_switch)
 - [recovery_sync_batch](#recovery_sync_batch)
 - [readonly](#readonly)
 - [no_recovery](#no_recovery)
@@ -120,17 +119,6 @@ OSD, чтобы успевать очищать журнал - без них OSD
 для ускорения или замедления восстановления и перебалансировки данных, но
 в планах реализация других параметров.
 
-## recovery_pg_switch
-
-- Тип: целое число
-- Значение по умолчанию: 128
-
-Число операций восстановления перед переключением на восстановление другой PG.
-Идея заключается в том, чтобы восстанавливать все PG одновременно для более
-равномерного распределения места и нагрузки, но при этом всё равно выигрывать
-от глубины очереди восстановления, большей, чем 1. Деградированные PG в любом
-случае сканируются первыми.
-
 ## recovery_sync_batch
 
 - Тип: целое число

docs/config/pool.en.md

@@ -33,9 +33,6 @@ Parameters:
 - [pg_count](#pg_count)
 - [failure_domain](#failure_domain)
 - [max_osd_combinations](#max_osd_combinations)
-- [block_size](#block_size)
-- [bitmap_granularity](#bitmap_granularity)
-- [immediate_commit](#immediate_commit)
 - [pg_stripe_size](#pg_stripe_size)
 - [root_node](#root_node)
 - [osd_tags](#osd_tags)
@@ -82,7 +79,7 @@ Parent node reference is required for intermediate tree nodes.
 Separate OSD settings are set in etc keys `/vitastor/config/osd/<number>`
 in JSON format `{"<key>":<value>}`.
 
-As of now, two settings are supported:
+As of now, there is only one setting:
 
 ## reweight
 
@@ -96,15 +93,6 @@ This means an OSD configured with reweight lower than 1 receives less PGs than
 it normally would. An OSD with reweight = 0 won't store any data. You can set
 reweight to 0 to trigger rebalance and remove all data from an OSD.
 
-## tags
-
-- Type: string or array of strings
-
-Sets tag or multiple tags for this OSD. Tags can be used to group OSDs into
-subsets and then use a specific subset for pool instead of all OSDs.
-For example you can mark SSD OSDs with tag "ssd" and HDD OSDs with "hdd" and
-such tags will work as device classes.
-
 # Pool parameters
 
 ## name
@@ -198,43 +186,6 @@ number of combinations to generate when optimising PG placement.
 
 This parameter usually doesn't require to be changed.
 
-## block_size
-
-- Type: integer
-- Default: 131072
-
-Block size for this pool. The value from /vitastor/config/global is used when
-unspecified. If your cluster has OSDs with different block sizes then pool must
-be restricted by [osd_tags](#osd_tags) to only include OSDs with matching block
-size.
-
-Read more about this parameter in [Cluster-Wide Disk Layout Parameters](layout-cluster.en.md#block_size).
-
-## bitmap_granularity
-
-- Type: integer
-- Default: 4096
-
-"Sector" size of virtual disks in this pool. The value from
-/vitastor/config/global is used when unspecified. Similar to block_size, the
-pool must be restricted by [osd_tags](#osd_tags) to only include OSDs with
-matching bitmap_granularity.
-
-Read more about this parameter in [Cluster-Wide Disk Layout Parameters](layout-cluster.en.md#bitmap_granularity).
-
-## immediate_commit
-
-- Type: string, one of "all", "small" and "none"
-- Default: none
-
-Immediate commit setting for this pool. The value from /vitastor/config/global
-is used when unspecified. Similar to block_size, the pool must be restricted by
-[osd_tags](#osd_tags) to only include OSDs with compatible immediate_commit.
-Compatible means that a pool with non-immediate commit will work with OSDs with
-immediate commit enabled, but not vice versa.
-
-Read more about this parameter in [Cluster-Wide Disk Layout Parameters](layout-cluster.en.md#immediate_commit).
-
 ## pg_stripe_size
 
 - Type: integer

docs/config/pool.ru.md

@@ -32,9 +32,6 @@
 - [pg_count](#pg_count)
 - [failure_domain](#failure_domain)
 - [max_osd_combinations](#max_osd_combinations)
-- [block_size](#block_size)
-- [bitmap_granularity](#bitmap_granularity)
-- [immediate_commit](#immediate_commit)
 - [pg_stripe_size](#pg_stripe_size)
 - [root_node](#root_node)
 - [osd_tags](#osd_tags)
@@ -81,10 +78,7 @@
 Настройки отдельных OSD задаются в ключах etcd `/vitastor/config/osd/<number>`
 в JSON-формате `{"<key>":<value>}`.
 
-На данный момент поддерживаются две настройки:
-
-- [reweight](#reweight)
-- [tags](#tags)
+На данный момент поддерживается одна настройка:
 
 ## reweight
 
@@ -99,15 +93,6 @@
 хранении данных вообще. Вы можете установить reweight в 0, чтобы убрать
 все данные с OSD.
 
-## tags
-
-- Тип: строка или массив строк
-
-Задаёт тег или набор тегов для данного OSD. Теги можно использовать, чтобы
-делить OSD на множества и потом размещать пул только на части OSD, а не на
-всех. Можно, например, пометить SSD OSD тегом "ssd", а HDD тегом "hdd", в
-этом смысле теги работают аналогично классам устройств.
-
 # Параметры
 
 ## name
@@ -200,51 +185,13 @@ PG в Vitastor эферемерны, то есть вы можете менят
 
 Обычно данный параметр не требует изменений.
 
-## block_size
-
-- Тип: целое число
-- По умолчанию: 131072
-
-Размер блока для данного пула. Если не задан, используется значение из
-/vitastor/config/global. Если в вашем кластере есть OSD с разными размерами
-блока, пул должен быть ограничен только OSD, блок которых равен блоку пула,
-с помощью [osd_tags](#osd_tags).
-
-О самом параметре читайте в разделе [Дисковые параметры уровня кластера](layout-cluster.ru.md#block_size).
-
-## bitmap_granularity
-
-- Тип: целое число
-- По умолчанию: 4096
-
-Размер "сектора" виртуальных дисков в данном пуле. Если не задан, используется
-значение из /vitastor/config/global. Аналогично block_size, пул должен быть
-ограничен OSD со значением bitmap_granularity, равным значению пула, с помощью
-[osd_tags](#osd_tags).
-
-О самом параметре читайте в разделе [Дисковые параметры уровня кластера](layout-cluster.ru.md#bitmap_granularity).
-
-## immediate_commit
-
-- Тип: строка "all", "small" или "none"
-- По умолчанию: none
-
-Настройка мгновенного коммита для данного пула. Если не задана, используется
-значение из /vitastor/config/global. Аналогично block_size, пул должен быть
-ограничен OSD со значением bitmap_granularity, совместимым со значением пула, с
-помощью [osd_tags](#osd_tags). Совместимость означает, что пул с отключенным
-мгновенным коммитом может работать на OSD с включённым мгновенным коммитом, но
-не наоборот.
-
-О самом параметре читайте в разделе [Дисковые параметры уровня кластера](layout-cluster.ru.md#immediate_commit).
-
 ## pg_stripe_size
 
 - Тип: целое число
 - По умолчанию: 0
 
 Данный параметр задаёт размер полосы "нарезки" образов на PG. Размер полосы не может
-быть меньше, чем [block_size](#block_size), умноженный на
+быть меньше, чем [block_size](layout-cluster.ru.md#block_size), умноженный на
 (pg_size - parity_chunks) для EC-пулов или 1 для реплицированных пулов. То же
 значение используется по умолчанию.
 

docs/config/src/layout-cluster.en.md

@@ -2,13 +2,3 @@
 
 These parameters apply to clients and OSDs, are fixed at the moment of OSD drive
 initialization and can't be changed after it without losing data.
-
-OSDs with different values of these parameters (for example, SSD and SSD+HDD
-OSDs) can coexist in one Vitastor cluster within different pools. Each pool can
-only include OSDs with identical settings of these parameters.
-
-These parameters, when set to a non-default value, must also be specified in
-etcd for clients to be aware of their values, either in /vitastor/config/global
-or in pool configuration. Pool configuration overrides the global setting.
-If the value for a pool in etcd doesn't match on-disk OSD configuration, the
-OSD will refuse to start PGs of that pool.

docs/config/src/layout-cluster.ru.md

@@ -2,13 +2,3 @@
 
 Данные параметры используются клиентами и OSD, задаются в момент инициализации
 диска OSD и не могут быть изменены после этого без потери данных.
-
-OSD с разными значениями данных параметров (например, SSD и гибридные SSD+HDD
-OSD) могут сосуществовать в одном кластере Vitastor в разных пулах. Один пул
-может включать только OSD с одинаковыми настройками этих параметров.
-
-Данные параметры, отличаясь от значения по умолчанию, должны также быть заданы
-в etcd, чтобы клиенты могли узнать их значение, либо в глобальной конфигурации
-/vitastor/config/global, либо в настройках пулов. Настройки пула переопределяют
-глобальное значение. Если значение в настройках пула не будет соответствовать
-конфигурации OSD, OSD откажется запускать PG данного пула.

docs/config/src/layout-cluster.yml

@@ -2,28 +2,46 @@
   type: int
   default: 131072
   info: |
-    Size of objects (data blocks) into which all physical and virtual drives
-    (within a pool) are subdivided in Vitastor. One of current main settings
-    in Vitastor, affects memory usage, write amplification and I/O load
-    distribution effectiveness.
+    Size of objects (data blocks) into which all physical and virtual drives are
+    subdivided in Vitastor. One of current main settings in Vitastor, affects
+    memory usage, write amplification and I/O load distribution effectiveness.
 
     Recommended default block size is 128 KB for SSD and 4 MB for HDD. In fact,
     it's possible to use 4 MB for SSD too - it will lower memory usage, but
     may increase average WA and reduce linear performance.
 
+    OSDs with different block sizes (for example, SSD and SSD+HDD OSDs) can
+    currently coexist in one etcd instance only within separate Vitastor
+    clusters with different etcd_prefix'es.
+
+    Also block size can't be changed after OSD initialization without losing
+    data.
+
+    You must always specify block_size in etcd in /vitastor/config/global if
+    you change it so all clients can know about it.
+
     OSD memory usage is roughly (SIZE / BLOCK * 68 bytes) which is roughly
     544 MB per 1 TB of used disk space with the default 128 KB block size.
   info_ru: |
     Размер объектов (блоков данных), на которые делятся физические и виртуальные
-    диски в Vitastor (в рамках каждого пула). Одна из ключевых на данный момент
-    настроек, влияет на потребление памяти, объём избыточной записи (write
-    amplification) и эффективность распределения нагрузки по OSD.
+    диски в Vitastor. Одна из ключевых на данный момент настроек, влияет на
+    потребление памяти, объём избыточной записи (write amplification) и
+    эффективность распределения нагрузки по OSD.
 
     Рекомендуемые по умолчанию размеры блока - 128 килобайт для SSD и 4
     мегабайта для HDD. В принципе, для SSD можно тоже использовать 4 мегабайта,
     это понизит использование памяти, но ухудшит распределение нагрузки и в
     среднем увеличит WA.
 
+    OSD с разными размерами блока (например, SSD и SSD+HDD OSD) на данный
+    момент могут сосуществовать в рамках одного etcd только в виде двух независимых
+    кластеров Vitastor с разными etcd_prefix.
+
+    Также размер блока нельзя менять после инициализации OSD без потери данных.
+
+    Если вы меняете размер блока, обязательно прописывайте его в etcd в
+    /vitastor/config/global, дабы все клиенты его знали.
+
     Потребление памяти OSD составляет примерно (РАЗМЕР / БЛОК * 68 байт),
     т.е. примерно 544 МБ памяти на 1 ТБ занятого места на диске при
     стандартном 128 КБ блоке.
@@ -36,14 +54,25 @@
     an allocation bitmap for each object containing 2 bits per each
     (bitmap_granularity) bytes.
 
-    Can't be smaller than the OSD data device sector.
+    This parameter can't be changed after OSD initialization without losing
+    data. Also it's fixed for the whole Vitastor cluster i.e. two different
+    values can't be used in a single Vitastor cluster.
+
+    Clients MUST be aware of this parameter value, so put it into etcd key
+    /vitastor/config/global if you change it for any reason.
   info_ru: |
     Требуемое выравнивание записи на виртуальные диски (размер их "сектора").
     Должен быть кратен disk_alignment. Называется гранулярностью битовой карты
     потому, что Vitastor хранит битовую карту для каждого объекта, содержащую
     по 2 бита на каждые (bitmap_granularity) байт.
 
-    Не может быть меньше размера сектора дисков данных OSD.
+    Данный параметр нельзя менять после инициализации OSD без потери данных.
+    Также он фиксирован для всего кластера Vitastor, т.е. разные значения
+    не могут сосуществовать в одном кластере.
+
+    Клиенты ДОЛЖНЫ знать правильное значение этого параметра, так что если вы
+    его меняете, обязательно прописывайте изменённое значение в etcd в ключ
+    /vitastor/config/global.
 - name: immediate_commit
   type: string
   default: false
@@ -85,9 +114,10 @@
     SSD cache or "media-cache" - for example, a lot of Seagate EXOS drives have
     it (they have internal SSD cache even though it's not stated in datasheets).
 
-    Setting this parameter to "all" or "small" in OSD parameters requires enabling
-    disable_journal_fsync and disable_meta_fsync, setting it to "all" also requires
-    enabling disable_data_fsync.
+    This parameter must be set both in etcd in /vitastor/config/global and in
+    OSD command line or configuration. Setting it to "all" or "small" requires
+    enabling disable_journal_fsync and disable_meta_fsync, setting it to "all"
+    also requires enabling disable_data_fsync.
 
     TLDR: For optimal performance, set immediate_commit to "all" if you only use
     SSDs with supercapacitor-based power loss protection (nonvolatile
@@ -138,7 +168,8 @@
     многих дисках Seagate EXOS (у них есть внутренний SSD-кэш, хотя это и не
     указано в спецификациях).
 
-    Указание "all" или "small" в настройках / командной строке OSD требует
+    Данный параметр нужно указывать и в etcd в /vitastor/config/global, и в
+    командной строке или конфигурации OSD. Значения "all" и "small" требуют
     включения disable_journal_fsync и disable_meta_fsync, значение "all" также
     требует включения disable_data_fsync.
 
@@ -148,3 +179,22 @@
     такие SSD для всех журналов, но не для данных - можете установить параметр
     в "small". Если и какие-то из дисков журналов имеют волатильный кэш записи -
     оставьте параметр пустым.
+- name: client_dirty_limit
+  type: int
+  default: 33554432
+  info: |
+    Without immediate_commit=all this parameter sets the limit of "dirty"
+    (not committed by fsync) data allowed by the client before forcing an
+    additional fsync and committing the data. Also note that the client always
+    holds a copy of uncommitted data in memory so this setting also affects
+    RAM usage of clients.
+
+    This parameter doesn't affect OSDs themselves.
+  info_ru: |
+    При работе без immediate_commit=all - это лимит объёма "грязных" (не
+    зафиксированных fsync-ом) данных, при достижении которого клиент будет
+    принудительно вызывать fsync и фиксировать данные. Также стоит иметь в виду,
+    что в этом случае до момента fsync клиент хранит копию незафиксированных
+    данных в памяти, то есть, настройка влияет на потребление памяти клиентами.
+
+    Параметр не влияет на сами OSD.

docs/config/src/network.yml

@@ -223,22 +223,3 @@
     detect disconnections quickly.
   info_ru: |
     Интервал проверки живости вебсокет-подключений к etcd.
-- name: client_dirty_limit
-  type: int
-  default: 33554432
-  info: |
-    Without immediate_commit=all this parameter sets the limit of "dirty"
-    (not committed by fsync) data allowed by the client before forcing an
-    additional fsync and committing the data. Also note that the client always
-    holds a copy of uncommitted data in memory so this setting also affects
-    RAM usage of clients.
-
-    This parameter doesn't affect OSDs themselves.
-  info_ru: |
-    При работе без immediate_commit=all - это лимит объёма "грязных" (не
-    зафиксированных fsync-ом) данных, при достижении которого клиент будет
-    принудительно вызывать fsync и фиксировать данные. Также стоит иметь в виду,
-    что в этом случае до момента fsync клиент хранит копию незафиксированных
-    данных в памяти, то есть, настройка влияет на потребление памяти клиентами.
-
-    Параметр не влияет на сами OSD.

docs/config/src/osd.yml

@@ -102,20 +102,6 @@
     момент времени. На данный момент единственный параметр, который можно менять
     для ускорения или замедления восстановления и перебалансировки данных, но
     в планах реализация других параметров.
-- name: recovery_pg_switch
-  type: int
-  default: 128
-  info: |
-    Number of recovery operations before switching to recovery of the next PG.
-    The idea is to mix all PGs during recovery for more even space and load
-    distribution but still benefit from recovery queue depth greater than 1.
-    Degraded PGs are anyway scanned first.
-  info_ru: |
-    Число операций восстановления перед переключением на восстановление другой PG.
-    Идея заключается в том, чтобы восстанавливать все PG одновременно для более
-    равномерного распределения места и нагрузки, но при этом всё равно выигрывать
-    от глубины очереди восстановления, большей, чем 1. Деградированные PG в любом
-    случае сканируются первыми.
 - name: recovery_sync_batch
   type: int
   default: 16

docs/installation/proxmox.en.md

@@ -6,10 +6,10 @@
 
 # Proxmox VE
 
-To enable Vitastor support in Proxmox Virtual Environment (6.4-7.3 are supported):
+To enable Vitastor support in Proxmox Virtual Environment (6.4 and 7.1 are supported):
 
-- Add the corresponding Vitastor Debian repository into sources.list on Proxmox hosts:
-  buster for 6.4, bullseye for 7.3, pve7.1 for 7.1, pve7.2 for 7.2
+- Add the corresponding Vitastor Debian repository into sources.list on Proxmox hosts
+  (buster for 6.4, bullseye for 7.1)
 - Install vitastor-client, pve-qemu-kvm, pve-storage-vitastor (* or see note) packages from Vitastor repository
 - Define storage in `/etc/pve/storage.cfg` (see below)
 - Block network access from VMs to Vitastor network (to OSDs and etcd),
@@ -35,5 +35,5 @@ vitastor: vitastor
     vitastor_nbd 0
 ```
 
-\* Note: you can also manually copy [patches/VitastorPlugin.pm](patches/VitastorPlugin.pm) to Proxmox hosts
+\* Note: you can also manually copy [patches/PVE_VitastorPlugin.pm](patches/PVE_VitastorPlugin.pm) to Proxmox hosts
 as `/usr/share/perl5/PVE/Storage/Custom/VitastorPlugin.pm` instead of installing pve-storage-vitastor.

docs/installation/proxmox.ru.md

@@ -6,10 +6,10 @@
 
 # Proxmox
 
-Чтобы подключить Vitastor к Proxmox Virtual Environment (поддерживаются версии 6.4-7.3):
+Чтобы подключить Vitastor к Proxmox Virtual Environment (поддерживаются версии 6.4 и 7.1):
 
-- Добавьте соответствующий Debian-репозиторий Vitastor в sources.list на хостах Proxmox:
-  buster для 6.4, bullseye для 7.3, pve7.1 для 7.1, pve7.2 для 7.2
+- Добавьте соответствующий Debian-репозиторий Vitastor в sources.list на хостах Proxmox
+  (buster для 6.4, bullseye для 7.1)
 - Установите пакеты vitastor-client, pve-qemu-kvm, pve-storage-vitastor (* или см. сноску) из репозитория Vitastor
 - Определите тип хранилища в `/etc/pve/storage.cfg` (см. ниже)
 - Обязательно заблокируйте доступ от виртуальных машин к сети Vitastor (OSD и etcd), т.к. Vitastor (пока) не поддерживает аутентификацию
@@ -35,5 +35,5 @@ vitastor: vitastor
 ```
 
 \* Примечание: вместо установки пакета pve-storage-vitastor вы можете вручную скопировать файл
-[patches/VitastorPlugin.pm](patches/VitastorPlugin.pm) на хосты Proxmox как
+[patches/PVE_VitastorPlugin.pm](patches/PVE_VitastorPlugin.pm) на хосты Proxmox как
 `/usr/share/perl5/PVE/Storage/Custom/VitastorPlugin.pm`.

docs/intro/architecture.ru.md

@@ -127,7 +127,7 @@
   запросы записи клиенты копируют в памяти и при потере соединения и повторном соединении
   с OSD повторяют из памяти. Скопированные в память данные удаляются при успешном fsync,
   а чтобы хранение этих данных не приводило к чрезмерному потреблению памяти, клиенты
-  автоматически выполняют fsync каждые [client_dirty_limit](../config/network.ru.md#client_dirty_limit)
+  автоматически выполняют fsync каждые [client_dirty_limit](../config/layout-cluster.ru.md#client_dirty_limit)
   записанных байт.
 
 ## Схожесть с Ceph

docs/intro/features.en.md

@@ -34,7 +34,6 @@
 
 - [Debian and CentOS packages](../installation/packages.en.md)
 - [Image management CLI (vitastor-cli)](../usage/cli.en.md)
-- [Disk management CLI (vitastor-disk)](docs/usage/disk.en.md)
 - Generic user-space client library
 - [Native QEMU driver](../usage/qemu.en.md)
 - [Loadable fio engine for benchmarks](../usage/fio.en.md)
@@ -48,6 +47,7 @@
 
 The following features are planned for the future:
 
+- Better OSD creation and auto-start tools
 - Other administrative tools
 - Web GUI
 - OpenNebula plugin

docs/intro/features.ru.md

@@ -36,7 +36,6 @@
 
 - [Пакеты для Debian и CentOS](../installation/packages.ru.md)
 - [Консольный интерфейс управления образами (vitastor-cli)](../usage/cli.ru.md)
-- [Инструмент управления дисками (vitastor-disk)](docs/usage/disk.ru.md)
 - Общая пользовательская клиентская библиотека для работы с кластером
 - [Драйвер диска для QEMU](../usage/qemu.ru.md)
 - [Драйвер диска для утилиты тестирования производительности fio](../usage/fio.ru.md)
@@ -48,6 +47,7 @@
 
 ## Планы развития
 
+- Более корректные скрипты разметки дисков и автоматического запуска OSD
 - Другие инструменты администрирования
 - Web-интерфейс
 - Плагин для OpenNebula

docs/intro/quickstart.en.md

@@ -26,14 +26,9 @@
 ## Configure monitors
 
 On the monitor hosts:
-- Put identical etcd_address into `/etc/vitastor/vitastor.conf`. Example:
-  ```
-  {
-    "etcd_address": ["10.200.1.10:2379","10.200.1.11:2379","10.200.1.12:2379"]
-  }
-  ```
-- Create systemd units for etcd by running: `/usr/lib/vitastor/mon/make-etcd`
-- Start etcd and monitors: `systemctl enable --now etcd vitastor-mon`
+- Edit variables at the top of `/usr/lib/vitastor/mon/make-units.sh` to desired values.
+- Create systemd units for the monitor and etcd: `/usr/lib/vitastor/mon/make-units.sh`
+- Start etcd and monitors: `systemctl start etcd vitastor-mon`
 
 ## Configure OSDs
 
@@ -45,9 +40,11 @@ On the monitor hosts:
   }
   ```
 - Initialize OSDs:
-  - SSD-only: `vitastor-disk prepare /dev/sdXXX [/dev/sdYYY ...]`
-  - Hybrid, SSD+HDD: `vitastor-disk prepare --hybrid /dev/sdXXX [/dev/sdYYY ...]`.
-    Pass all your devices (HDD and SSD) to this script — it will partition disks and initialize journals on its own.
+  - Simplest, SSD-only: `/usr/lib/vitastor/mon/make-osd.sh /dev/disk/by-partuuid/XXX [/dev/disk/by-partuuid/YYY ...]`
+    **Warning!** This very simple script by default makes units for server-grade SSDs with write-through cache!
+    If it's not your case, you MUST remove disable_data_fsync and immediate_commit from systemd units.
+  - Hybrid, HDD+SSD: `/usr/lib/vitastor/mon/make-osd-hybrid.js /dev/sda /dev/sdb ...` — pass all your
+    devices (HDD and SSD) to this script — it will partition disks and initialize journals on its own.
     This script skips HDDs which are already partitioned so if you want to use non-empty disks for
     Vitastor you should first wipe them with `wipefs -a`. SSDs with GPT partition table are not skipped,
     but some free unpartitioned space must be available because the script creates new partitions for journals.
@@ -70,7 +67,7 @@ For EC pools the configuration should look like the following:
 
 ```
 etcdctl --endpoints=... put /vitastor/config/pools '{"2":{"name":"ecpool",
-  "scheme":"ec","pg_size":4,"parity_chunks":2,"pg_minsize":2,"pg_count":256,"failure_domain":"host"}'
+  "scheme":"ec","pg_size":4,"parity_chunks":2,"pg_minsize":2,"pg_count":256,"failure_domain":"host"}`
 ```
 
 After you do this, one of the monitors will configure PGs and OSDs will start them.

docs/intro/quickstart.ru.md

@@ -26,14 +26,16 @@
 ## Настройте мониторы
 
 На хостах, выделенных под мониторы:
-- Пропишите одинаковые etcd_address в `/etc/vitastor/vitastor.conf`. Например:
+- Пропишите нужные вам значения в файле `/usr/lib/vitastor/mon/make-units.sh`
+- Создайте юниты systemd для etcd и мониторов: `/usr/lib/vitastor/mon/make-units.sh`
+- Запустите etcd и мониторы: `systemctl start etcd vitastor-mon`
+- Пропишите etcd_address и osd_network в `/etc/vitastor/vitastor.conf`. Например:
   ```
   {
-    "etcd_address": ["10.200.1.10:2379","10.200.1.11:2379","10.200.1.12:2379"]
+    "etcd_address": ["10.200.1.10:2379","10.200.1.11:2379","10.200.1.12:2379"],
+    "osd_network": "10.200.1.0/24"
   }
   ```
-- Инициализируйте сервисы etcd, запустив `/usr/lib/vitastor/mon/make-etcd`
-- Запустите etcd и мониторы: `systemctl enable --now etcd vitastor-mon`
 
 ## Настройте OSD
 
@@ -45,10 +47,12 @@
   }
   ```
 - Инициализуйте OSD:
-  - SSD: `vitastor-disk prepare /dev/sdXXX [/dev/sdYYY ...]`
-  - Гибридные, SSD+HDD: `vitastor-disk prepare --hybrid /dev/sdXXX [/dev/sdYYY ...]`.
-    Передайте все ваши SSD и HDD скрипту в командной строке подряд, скрипт автоматически выделит
-    разделы под журналы на SSD и данные на HDD. Скрипт пропускает HDD, на которых уже есть разделы
+  - SSD: `/usr/lib/vitastor/make-osd.sh /dev/disk/by-partuuid/XXX [/dev/disk/by-partuuid/YYY ...]`. \
+    **Внимание!** Скрипт по умолчанию рассчитан на то, что у вас диски с конденсаторами и отключённым
+    кэшем! Если это не так, из юнитов systemd нужно убрать строчки disable_data_fsync и immediate_commit!
+  - Гибридные, HDD+SSD: `/usr/lib/vitastor/mon/make-osd-hybrid.js /dev/sda /dev/sdb ...` - передайте
+    все ваши SSD и HDD скрипту в командной строке подряд, скрипт автоматически выделит разделы под
+    журналы на SSD и данные на HDD. Скрипт пропускает HDD, на которых уже есть разделы
     или вообще какие-то данные, поэтому если диски непустые, сначала очистите их с помощью
     `wipefs -a`. SSD с таблицей разделов не пропускаются, но так как скрипт создаёт новые разделы
     для журналов, на SSD должно быть доступно свободное нераспределённое место.
@@ -71,7 +75,7 @@ etcdctl --endpoints=... put /vitastor/config/pools '{"1":{"name":"testpool",
 
 ```
 etcdctl --endpoints=... put /vitastor/config/pools '{"2":{"name":"ecpool",
-  "scheme":"ec","pg_size":4,"parity_chunks":2,"pg_minsize":2,"pg_count":256,"failure_domain":"host"}'
+  "scheme":"ec","pg_size":4,"parity_chunks":2,"pg_minsize":2,"pg_count":256,"failure_domain":"host"}`
 ```
 
 После этого один из мониторов должен сконфигурировать PG, а OSD должны запустить их.

docs/usage/cli.en.md

@@ -20,7 +20,7 @@ It supports the following commands:
 - [rm-data](#rm-data)
 - [merge-data](#merge-data)
 - [alloc-osd](#alloc-osd)
-- [rm-osd](#rm-osd)
+- [simple-offsets](#simple-offsets)
 
 Global options:
 
@@ -38,9 +38,9 @@ Global options:
 
 `vitastor-cli status`
 
-Show cluster status.
+Показать состояние кластера.
 
-Example output:
+Пример вывода:
 
 ```
   cluster:
@@ -65,9 +65,9 @@ Example output:
 
 `vitastor-cli df`
 
-Show pool space statistics.
+Показать список пулов и занятое место.
 
-Example output:
+Пример вывода:
 
 ```
 NAME      SCHEME  PGS  TOTAL    USED    AVAILABLE  USED%   EFFICIENCY
@@ -76,26 +76,27 @@ size1     1/1     32   199.9 G  10 G    121.5 G    39.23%  100%
 kaveri    2/1     32   0 B      10 G    0 B        100%    0%
 ```
 
-In the example above, "kaveri" pool has "zero" efficiency because all its OSD are down.
+В примере у пула "kaveri" эффективность равна нулю, так как все OSD выключены.
 
 ## ls
 
 `vitastor-cli ls [-l] [-p POOL] [--sort FIELD] [-r] [-n N] [<glob> ...]`
 
-List images (only matching `<glob>` pattern(s) if passed).
+Показать список образов, если переданы шаблоны `<glob>`, то только с именами,
+соответствующими этим шаблонам (стандартные ФС-шаблоны с * и ?).
 
-Options:
+Опции:
 
 ```
--p|--pool POOL  Filter images by pool ID or name
--l|--long       Also report allocated size and I/O statistics
---del           Also include delete operation statistics
---sort FIELD    Sort by specified field (name, size, used_size, <read|write|delete>_<iops|bps|lat|queue>)
--r|--reverse    Sort in descending order
--n|--count N    Only list first N items
+-p|--pool POOL  Фильтровать образы по пулу (ID или имени)
+-l|--long       Также выводить статистику занятого места и ввода-вывода
+--del           Также выводить статистику операций удаления
+--sort FIELD    Сортировать по заданному полю (name, size, used_size, <read|write|delete>_<iops|bps|lat|queue>)
+-r|--reverse    Сортировать в обратном порядке
+-n|--count N    Показывать только первые N записей
 ```
 
-Example output:
+Пример вывода:
 
 ```
 NAME                 POOL      SIZE  USED    READ   IOPS  QUEUE  LAT   WRITE  IOPS  QUEUE  LAT   FLAGS  PARENT
@@ -112,78 +113,94 @@ bench-kaveri         kaveri    10 G  10 G    0 B/s  0     0      0 us  0 B/s  0
 
 `vitastor-cli create -s|--size <size> [-p|--pool <id|name>] [--parent <parent_name>[@<snapshot>]] <name>`
 
-Create an image. You may use K/M/G/T suffixes for `<size>`. If `--parent` is specified,
-a copy-on-write image clone is created. Parent must be a snapshot (readonly image).
-Pool must be specified if there is more than one pool.
+Создать образ. Для размера `<size>` можно использовать суффиксы K/M/G/T (килобайт-мегабайт-гигабайт-терабайт).
+Если указана опция `--parent`, создаётся клон образа. Родитель `<parent_name>[@<snapshot>]` должен быть
+снимком (или просто немодифицируемым образом). Пул обязательно указывать, если в кластере больше одного пула.
 
 ```
 vitastor-cli create --snapshot <snapshot> [-p|--pool <id|name>] <image>
 vitastor-cli snap-create [-p|--pool <id|name>] <image>@<snapshot>
 ```
 
-Create a snapshot of image `<name>` (either form can be used). May be used live if only a single writer is active.
+Создать снимок образа `<name>` (можно использовать любую форму команды). Снимок можно создавать без остановки
+клиентов, если пишущий клиент максимум 1.
 
 ## modify
 
 `vitastor-cli modify <name> [--rename <new-name>] [--resize <size>] [--readonly | --readwrite] [-f|--force]`
 
-Rename, resize image or change its readonly status. Images with children can't be made read-write.
-If the new size is smaller than the old size, extra data will be purged.
-You should resize file system in the image, if present, before shrinking it.
+Изменить размер, имя образа или флаг "только для чтения". Снимать флаг "только для чтения"
+и уменьшать размер образов, у которых есть дочерние клоны, без `--force` нельзя.
+
+Если новый размер меньше старого, "лишние" данные будут удалены, поэтому перед уменьшением
+образа сначала уменьшите файловую систему в нём.
 
 ```
--f|--force  Proceed with shrinking or setting readwrite flag even if the image has children.
+-f|--force  Разрешить уменьшение или перевод в чтение-запись образа, у которого есть клоны.
 ```
 
 ## rm
 
 `vitastor-cli rm <from> [<to>] [--writers-stopped]`
 
-Remove `<from>` or all layers between `<from>` and `<to>` (`<to>` must be a child of `<from>`),
-rebasing all their children accordingly. --writers-stopped allows merging to be a bit
-more effective in case of a single 'slim' read-write child and 'fat' removed parent:
-the child is merged into parent and parent is renamed to child in that case.
-In other cases parent layers are always merged into children.
+Удалить образ `<from>` или все слои от `<from>` до `<to>` (`<to>` должен быть дочерним
+образом `<from>`), одновременно меняя родительские образы их клонов (если таковые есть).
+
+`--writers-stopped` позволяет чуть более эффективно удалять образы в частом случае, когда
+у удаляемой цепочки есть только один дочерний образ, содержащий небольшой объём данных.
+В этом случае дочерний образ вливается в родительский и удаляется, а родительский
+переименовывается в дочерний.
+
+В других случаях родительские слои вливаются в дочерние.
 
 ## flatten
 
 `vitastor-cli flatten <layer>`
 
-Flatten a layer, i.e. merge data and detach it from parents.
+Сделай образ `<layer>` плоским, то есть, скопировать в него данные и разорвать его
+соединение с родительскими.
 
 ## rm-data
 
 `vitastor-cli rm-data --pool <pool> --inode <inode> [--wait-list] [--min-offset <offset>]`
 
-Remove inode data without changing metadata.
+Удалить данные инода, не меняя метаданные образов.
 
 ```
---wait-list   Retrieve full objects listings before starting to remove objects.
-              Requires more memory, but allows to show correct removal progress.
---min-offset  Purge only data starting with specified offset.
+--wait-list   Сначала запросить полный листинг объектов, а потом начать удалять.
+              Требует больше памяти, но позволяет правильно печатать прогресс удаления.
+--min-offset  Удалять только данные, начиная с заданного смещения.
 ```
 
 ## merge-data
 
 `vitastor-cli merge-data <from> <to> [--target <target>]`
 
-Merge layer data without changing metadata. Merge `<from>`..`<to>` to `<target>`.
-`<to>` must be a child of `<from>` and `<target>` may be one of the layers between
-`<from>` and `<to>`, including `<from>` and `<to>`.
+Слить данные слоёв, не меняя метаданные. Вливает данные из слоёв от `<from>` до `<to>`
+в целевой образ `<target>`. `<to>` должен быть дочерним образом `<from>`, а `<target>`
+должен быть одним из слоёв между `<from>` и `<to>`, включая сами `<from>` и `<to>`.
 
 ## alloc-osd
 
 `vitastor-cli alloc-osd`
 
-Allocate a new OSD number and reserve it by creating empty `/osd/stats/<n>` key.
+Атомарно выделить новый номер OSD и зарезервировать его, создав в etcd пустой
+ключ `/osd/stats/<n>`.
 
-## rm-osd
+## simple-offsets
 
-`vitastor-cli rm-osd [--force] [--allow-data-loss] [--dry-run] <osd_id> [osd_id...]`
+`vitastor-cli simple-offsets <device>`
 
-Remove metadata and configuration for specified OSD(s) from etcd.
+Рассчитать смещения для простого и тупого создания OSD на диске (без суперблока).
 
-Refuses to remove OSDs with data without `--force` and `--allow-data-loss`.
+Опции (см. также [Дисковые параметры уровня кластера](../config/layout-cluster.ru.md)):
 
-With `--dry-run` only checks if deletion is possible without data loss and
-redundancy degradation.
+```
+--object_size 128k       Размер блока хранилища
+--bitmap_granularity 4k  Гранулярность битовых карт
+--journal_size 32M       Размер журнала
+--device_block_size 4k   Размер блока устройства
+--journal_offset 0       Смещение журнала
+--device_size 0          Размер устройства
+--format text            Формат результата: json, options, env или text
+```

docs/usage/cli.ru.md

@@ -21,7 +21,7 @@ vitastor-cli - интерфейс командной строки для адм
 - [rm-data](#rm-data)
 - [merge-data](#merge-data)
 - [alloc-osd](#alloc-osd)
-- [rm-osd](#rm-osd)
+- [simple-offsets](#simple-offsets)
 
 Глобальные опции:
 
@@ -39,9 +39,9 @@ vitastor-cli - интерфейс командной строки для адм
 
 `vitastor-cli status`
 
-Показать состояние кластера.
+Show cluster status.
 
-Пример вывода:
+Example output:
 
 ```
   cluster:
@@ -66,9 +66,9 @@ vitastor-cli - интерфейс командной строки для адм
 
 `vitastor-cli df`
 
-Показать список пулов и занятое место.
+Show pool space statistics.
 
-Пример вывода:
+Example output:
 
 ```
 NAME      SCHEME  PGS  TOTAL    USED    AVAILABLE  USED%   EFFICIENCY
@@ -77,27 +77,26 @@ size1     1/1     32   199.9 G  10 G    121.5 G    39.23%  100%
 kaveri    2/1     32   0 B      10 G    0 B        100%    0%
 ```
 
-В примере у пула "kaveri" эффективность равна нулю, так как все OSD выключены.
+In the example above, "kaveri" pool has "zero" efficiency because all its OSD are down.
 
 ## ls
 
 `vitastor-cli ls [-l] [-p POOL] [--sort FIELD] [-r] [-n N] [<glob> ...]`
 
-Показать список образов, если переданы шаблоны `<glob>`, то только с именами,
-соответствующими этим шаблонам (стандартные ФС-шаблоны с * и ?).
+List images (only matching `<glob>` pattern(s) if passed).
 
-Опции:
+Options:
 
 ```
--p|--pool POOL  Фильтровать образы по пулу (ID или имени)
--l|--long       Также выводить статистику занятого места и ввода-вывода
---del           Также выводить статистику операций удаления
---sort FIELD    Сортировать по заданному полю (name, size, used_size, <read|write|delete>_<iops|bps|lat|queue>)
--r|--reverse    Сортировать в обратном порядке
--n|--count N    Показывать только первые N записей
+-p|--pool POOL  Filter images by pool ID or name
+-l|--long       Also report allocated size and I/O statistics
+--del           Also include delete operation statistics
+--sort FIELD    Sort by specified field (name, size, used_size, <read|write|delete>_<iops|bps|lat|queue>)
+-r|--reverse    Sort in descending order
+-n|--count N    Only list first N items
 ```
 
-Пример вывода:
+Example output:
 
 ```
 NAME                 POOL      SIZE  USED    READ   IOPS  QUEUE  LAT   WRITE  IOPS  QUEUE  LAT   FLAGS  PARENT
@@ -114,87 +113,85 @@ bench-kaveri         kaveri    10 G  10 G    0 B/s  0     0      0 us  0 B/s  0
 
 `vitastor-cli create -s|--size <size> [-p|--pool <id|name>] [--parent <parent_name>[@<snapshot>]] <name>`
 
-Создать образ. Для размера `<size>` можно использовать суффиксы K/M/G/T (килобайт-мегабайт-гигабайт-терабайт).
-Если указана опция `--parent`, создаётся клон образа. Родитель `<parent_name>[@<snapshot>]` должен быть
-снимком (или просто немодифицируемым образом). Пул обязательно указывать, если в кластере больше одного пула.
+Create an image. You may use K/M/G/T suffixes for `<size>`. If `--parent` is specified,
+a copy-on-write image clone is created. Parent must be a snapshot (readonly image).
+Pool must be specified if there is more than one pool.
 
 ```
 vitastor-cli create --snapshot <snapshot> [-p|--pool <id|name>] <image>
 vitastor-cli snap-create [-p|--pool <id|name>] <image>@<snapshot>
 ```
 
-Создать снимок образа `<name>` (можно использовать любую форму команды). Снимок можно создавать без остановки
-клиентов, если пишущий клиент максимум 1.
+Create a snapshot of image `<name>` (either form can be used). May be used live if only a single writer is active.
 
 ## modify
 
 `vitastor-cli modify <name> [--rename <new-name>] [--resize <size>] [--readonly | --readwrite] [-f|--force]`
 
-Изменить размер, имя образа или флаг "только для чтения". Снимать флаг "только для чтения"
-и уменьшать размер образов, у которых есть дочерние клоны, без `--force` нельзя.
-
-Если новый размер меньше старого, "лишние" данные будут удалены, поэтому перед уменьшением
-образа сначала уменьшите файловую систему в нём.
+Rename, resize image or change its readonly status. Images with children can't be made read-write.
+If the new size is smaller than the old size, extra data will be purged.
+You should resize file system in the image, if present, before shrinking it.
 
 ```
--f|--force  Разрешить уменьшение или перевод в чтение-запись образа, у которого есть клоны.
+-f|--force  Proceed with shrinking or setting readwrite flag even if the image has children.
 ```
 
 ## rm
 
 `vitastor-cli rm <from> [<to>] [--writers-stopped]`
 
-Удалить образ `<from>` или все слои от `<from>` до `<to>` (`<to>` должен быть дочерним
-образом `<from>`), одновременно меняя родительские образы их клонов (если таковые есть).
-
-`--writers-stopped` позволяет чуть более эффективно удалять образы в частом случае, когда
-у удаляемой цепочки есть только один дочерний образ, содержащий небольшой объём данных.
-В этом случае дочерний образ вливается в родительский и удаляется, а родительский
-переименовывается в дочерний.
-
-В других случаях родительские слои вливаются в дочерние.
+Remove `<from>` or all layers between `<from>` and `<to>` (`<to>` must be a child of `<from>`),
+rebasing all their children accordingly. --writers-stopped allows merging to be a bit
+more effective in case of a single 'slim' read-write child and 'fat' removed parent:
+the child is merged into parent and parent is renamed to child in that case.
+In other cases parent layers are always merged into children.
 
 ## flatten
 
 `vitastor-cli flatten <layer>`
 
-Сделай образ `<layer>` плоским, то есть, скопировать в него данные и разорвать его
-соединение с родительскими.
+Flatten a layer, i.e. merge data and detach it from parents.
 
 ## rm-data
 
 `vitastor-cli rm-data --pool <pool> --inode <inode> [--wait-list] [--min-offset <offset>]`
 
-Удалить данные инода, не меняя метаданные образов.
+Remove inode data without changing metadata.
 
 ```
---wait-list   Сначала запросить полный листинг объектов, а потом начать удалять.
-              Требует больше памяти, но позволяет правильно печатать прогресс удаления.
---min-offset  Удалять только данные, начиная с заданного смещения.
+--wait-list   Retrieve full objects listings before starting to remove objects.
+              Requires more memory, but allows to show correct removal progress.
+--min-offset  Purge only data starting with specified offset.
 ```
 
 ## merge-data
 
 `vitastor-cli merge-data <from> <to> [--target <target>]`
 
-Слить данные слоёв, не меняя метаданные. Вливает данные из слоёв от `<from>` до `<to>`
-в целевой образ `<target>`. `<to>` должен быть дочерним образом `<from>`, а `<target>`
-должен быть одним из слоёв между `<from>` и `<to>`, включая сами `<from>` и `<to>`.
+Merge layer data without changing metadata. Merge `<from>`..`<to>` to `<target>`.
+`<to>` must be a child of `<from>` and `<target>` may be one of the layers between
+`<from>` and `<to>`, including `<from>` and `<to>`.
 
 ## alloc-osd
 
 `vitastor-cli alloc-osd`
 
-Атомарно выделить новый номер OSD и зарезервировать его, создав в etcd пустой
-ключ `/osd/stats/<n>`.
+Allocate a new OSD number and reserve it by creating empty `/osd/stats/<n>` key.
 
-## rm-osd
+## simple-offsets
 
-`vitastor-cli rm-osd [--force] [--allow-data-loss] [--dry-run] <osd_id> [osd_id...]`
+`vitastor-cli simple-offsets <device>`
 
-Удалить метаданные и конфигурацию для заданных OSD из etcd.
+Calculate offsets for simple&stupid (no superblock) OSD deployment.
 
-Отказывается удалять OSD с данными без опций `--force` и `--allow-data-loss`.
+Options (see also [Cluster-Wide Disk Layout Parameters](../config/layout-cluster.en.md)):
 
-С опцией `--dry-run` только проверяет, возможно ли удаление без потери данных и деградации
-избыточности.
+```
+--object_size 128k       Set blockstore block size
+--bitmap_granularity 4k  Set bitmap granularity
+--journal_size 32M       Set journal size
+--device_block_size 4k   Set device block size
+--journal_offset 0       Set journal offset
+--device_size 0          Set device size
+--format text            Result format: json, options, env, or text
+```

docs/usage/disk.en.md

@@ -1,257 +0,0 @@
-[Documentation](../../README.md#documentation) → Usage → Disk Tool
-
------
-
-[Читать на русском](disk.ru.md)
-
-# Disk management tool
-
-vitastor-disk is a command-line tool for physical Vitastor disk management.
-
-It supports the following commands:
-
-- [prepare](#prepare)
-- [upgrade-simple](#upgrade-simple)
-- [resize](#resize)
-- [start/stop/restart/enable/disable](#start/stop/restart/enable/disable)
-- [purge](#purge)
-- [read-sb](#read-sb)
-- [write-sb](#write-sb)
-- [udev](#udev)
-- [exec-osd](#exec-osd)
-- [pre-exec](#pre-exec)
-- Debugging:
-  - [dump-journal](#dump-journal)
-  - [write-journal](#write-journal)
-  - [dump-meta](#dump-meta)
-  - [write-meta](#write-meta)
-- [simple-offsets](#simple-offsets)
-
-## prepare
-
-`vitastor-disk prepare [OPTIONS] [devices...]`
-
-Initialize disk(s) for Vitastor OSD(s).
-
-There are two modes of this command. In the first mode, you pass `<devices>` which
-must be raw disks (not partitions). They are partitioned automatically and OSDs
-are initialized on all of them.
-
-In the second mode, you omit `<devices>` and pass `--data_device`, `--journal_device`
-and/or `--meta_device` which must be already existing partitions identified by their
-GPT partition UUIDs. In this case a single OSD is created.
-
-Requires `vitastor-cli`, `wipefs`, `sfdisk` and `partprobe` (from parted) utilities.
-
-Options (automatic mode):
-
-```
---osd_per_disk <N>
-  Create <N> OSDs on each disk (default 1)
---hybrid
-  Prepare hybrid (HDD+SSD) OSDs using provided devices. SSDs will be used for
-  journals and metadata, HDDs will be used for data. Partitions for journals and
-  metadata will be created automatically. Whether disks are SSD or HDD is decided
-  by the `/sys/block/.../queue/rotational` flag. In hybrid mode, default object
-  size is 1 MB instead of 128 KB, default journal size is 1 GB instead of 32 MB,
-  and throttle_small_writes is enabled by default.
---disable_data_fsync auto
-  Disable data device cache and fsync (1/yes/true = on, default auto)
---disable_meta_fsync auto
-  Disable metadata/journal device cache and fsync (default auto)
---meta_reserve 2x,1G
-  New metadata partitions in --hybrid mode are created larger than actual
-  metadata size to ease possible future extension. The default is to allocate
-  2 times more space and at least 1G. Use this option to override.
---max_other 10%
-  Use disks for OSD data even if they already have non-Vitastor partitions,
-  but only if these take up no more than this percent of disk space.
-```
-
-Options (single-device mode):
-
-```
---data_device <DEV>        Use partition <DEV> for data
---meta_device <DEV>        Use partition <DEV> for metadata (optional)
---journal_device <DEV>     Use partition <DEV> for journal (optional)
---disable_data_fsync 0     Disable data device cache and fsync (default off)
---disable_meta_fsync 0     Disable metadata device cache and fsync (default off)
---disable_journal_fsync 0  Disable journal device cache and fsync (default off)
---force                    Bypass partition safety checks (for emptiness and so on)
-```
-
-Options (both modes):
-
-```
---journal_size 1G/32M      Set journal size (area or partition size)
---block_size 1M/128k       Set blockstore object size
---bitmap_granularity 4k    Set bitmap granularity
---data_device_block 4k     Override data device block size
---meta_device_block 4k     Override metadata device block size
---journal_device_block 4k  Override journal device block size
-```
-
-[immediate_commit](../config/layout-cluster.en.md#immediate_commit) setting is
-automatically derived from "disable fsync" options. It's set to "all" when fsync
-is disabled on all devices, and to "small" if fsync is only disabled on journal device.
-
-When data/meta/journal fsyncs are disabled, the OSD startup script automatically
-checks the device cache status on start and tries to disable cache for SATA/SAS disks.
-If it doesn't succeed it issues a warning in the system log.
-
-You can also pass other OSD options here as arguments and they'll be persisted
-to the superblock: max_write_iodepth, max_write_iodepth, min_flusher_count,
-max_flusher_count, inmemory_metadata, inmemory_journal, journal_sector_buffer_count,
-journal_no_same_sector_overwrites, throttle_small_writes, throttle_target_iops,
-throttle_target_mbs, throttle_target_parallelism, throttle_threshold_us.
-See [Runtime OSD Parameters](../config/osd.en.md) for details.
-
-## upgrade-simple
-
-`vitastor-disk upgrade-simple <UNIT_FILE|OSD_NUMBER>`
-
-Upgrade an OSD created by old (0.7.1 and older) `make-osd.sh` or `make-osd-hybrid.js` scripts.
-
-Adds superblocks to OSD devices, disables old `vitastor-osdN` unit and replaces it with `vitastor-osd@N`.
-Can be invoked with an osd number of with a path to systemd service file `UNIT_FILE` which
-must be `/etc/systemd/system/vitastor-osd<OSD_NUMBER>.service`.
-
-Note that the procedure isn't atomic and may ruin OSD data in case of an interrupt,
-so don't upgrade all your OSDs in parallel.
-
-Requires the `sfdisk` utility.
-
-## resize
-
-`vitastor-disk resize <ALL_OSD_PARAMETERS> <NEW_LAYOUT> [--iodepth 32]`
-
-Resize data area and/or rewrite/move journal and metadata.
-
-`ALL_OSD_PARAMETERS` must include all (at least all disk-related)
-parameters from OSD command line (i.e. from systemd unit or superblock).
-
-`NEW_LAYOUT` may include new disk layout parameters:
-
-```
---new_data_offset SIZE     resize data area so it starts at SIZE
---new_data_len SIZE        resize data area to SIZE bytes
---new_meta_device PATH     use PATH for new metadata
---new_meta_offset SIZE     make new metadata area start at SIZE
---new_meta_len SIZE        make new metadata area SIZE bytes long
---new_journal_device PATH  use PATH for new journal
---new_journal_offset SIZE  make new journal area start at SIZE
---new_journal_len SIZE     make new journal area SIZE bytes long
-```
-
-SIZE may include k/m/g/t suffixes. If any of the new layout parameter
-options are not specified, old values will be used.
-
-## start/stop/restart/enable/disable
-
-`vitastor-disk start|stop|restart|enable|disable [--now] <device> [device2 device3 ...]`
-
-Manipulate Vitastor OSDs using systemd by their device paths.
-
-Commands are passed to `systemctl` with `vitastor-osd@<num>` units as arguments.
-
-When `--now` is added to enable/disable, OSDs are also immediately started/stopped.
-
-## purge
-
-`vitastor-disk purge [--force] [--allow-data-loss] <device> [device2 device3 ...]`
-
-Purge Vitastor OSD(s) on specified device(s). Uses `vitastor-cli rm-osd` to check
-if deletion is possible without data loss and to actually remove metadata from etcd.
-`--force` and `--allow-data-loss` options may be used to ignore safety check results.
-
-Requires `vitastor-cli`, `sfdisk` and `partprobe` (from parted) utilities.
-
-## read-sb
-
-`vitastor-disk read-sb [--force] <device>`
-
-Try to read Vitastor OSD superblock from `<device>` and print it in JSON format.
-`--force` allows to ignore validation errors.
-
-## write-sb
-
-`vitastor-disk write-sb <device>`
-
-Read JSON from STDIN and write it into Vitastor OSD superblock on `<device>`.
-
-## udev
-
-`vitastor-disk udev <device>`
-
-Try to read Vitastor OSD superblock from `<device>` and print variables for udev.
-
-## exec-osd
-
-`vitastor-disk exec-osd <device>`
-
-Read Vitastor OSD superblock from `<device>` and start the OSD with parameters from it.
-
-Intended for use from startup scripts (i.e. from systemd units).
-
-## pre-exec
-
-`vitastor-disk pre-exec <device>`
-
-Read Vitastor OSD superblock from `<device>` and perform pre-start checks for the OSD.
-
-For now, this only checks that device cache is in write-through mode if fsync is disabled.
-
-Intended for use from startup scripts (i.e. from systemd units).
-
-## dump-journal
-
-`vitastor-disk dump-journal [OPTIONS] <journal_file> <journal_block_size> <offset> <size>`
-
-Dump journal in human-readable or JSON (if `--json` is specified) format.
-
-Options:
-
-```
---all             Scan the whole journal area for entries and dump them, even outdated ones
---json            Dump journal in JSON format
---format entries  (Default) Dump actual journal entries as an array, without data
---format data     Same as "entries", but also include small write data
---format blocks   Dump as an array of journal blocks each containing array of entries
-```
-
-## write-journal
-
-`vitastor-disk write-journal <journal_file> <journal_block_size> <bitmap_size> <offset> <size>`
-
-Write journal from JSON taken from standard input in the same format as produced by
-`dump-journal --json --format data`.
-
-## dump-meta
-
-`vitastor-disk dump-meta <meta_file> <meta_block_size> <offset> <size>`
-
-Dump metadata in JSON format.
-
-## write-meta
-
-`vitastor-disk write-meta <meta_file> <offset> <size>`
-
-Write metadata from JSON taken from standard input in the same format as produced by `dump-meta`.
-
-## simple-offsets
-
-`vitastor-disk simple-offsets <device>`
-
-Calculate offsets for old simple&stupid (no superblock) OSD deployment.
-
-Options (see also [Cluster-Wide Disk Layout Parameters](../config/layout-cluster.en.md)):
-
-```
---object_size 128k       Set blockstore block size
---bitmap_granularity 4k  Set bitmap granularity
---journal_size 32M       Set journal size
---device_block_size 4k   Set device block size
---journal_offset 0       Set journal offset
---device_size 0          Set device size
---format text            Result format: json, options, env, or text
-```

docs/usage/disk.ru.md

@@ -1,262 +0,0 @@
-[Документация](../../README-ru.md#документация) → Использование → Управление дисками
-
------
-
-[Read in English](disk.en.md)
-
-# Инструмент управления дисками
-
-vitastor-disk - инструмент командной строки для управления дисками Vitastor OSD.
-
-Поддерживаются следующие команды:
-
-- [prepare](#prepare)
-- [upgrade-simple](#upgrade-simple)
-- [resize](#resize)
-- [start/stop/restart/enable/disable](#start/stop/restart/enable/disable)
-- [purge](#purge)
-- [read-sb](#read-sb)
-- [write-sb](#write-sb)
-- [udev](#udev)
-- [exec-osd](#exec-osd)
-- [pre-exec](#pre-exec)
-- Для отладки:
-  - [dump-journal](#dump-journal)
-  - [write-journal](#write-journal)
-  - [dump-meta](#dump-meta)
-  - [write-meta](#write-meta)
-- [simple-offsets](#simple-offsets)
-
-## prepare
-
-`vitastor-disk prepare [OPTIONS] [devices...]`
-
-Подготовить диск(и) для OSD Vitastor.
-
-У команды есть 2 режима. В первом режиме вы указываете список устройств `<devices>`,
-которые должны быть целыми дисками (не разделами). На них автоматически создаются
-разделы и инициализируются OSD.
-
-Во втором режиме вместо списка устройств вы указываете пути к отдельным устройствам
-`--data_device`, `--journal_device` и/или `--meta_device`, которые должны быть
-уже существующими GPT-разделами. В этом случае инициализируется ровно один OSD.
-
-Команде требуются утилиты `vitastor-cli`, `wipefs`, `sfdisk` и `partprobe` (из состава parted).
-
-Опции для автоматического режима:
-
-```
---osd_per_disk <N>
-  Создавать по несколько (<N>) OSD на каждом диске (по умолчанию 1)
---hybrid
-  Инициализировать гибридные (HDD+SSD) OSD на указанных дисках. SSD будут
-  использованы для журналов и метаданных, а HDD - для данных. Разделы для журналов
-  и метаданных будут созданы автоматически. Является ли диск SSD или HDD, определяется
-  по флагу `/sys/block/.../queue/rotational`. В гибридном режиме по умолчанию
-  используется размер объекта 1 МБ вместо 128 КБ, размер журнала 1 ГБ вместо 32 МБ
-  и включённый throttle_small_writes.
---disable_data_fsync auto
-  Отключать кэш и fsync-и для устройств данных. (1/yes/true = да, по умолчанию автоопределение)
---disable_meta_fsync auto
-  Отключать кэш и fsync-и для журналов и метаданных (по умолчанию автоопределение)
---meta_reserve 2x,1G
-  В гибридном режиме для метаданных выделяется больше места, чем нужно на самом
-  деле, чтобы оставить запас под будущее расширение. По умолчанию выделяется
-  в 2 раза больше места, и не менее 1 ГБ. Чтобы изменить это поведение,
-  воспользуйтесь данной опцией.
---max_other 10%
-  Использовать диски под данные OSD, даже если на них уже есть не-Vitastor-овые
-  разделы, но только в случае, если они занимают не более данного процента диска.
-```
-
-Опции для режима одного OSD:
-
-```
---data_device <DEV>        Использовать раздел <DEV> для данных
---meta_device <DEV>        Использовать раздел <DEV> для метаданных (опционально)
---journal_device <DEV>     Использовать раздел <DEV> для журнала (опционально)
---disable_data_fsync 0     Отключить кэш и fsync устройства данных (по умолчанию нет)
---disable_meta_fsync 0     Отключить кэш и fsync метаданных (по умолчанию нет)
---disable_journal_fsync 0  Отключить кэш и fsync журнала (по умолчанию нет)
---force                    Пропустить проверки разделов (на пустоту и т.п.)
-```
-
-Опции для обоих режимов:
-
-```
---journal_size 1G/32M      Задать размер журнала (области или раздела журнала)
---block_size 1M/128k       Задать размер объекта хранилища
---bitmap_granularity 4k    Задать гранулярность битовых карт
---data_device_block 4k     Задать размер блока устройства данных
---meta_device_block 4k     Задать размер блока метаданных
---journal_device_block 4k  Задать размер блока журнала
-```
-
-Настройка [immediate_commit](../config/layout-cluster.ru.md#immediate_commit)
-автоматически выводится из опций отключения кэша - она устанавливается в "all", если кэш
-отключён на всех устройствах, и в "small", если он отключён только на устройстве журнала.
-
-Когда fsync данных/метаданных/журнала отключён, скрипты запуска OSD автоматически
-проверяют состояние кэша диска и стараются его отключить для SATA/SAS дисков. Если
-это не удаётся, в системный журнал выводится предупреждение.
-
-Вы можете передать данной команде и некоторые другие опции OSD в качестве аргументов
-и они тоже будут сохранены в суперблок: max_write_iodepth, max_write_iodepth, min_flusher_count,
-max_flusher_count, inmemory_metadata, inmemory_journal, journal_sector_buffer_count,
-journal_no_same_sector_overwrites, throttle_small_writes, throttle_target_iops,
-throttle_target_mbs, throttle_target_parallelism, throttle_threshold_us.
-Читайте об этих параметрах подробнее в разделе [Изменяемые параметры OSD](../config/osd.ru.md).
-
-## upgrade-simple
-
-`vitastor-disk upgrade-simple <UNIT_FILE|OSD_NUMBER>`
-
-Обновить OSD, созданный старыми (0.7.1 и старее) скриптами `make-osd.sh` и `make-osd-hybrid.js`.
-
-Добавляет суперблок на разделы OSD, отключает старый сервис `vitastor-osdN` и заменяет его на `vitastor-osd@N`.
-
-Можно вызывать, указывая либо номер OSD, либо путь к файлу сервиса `UNIT_FILE`, но он обязан
-иметь вид `/etc/systemd/system/vitastor-osd<OSD_NUMBER>.service`.
-
-Имейте в виду, что процедура обновления не атомарна и при прерывании может уничтожить данные OSD,
-так что обновляйте ваши OSD по очереди.
-
-Команде требуется утилита `sfdisk`.
-
-## resize
-
-`vitastor-disk resize <ALL_OSD_PARAMETERS> <NEW_LAYOUT> [--iodepth 32]`
-
-Изменить размер области данных и/или переместить журнал и метаданные.
-
-В `ALL_OSD_PARAMETERS` нужно указать все относящиеся к диску параметры OSD
-из суперблока OSD или из файла сервиса systemd (в старых версиях).
-
-В `NEW_LAYOUT` нужно указать новые параметры расположения данных:
-
-```
---new_data_offset РАЗМЕР     сдвинуть начало области данных на РАЗМЕР байт
---new_data_len РАЗМЕР        изменить размер области данных до РАЗМЕР байт
---new_meta_device ПУТЬ       использовать ПУТЬ как новое устройство метаданных
---new_meta_offset РАЗМЕР     разместить новые метаданные по смещению РАЗМЕР байт
---new_meta_len РАЗМЕР        сделать новые метаданные размером РАЗМЕР байт
---new_journal_device ПУТЬ    использовать ПУТЬ как новое устройство журнала
---new_journal_offset РАЗМЕР  разместить новый журнал по смещению РАЗМЕР байт
---new_journal_len РАЗМЕР     сделать новый журнал размером РАЗМЕР байт
-```
-
-РАЗМЕР может быть указан с суффиксами k/m/g/t. Если любой из новых параметров
-расположения не указан, он принимается равным старому значению.
-
-## start/stop/restart/enable/disable
-
-`vitastor-disk start|stop|restart|enable|disable [--now] <device> [device2 device3 ...]`
-
-Команды управления OSD по путям дисков через systemd.
-
-Команды транслируются `systemctl` с сервисами `vitastor-osd@<num>` в виде аргументов.
-
-Когда к командам включения/выключения добавляется параметр `--now`, OSD также сразу
-запускаются/останавливаются.
-
-## purge
-
-`vitastor-disk purge [--force] [--allow-data-loss] <device> [device2 device3 ...]`
-
-Удалить OSD на заданном диске/дисках. Использует `vitastor-cli rm-osd` для проверки
-возможности удаления без потери данных и для удаления OSD из etcd. Опции `--force`
-и `--allow-data-loss` служат для обхода данной защиты в случае необходимости.
-
-Команде требуются утилиты `vitastor-cli`, `sfdisk` и `partprobe` (из состава parted).
-
-## read-sb
-
-`vitastor-disk read-sb [--force] <device>`
-
-Прочитать суперблок OSD с диска `<device>` и вывести его в формате JSON.
-
-Опция `--force` позволяет читать суперблок, даже если он считается некорректным
-из-за ошибок валидации.
-
-## write-sb
-
-`vitastor-disk write-sb <device>`
-
-Прочитать JSON со стандартного ввода и записать его в суперблок OSD на диск `<device>`.
-
-## udev
-
-`vitastor-disk udev <device>`
-
-Прочитать суперблок OSD с диска `<device>` и вывести переменные для udev.
-
-## exec-osd
-
-`vitastor-disk exec-osd <device>`
-
-Прочитать суперблок OSD с диска `<device>` и запустить исполняемый файл OSD с параметрами оттуда.
-
-Команда предназначена для использования из скриптов запуска (например, из сервисов systemd).
-
-## pre-exec
-
-`vitastor-disk pre-exec <device>`
-
-Прочитать суперблок OSD с диска `<device>` и провести проверки OSD перед запуском.
-
-На данный момент только отключает кэш диска или проверяет, что он отключён, если в параметрах
-OSD отключены fsync-и.
-
-Команда предназначена для использования из скриптов запуска (например, из сервисов systemd).
-
-## dump-journal
-
-`vitastor-disk dump-journal [OPTIONS] <journal_file> <journal_block_size> <offset> <size>`
-
-Вывести журнал в человекочитаемом или в JSON (с опцией `--json`) виде.
-
-Опции:
-
-```
---all             Просканировать всю область журнала и вывести даже старые записи
---json            Вывести журнал в формате JSON
---format entries  (По умолчанию) Вывести только актуальные записи журнала без данных
---format data     Вывести только актуальные записи журнала с данными
---format blocks   Вывести массив блоков журнала, а в каждом массив актуальных записей без данных
-```
-
-## write-journal
-
-`vitastor-disk write-journal <journal_file> <journal_block_size> <bitmap_size> <offset> <size>`
-
-Записать журнал из JSON со стандартного ввода в формате, аналогичном `dump-journal --json --format data`.
-
-## dump-meta
-
-`vitastor-disk dump-meta <meta_file> <meta_block_size> <offset> <size>`
-
-Вывести метаданные в формате JSON.
-
-## write-meta
-
-`vitastor-disk write-meta <meta_file> <offset> <size>`
-
-Записать метаданные из JSON со стандартного ввода в формате, аналогичном `dump-meta`.
-
-## simple-offsets
-
-`vitastor-disk simple-offsets <device>`
-
-Рассчитать смещения для старого ("простого и тупого") создания OSD на диске (без суперблока).
-
-Опции (см. также [Дисковые параметры уровня кластера](../config/layout-cluster.ru.md)):
-
-```
---object_size 128k       Размер блока хранилища
---bitmap_granularity 4k  Гранулярность битовых карт
---journal_size 32M       Размер журнала
---device_block_size 4k   Размер блока устройства
---journal_offset 0       Смещение журнала
---device_size 0          Размер устройства
---format text            Формат результата: json, options, env или text
-```

mon/90-vitastor.rules

@@ -1,7 +0,0 @@
-SUBSYSTEM=="block", ENV{ID_PART_ENTRY_TYPE}=="e7009fac-a5a1-4d72-af72-53de13059903", \
-    OWNER="vitastor", GROUP="vitastor", \
-    IMPORT{program}="/usr/bin/vitastor-disk udev $devnode", \
-    SYMLINK+="vitastor/$env{VITASTOR_ALIAS}"
-
-ENV{VITASTOR_OSD_NUM}!="", ACTION=="add", RUN{program}+="/usr/bin/systemctl enable --now vitastor-osd@$env{VITASTOR_OSD_NUM}"
-ENV{VITASTOR_OSD_NUM}!="", ACTION=="remove", RUN{program}+="/usr/bin/systemctl disable --now vitastor-osd@$env{VITASTOR_OSD_NUM}"

mon/PGUtil.js

@@ -21,7 +21,7 @@ function add_pg_history(new_pg_history, new_pg, prev_pgs, prev_pg_history, old_p
     {
         for (const pg of oh.osd_sets)
         {
-            nh.osd_sets[pg.join(' ')] = pg.map(osd_num => Number(osd_num));
+            nh.osd_sets[pg.join(' ')] = pg;
         }
     }
     if (oh && oh.all_peers && oh.all_peers.length)

mon/make-etcd

@@ -1,110 +0,0 @@
-#!/usr/bin/node
-// Simple systemd unit generator for etcd
-// Copyright (c) Vitaliy Filippov, 2019+
-// License: MIT
-
-// USAGE:
-// 1) Put the same etcd_address into /etc/vitastor/vitastor.conf on all monitor nodes
-// 2) Run ./make-etcd.js. It will create the etcd service on one of specified IPs
-
-const child_process = require('child_process');
-const fs = require('fs');
-const os = require('os');
-
-run().catch(e => { console.error(e); process.exit(1); });
-
-async function run()
-{
-    const config_path = process.argv[2] || '/etc/vitastor/vitastor.conf';
-    if (config_path == '-h' || config_path == '--help')
-    {
-        console.log(
-            'Initialize systemd etcd service for Vitastor\n'+
-            '(c) Vitaliy Filippov, 2019+ (MIT)\n'+
-            '\n'+
-            'USAGE:\n'+
-            '1) Put the same etcd_address into /etc/vitastor/vitastor.conf on all monitor nodes\n'+
-            '2) Run '+process.argv[1]+' [config_path]\n'
-        );
-        process.exit(0);
-    }
-    if (!fs.existsSync(config_path))
-    {
-        console.log(config_path+' is missing');
-        process.exit(1);
-    }
-    if (fs.existsSync("/etc/systemd/system/etcd.service"))
-    {
-        console.log("/etc/systemd/system/etcd.service already exists");
-        process.exit(1);
-    }
-    const config = JSON.parse(fs.readFileSync(config_path, { encoding: 'utf-8' }));
-    if (!config.etcd_address)
-    {
-        console.log("etcd_address is missing in "+config_path);
-        process.exit(1);
-    }
-    const etcds = (config.etcd_address instanceof Array ? config.etcd_address : (''+config.etcd_address).split(/,/))
-        .map(s => (''+s).replace(/^https?:\/\/\[?|\]?(:\d+)?(\/.*)?$/g, '').toLowerCase());
-    const num = select_local_etcd(etcds);
-    if (num < 0)
-    {
-        console.log('No matching IPs in etcd_address from '+config_path);
-        process.exit(0);
-    }
-    const etcd_cluster = etcds.map((e, i) => `etcd${i}=http://${e}:2380`).join(',');
-    await system(`mkdir -p /var/lib/etcd${num}.etcd`);
-    fs.writeFileSync(
-        "/etc/systemd/system/etcd.service",
-`[Unit]
-Description=etcd for vitastor
-After=network-online.target local-fs.target time-sync.target
-Wants=network-online.target local-fs.target time-sync.target
-
-[Service]
-Restart=always
-ExecStart=/usr/local/bin/etcd -name etcd${num} --data-dir /var/lib/etcd${num}.etcd \\
-    --advertise-client-urls http://${etcds[num]}:2379 --listen-client-urls http://${etcds[num]}:2379 \\
-    --initial-advertise-peer-urls http://${etcds[num]}:2380 --listen-peer-urls http://${etcds[num]}:2380 \\
-    --initial-cluster-token vitastor-etcd-1 --initial-cluster ${etcd_cluster} \\
-    --initial-cluster-state new --max-txn-ops=100000 --max-request-bytes=104857600 \\
-    --auto-compaction-retention=10 --auto-compaction-mode=revision
-WorkingDirectory=/var/lib/etcd${num}.etcd
-ExecStartPre=+chown -R etcd /var/lib/etcd${num}.etcd
-User=etcd
-PrivateTmp=false
-TasksMax=infinity
-Restart=always
-StartLimitInterval=0
-RestartSec=10
-
-[Install]
-WantedBy=local.target
-`);
-    await system(`useradd etcd`);
-    await system(`systemctl daemon-reload`);
-    await system(`systemctl enable etcd`);
-    await system(`systemctl start etcd`);
-    process.exit(0);
-}
-
-function select_local_etcd(etcds)
-{
-    const ifaces = os.networkInterfaces();
-    for (const ifname in ifaces)
-        for (const iface of ifaces[ifname])
-            for (let i = 0; i < etcds.length; i++)
-                if (etcds[i] == iface.address.toLowerCase())
-                    return i;
-    return -1;
-}
-
-async function system(cmd)
-{
-    const cp = child_process.spawn(cmd, { shell: true, stdio: [ 0, 1, 2 ] });
-    let finish_cb;
-    cp.on('exit', () => finish_cb && finish_cb());
-    if (cp.exitCode == null)
-        await new Promise(ok => finish_cb = ok);
-    return cp.exitCode;
-}

mon/make-osd-hybrid.js

@@ -0,0 +1,414 @@
+#!/usr/bin/nodejs
+// systemd unit generator for hybrid (HDD+SSD) vitastor OSDs
+// Copyright (c) Vitaliy Filippov, 2019+
+// License: VNPL-1.1
+
+// USAGE: nodejs make-osd-hybrid.js [--disable_ssd_cache 0] [--disable_hdd_cache 0] /dev/sda /dev/sdb /dev/sdc /dev/sdd ...
+// I.e. - just pass all HDDs and SSDs mixed, the script will decide where
+// to put journals on its own
+
+const fs = require('fs');
+const fsp = fs.promises;
+const child_process = require('child_process');
+
+const options = {
+    debug: 1,
+    journal_size: 1024*1024*1024,
+    min_meta_size: 1024*1024*1024,
+    object_size: 1024*1024,
+    bitmap_granularity: 4096,
+    device_block_size: 4096,
+    disable_ssd_cache: 1,
+    disable_hdd_cache: 1,
+};
+
+run().catch(console.fatal);
+
+async function run()
+{
+    const device_list = parse_options();
+    await system_or_die("mkdir -p /var/log/vitastor; chown vitastor /var/log/vitastor");
+    // Collect devices
+    const all_devices = await collect_devices(device_list);
+    const ssds = all_devices.filter(d => d.ssd);
+    const hdds = all_devices.filter(d => !d.ssd);
+    // Collect existing OSD units
+    const osd_units = await collect_osd_units();
+    // Count assigned HDD journals and unallocated space for each SSD
+    await check_journal_count(ssds, osd_units);
+    // Create new OSDs
+    await create_new_hybrid_osds(hdds, ssds, osd_units);
+    process.exit(0);
+}
+
+function parse_options()
+{
+    const devices = [];
+    const opt = {};
+    for (let i = 2; i < process.argv.length; i++)
+    {
+        const arg = process.argv[i];
+        if (arg == '--help' || arg == '-h')
+        {
+            opt.help = true;
+            break;
+        }
+        else if (arg.substr(0, 2) == '--')
+            opt[arg.substr(2)] = process.argv[++i];
+        else
+            devices.push(arg);
+    }
+    if (opt.help || !devices.length)
+    {
+        console.log(
+            'Prepare hybrid (HDD+SSD) Vitastor OSDs\n'+
+            '(c) Vitaliy Filippov, 2019+, license: VNPL-1.1\n\n'+
+            'USAGE: nodejs make-osd-hybrid.js [OPTIONS] /dev/sda /dev/sdb /dev/sdc ...\n'+
+            'Just pass all your SSDs and HDDs in any order, the script will distribute OSDs for you.\n\n'+
+            'OPTIONS (with defaults):\n'+
+            Object.keys(options).map(k => `  --${k} ${options[k]}`).join('\n')
+        );
+        process.exit(0);
+    }
+    for (const k in opt)
+        options[k] = opt[k];
+    return devices;
+}
+
+// Collect devices
+async function collect_devices(devices_to_check)
+{
+    const devices = [];
+    for (const dev of devices_to_check)
+    {
+        if (dev.substr(0, 5) != '/dev/')
+        {
+            console.log(`${dev} does not start with /dev/, skipping`);
+            continue;
+        }
+        if (!await file_exists('/sys/block/'+dev.substr(5)))
+        {
+            console.log(`${dev} is a partition, skipping`);
+            continue;
+        }
+        // Check if the device is an SSD
+        const rot = '/sys/block/'+dev.substr(5)+'/queue/rotational';
+        if (!await file_exists(rot))
+        {
+            console.log(`${dev} does not have ${rot} to check whether it's an SSD, skipping`);
+            continue;
+        }
+        const ssd = !parseInt(await fsp.readFile(rot, { encoding: 'utf-8' }));
+        // Check if the device has partition table
+        let [ has_partition_table, parts ] = await system(`sfdisk --dump ${dev} --json`);
+        if (has_partition_table != 0)
+        {
+            // Check if the device has any data
+            const [ has_data, out ] = await system(`blkid ${dev}`);
+            if (has_data == 0)
+            {
+                console.log(`${dev} contains data, skipping:\n  ${out.trim().replace(/\n/g, '\n  ')}`);
+                continue;
+            }
+        }
+        parts = parts ? JSON.parse(parts).partitiontable : null;
+        if (parts && parts.label != 'gpt')
+        {
+            console.log(`${dev} contains "${parts.label}" partition table, only GPT is supported, skipping`);
+            continue;
+        }
+        devices.push({
+            path: dev,
+            ssd,
+            parts,
+        });
+    }
+    return devices;
+}
+
+// Collect existing OSD units
+async function collect_osd_units()
+{
+    const units = [];
+    for (const unit of (await system("ls /etc/systemd/system/vitastor-osd*.service"))[1].trim().split('\n'))
+    {
+        if (!unit)
+        {
+            continue;
+        }
+        let cmd = /^ExecStart\s*=\s*(([^\n]*\\\n)*[^\n]*)/.exec(await fsp.readFile(unit, { encoding: 'utf-8' }));
+        if (!cmd)
+        {
+            console.log('ExecStart= not found in '+unit+', skipping')
+            continue;
+        }
+        let kv = {}, key;
+        cmd = cmd[1].replace(/^bash\s+-c\s+'/, '')
+            .replace(/>>\s*\S+2>\s*&1\s*'$/, '')
+            .replace(/\s*\\\n\s*/g, ' ')
+            .replace(/([^\s']+)|'([^']+)'/g, (m, m1, m2) =>
+            {
+                m1 = m1||m2;
+                if (key == null)
+                {
+                    if (m1.substr(0, 2) != '--')
+                    {
+                        console.log('Strange command line in '+unit+', stopping');
+                        process.exit(1);
+                    }
+                    key = m1.substr(2);
+                }
+                else
+                {
+                    kv[key] = m1;
+                    key = null;
+                }
+            });
+        units.push(kv);
+    }
+    return units;
+}
+
+// Count assigned HDD journals and unallocated space for each SSD
+async function check_journal_count(ssds, osd_units)
+{
+    const units_by_journal = osd_units.reduce((a, c) =>
+    {
+        if (c.journal_device)
+            a[c.journal_device] = c;
+        return a;
+    }, {});
+    for (const dev of ssds)
+    {
+        dev.journals = 0;
+        if (dev.parts)
+        {
+            for (const part of dev.parts.partitions)
+            {
+                if (part.uuid && units_by_journal['/dev/disk/by-partuuid/'+part.uuid.toLowerCase()])
+                {
+                    dev.journals++;
+                }
+            }
+            dev.free = free_from_parttable(dev.parts);
+        }
+        else
+        {
+            dev.free = parseInt(await system_or_die("blockdev --getsize64 "+dev.path));
+        }
+    }
+}
+
+async function create_new_hybrid_osds(hdds, ssds, osd_units)
+{
+    const units_by_disk = osd_units.reduce((a, c) => { a[c.data_device] = c; return a; }, {});
+    for (const dev of hdds)
+    {
+        if (!dev.parts)
+        {
+            // HDD is not partitioned yet, create a single partition
+            // + is the "default value" for sfdisk
+            await system_or_die('sfdisk '+dev.path, 'label: gpt\n\n+ +\n');
+            dev.parts = JSON.parse(await system_or_die('sfdisk --dump '+dev.path+' --json')).partitiontable;
+        }
+        if (dev.parts.partitions.length != 1)
+        {
+            console.log(dev.path+' has more than 1 partition, skipping');
+        }
+        else if ((dev.parts.partitions[0].start + dev.parts.partitions[0].size) != (1 + dev.parts.lastlba))
+        {
+            console.log(dev.path+'1 is not a whole-disk partition, skipping');
+        }
+        else if (!dev.parts.partitions[0].uuid)
+        {
+            console.log(dev.parts.partitions[0].node+' does not have UUID. Please repartition '+dev.path+' with GPT');
+        }
+        else if (!units_by_disk['/dev/disk/by-partuuid/'+dev.parts.partitions[0].uuid.toLowerCase()])
+        {
+            await create_hybrid_osd(dev, ssds);
+        }
+    }
+}
+
+async function create_hybrid_osd(dev, ssds)
+{
+    // Create a new OSD
+    // Calculate metadata size
+    const data_device = '/dev/disk/by-partuuid/'+dev.parts.partitions[0].uuid.toLowerCase();
+    const data_size = dev.parts.partitions[0].size * dev.parts.sectorsize;
+    const meta_entry_size = 24 + 2*options.object_size/options.bitmap_granularity/8;
+    const entries_per_block = Math.floor(options.device_block_size / meta_entry_size);
+    const object_count = Math.floor(data_size / options.object_size);
+    let meta_size = Math.ceil(1 + object_count / entries_per_block) * options.device_block_size;
+    // Leave some extra space for future metadata formats and round metadata area size to multiples of 1 MB
+    meta_size = 2*meta_size;
+    meta_size = Math.ceil(meta_size/1024/1024) * 1024*1024;
+    if (meta_size < options.min_meta_size)
+        meta_size = options.min_meta_size;
+    let journal_size = Math.ceil(options.journal_size/1024/1024) * 1024*1024;
+    // Pick an SSD for journal, balancing the number of journals across SSDs
+    let selected_ssd;
+    for (const ssd of ssds)
+        if (ssd.free >= (meta_size+journal_size) && (!selected_ssd || selected_ssd.journals > ssd.journals))
+            selected_ssd = ssd;
+    if (!selected_ssd)
+    {
+        console.error('Could not find free space for SSD journal and metadata for '+dev.path);
+        process.exit(1);
+    }
+    // Allocate an OSD number
+    const osd_num = (await system_or_die("vitastor-cli alloc-osd")).trim();
+    if (!osd_num)
+    {
+        console.error('Failed to run vitastor-cli alloc-osd');
+        process.exit(1);
+    }
+    console.log('Creating OSD '+osd_num+' on '+dev.path+' (HDD) with journal and metadata on '+selected_ssd.path+' (SSD)');
+    // Add two partitions: journal and metadata
+    const new_parts = await add_partitions(selected_ssd, [ journal_size, meta_size ]);
+    selected_ssd.journals++;
+    const journal_device = '/dev/disk/by-partuuid/'+new_parts[0].uuid.toLowerCase();
+    const meta_device = '/dev/disk/by-partuuid/'+new_parts[1].uuid.toLowerCase();
+    // Wait until the device symlinks appear
+    while (!await file_exists(journal_device))
+    {
+        await new Promise(ok => setTimeout(ok, 100));
+    }
+    while (!await file_exists(meta_device))
+    {
+        await new Promise(ok => setTimeout(ok, 100));
+    }
+    // Zero out metadata and journal
+    await system_or_die("dd if=/dev/zero of="+journal_device+" bs=1M count="+(journal_size/1024/1024)+" oflag=direct");
+    await system_or_die("dd if=/dev/zero of="+meta_device+" bs=1M count="+(meta_size/1024/1024)+" oflag=direct");
+    // Create unit file for the OSD
+    const has_scsi_cache_type = options.disable_ssd_cache &&
+        (await system("ls /sys/block/"+selected_ssd.path.substr(5)+"/device/scsi_disk/*/cache_type"))[0] == 0;
+    const write_through = options.disable_ssd_cache && (
+        has_scsi_cache_type || selected_ssd.path.substr(5, 4) == 'nvme'
+        && (await system_or_die("/sys/block/"+selected_ssd.path.substr(5)+"/queue/write_cache")).trim() == "write through");
+    await fsp.writeFile('/etc/systemd/system/vitastor-osd'+osd_num+'.service',
+`[Unit]
+Description=Vitastor object storage daemon osd.${osd_num}
+After=network-online.target local-fs.target time-sync.target
+Wants=network-online.target local-fs.target time-sync.target
+PartOf=vitastor.target
+
+[Service]
+LimitNOFILE=1048576
+LimitNPROC=1048576
+LimitMEMLOCK=infinity
+ExecStart=bash -c '/usr/bin/vitastor-osd \\
+    --osd_num ${osd_num} ${write_through
+        ? "--disable_meta_fsync 1 --disable_journal_fsync 1 --immediate_commit "+(options.disable_hdd_cache ? "all" : "small")
+        : ""} \\
+    --throttle_small_writes 1 \\
+    --disk_alignment ${options.device_block_size} \\
+    --journal_block_size ${options.device_block_size} \\
+    --meta_block_size ${options.device_block_size} \\
+    --journal_no_same_sector_overwrites true \\
+    --journal_sector_buffer_count 1024 \\
+    --block_size ${options.object_size} \\
+    --data_device ${data_device} \\
+    --journal_device ${journal_device} \\
+    --meta_device ${meta_device} >>/var/log/vitastor/osd${osd_num}.log 2>&1'
+WorkingDirectory=/
+ExecStartPre=+chown vitastor:vitastor ${data_device}
+ExecStartPre=+chown vitastor:vitastor ${journal_device}
+ExecStartPre=+chown vitastor:vitastor ${meta_device}${
+    has_scsi_cache_type
+    ? "\nExecStartPre=+bash -c 'D=$$$(readlink "+journal_device+"); echo write through > $$$(dirname /sys/block/*/$$\${D##*/})/device/scsi_disk/*/cache_type'"
+    : ""}${
+    options.disable_hdd_cache
+    ? "\nExecStartPre=+bash -c 'D=$$$(readlink "+data_device+"); echo write through > $$$(dirname /sys/block/*/$$\${D##*/})/device/scsi_disk/*/cache_type'"
+    : ""}
+User=vitastor
+PrivateTmp=false
+TasksMax=infinity
+Restart=always
+StartLimitInterval=0
+RestartSec=10
+
+[Install]
+WantedBy=vitastor.target
+`);
+    await system_or_die("systemctl enable vitastor-osd"+osd_num);
+}
+
+async function add_partitions(dev, sizes)
+{
+    let script = 'label: gpt\n\n';
+    if (dev.parts)
+    {
+        // Old partitions
+        for (const part of dev.parts.partitions)
+        {
+            script += part.node+': '+Object.keys(part).map(k => k == 'node' ? '' : k+'='+part[k]).filter(k => k).join(', ')+'\n';
+        }
+    }
+    // New partitions
+    for (const size of sizes)
+    {
+        script += '+ '+Math.ceil(size/1024)+'KiB\n';
+    }
+    await system_or_die('sfdisk '+dev.path, script);
+    // Get new partition table and find the new partition
+    const newpt = JSON.parse(await system_or_die('sfdisk --dump '+dev.path+' --json')).partitiontable;
+    const old_nodes = dev.parts ? dev.parts.partitions.reduce((a, c) => { a[c.uuid] = true; return a; }, {}) : {};
+    const new_nodes = newpt.partitions.filter(part => !old_nodes[part.uuid]);
+    if (new_nodes.length != sizes.length)
+    {
+        console.error('Failed to partition '+dev.path+': new partitions not found in table');
+        process.exit(1);
+    }
+    dev.parts = newpt;
+    dev.free = free_from_parttable(newpt);
+    return new_nodes;
+}
+
+function free_from_parttable(pt)
+{
+    let free = pt.lastlba + 1 - pt.firstlba;
+    for (const part of pt.partitions)
+    {
+        free -= part.size;
+    }
+    free *= pt.sectorsize;
+    return free;
+}
+
+async function system_or_die(cmd, input = '')
+{
+    let [ exitcode, stdout, stderr ] = await system(cmd, input);
+    if (exitcode != 0)
+    {
+        console.error(cmd+' failed: '+stderr);
+        process.exit(1);
+    }
+    return stdout;
+}
+
+async function system(cmd, input = '')
+{
+    if (options.debug)
+    {
+        process.stderr.write('+ '+cmd+(input ? " <<EOF\n"+input.replace(/\s*$/, '\n')+"EOF" : '')+'\n');
+    }
+    const cp = child_process.spawn(cmd, { shell: true });
+    let stdout = '', stderr = '', finish_cb;
+    cp.stdout.on('data', buf => stdout += buf.toString());
+    cp.stderr.on('data', buf => stderr += buf.toString());
+    cp.on('exit', () => finish_cb && finish_cb());
+    cp.stdin.write(input);
+    cp.stdin.end();
+    if (cp.exitCode == null)
+    {
+        await new Promise(ok => finish_cb = ok);
+    }
+    return [ cp.exitCode, stdout, stderr ];
+}
+
+async function file_exists(filename)
+{
+    return new Promise((ok, no) => fs.access(filename, fs.constants.R_OK, err => ok(!err)));
+}

mon/make-osd.sh

@@ -0,0 +1,66 @@
+#!/bin/bash
+# Very simple systemd unit generator for vitastor-osd services
+# Not the final solution yet, mostly for tests
+# Copyright (c) Vitaliy Filippov, 2019+
+# License: MIT
+
+# USAGE:
+# 1) Put etcd_address and osd_network into /etc/vitastor/vitastor.conf. Example:
+#    {
+#        "etcd_address":["http://10.200.1.10:2379/v3","http://10.200.1.11:2379/v3","http://10.200.1.12:2379/v3"],
+#        "osd_network":"10.200.1.0/24"
+#    }
+# 2) Run ./make-osd.sh /dev/disk/by-partuuid/xxx [ /dev/disk/by-partuuid/yyy]...
+
+set -e -x
+
+# Create OSDs on all passed devices
+for DEV in $*; do
+
+OSD_NUM=$(vitastor-cli alloc-osd)
+
+echo Creating OSD $OSD_NUM on $DEV
+
+OPT=$(vitastor-cli simple-offsets --format options $DEV | tr '\n' ' ')
+META=$(vitastor-cli simple-offsets --format json $DEV | jq .data_offset)
+dd if=/dev/zero of=$DEV bs=1048576 count=$(((META+1048575)/1048576)) oflag=direct
+
+mkdir -p /var/log/vitastor
+id vitastor &>/dev/null || useradd vitastor
+chown vitastor /var/log/vitastor
+
+cat >/etc/systemd/system/vitastor-osd$OSD_NUM.service <<EOF
+[Unit]
+Description=Vitastor object storage daemon osd.$OSD_NUM
+After=network-online.target local-fs.target time-sync.target
+Wants=network-online.target local-fs.target time-sync.target
+PartOf=vitastor.target
+
+[Service]
+LimitNOFILE=1048576
+LimitNPROC=1048576
+LimitMEMLOCK=infinity
+ExecStart=bash -c '/usr/bin/vitastor-osd \\
+    --osd_num $OSD_NUM \\
+    --disable_data_fsync 1 \\
+    --immediate_commit all \\
+    --disk_alignment 4096 --journal_block_size 4096 --meta_block_size 4096 \\
+    --journal_no_same_sector_overwrites true \\
+    --journal_sector_buffer_count 1024 \\
+    $OPT >>/var/log/vitastor/osd$OSD_NUM.log 2>&1'
+WorkingDirectory=/
+ExecStartPre=+chown vitastor:vitastor $DEV
+User=vitastor
+PrivateTmp=false
+TasksMax=infinity
+Restart=always
+StartLimitInterval=0
+RestartSec=10
+
+[Install]
+WantedBy=vitastor.target
+EOF
+
+systemctl enable vitastor-osd$OSD_NUM
+
+done

mon/make-units.sh

@@ -0,0 +1,86 @@
+#!/bin/bash
+# Very simple systemd unit generator for etcd & vitastor-mon services
+# Not the final solution yet, mostly for tests
+# Copyright (c) Vitaliy Filippov, 2019+
+# License: MIT
+
+# USAGE: ./make-units.sh
+
+IP_SUBSTR="10.200.1."
+ETCD_HOSTS="etcd0=http://10.200.1.10:2380,etcd1=http://10.200.1.11:2380,etcd2=http://10.200.1.12:2380"
+
+# determine IP
+IP=`ip -json a s | jq -r '.[].addr_info[] | select(.local | startswith("'$IP_SUBSTR'")) | .local'`
+[ "$IP" != "" ] || exit 1
+ETCD_NUM=${ETCD_HOSTS/$IP*/}
+[ "$ETCD_NUM" != "$ETCD_HOSTS" ] || exit 1
+ETCD_NUM=$(echo $ETCD_NUM | tr -d -c , | wc -c)
+
+# etcd
+useradd etcd
+
+mkdir -p /var/lib/etcd$ETCD_NUM.etcd
+cat >/etc/systemd/system/etcd.service <<EOF
+[Unit]
+Description=etcd for vitastor
+After=network-online.target local-fs.target time-sync.target
+Wants=network-online.target local-fs.target time-sync.target
+
+[Service]
+Restart=always
+ExecStart=/usr/local/bin/etcd -name etcd$ETCD_NUM --data-dir /var/lib/etcd$ETCD_NUM.etcd \\
+    --advertise-client-urls http://$IP:2379 --listen-client-urls http://$IP:2379 \\
+    --initial-advertise-peer-urls http://$IP:2380 --listen-peer-urls http://$IP:2380 \\
+    --initial-cluster-token vitastor-etcd-1 --initial-cluster $ETCD_HOSTS \\
+    --initial-cluster-state new --max-txn-ops=100000 --max-request-bytes=104857600 \\
+    --auto-compaction-retention=10 --auto-compaction-mode=revision
+WorkingDirectory=/var/lib/etcd$ETCD_NUM.etcd
+ExecStartPre=+chown -R etcd /var/lib/etcd$ETCD_NUM.etcd
+User=etcd
+PrivateTmp=false
+TasksMax=infinity
+Restart=always
+StartLimitInterval=0
+RestartSec=10
+
+[Install]
+WantedBy=local.target
+EOF
+
+systemctl daemon-reload
+systemctl enable etcd
+systemctl start etcd
+
+useradd vitastor
+chmod 755 /root
+
+# Vitastor target
+cat >/etc/systemd/system/vitastor.target <<EOF
+[Unit]
+Description=vitastor target
+[Install]
+WantedBy=multi-user.target
+EOF
+
+# Monitor unit
+ETCD_MON=$(echo $ETCD_HOSTS | perl -pe 's/:2380/:2379/g; s/etcd\d*=//g;')
+cat >/etc/systemd/system/vitastor-mon.service <<EOF
+[Unit]
+Description=Vitastor monitor
+After=network-online.target local-fs.target time-sync.target
+Wants=network-online.target local-fs.target time-sync.target
+
+[Service]
+Restart=always
+ExecStart=node /usr/lib/vitastor/mon/mon-main.js --etcd_url '$ETCD_MON' --etcd_prefix '/vitastor' --etcd_start_timeout 5
+WorkingDirectory=/
+User=vitastor
+PrivateTmp=false
+TasksMax=infinity
+Restart=always
+StartLimitInterval=0
+RestartSec=10
+
+[Install]
+WantedBy=vitastor.target
+EOF

mon/mon-main.js

@@ -12,7 +12,7 @@ for (let i = 2; i < process.argv.length; i++)
     if (process.argv[i] === '-h' || process.argv[i] === '--help')
     {
         console.error('USAGE: '+process.argv[0]+' '+process.argv[1]+' [--verbose 1]'+
-            ' [--etcd_address "http://127.0.0.1:2379,..."] [--config_path /etc/vitastor/vitastor.conf]'+
+            ' [--etcd_address "http://127.0.0.1:2379,..."] [--config_file /etc/vitastor/vitastor.conf]'+
             ' [--etcd_prefix "/vitastor"] [--etcd_start_timeout 5]');
         process.exit();
     }

mon/mon.js

@@ -157,12 +157,7 @@ const etcd_tree = {
                 pg_count: 100,
                 failure_domain: 'host',
                 max_osd_combinations: 10000,
-                // block_size, bitmap_granularity, immediate_commit must match all OSDs used in that pool
-                block_size: 131072,
-                bitmap_granularity: 4096,
-                // 'all'/'small'/'none', same as in OSD options
-                immediate_commit: 'none',
-                pg_stripe_size: 0,
+                pg_stripe_size: 4194304,
                 root_node?: 'rack1',
                 // restrict pool to OSDs having all of these tags
                 osd_tags?: 'nvme' | [ 'nvme', ... ],
@@ -328,13 +323,6 @@ const etcd_tree = {
             misplaced: uint64_t,
             degraded: uint64_t,
             incomplete: uint64_t,
-        },
-        object_bytes: {
-            total: uint64_t,
-            clean: uint64_t,
-            misplaced: uint64_t,
-            degraded: uint64_t,
-            incomplete: uint64_t,
         }, */
     },
     history: {
@@ -663,15 +651,12 @@ class Mon
     async save_last_clean()
     {
         // last_clean_pgs is used to avoid extra data move when observing a series of changes in the cluster
-        const new_clean_pgs = { items: {} };
-    next_pool:
         for (const pool_id in this.state.config.pools)
         {
-            new_clean_pgs.items[pool_id] = (this.state.history.last_clean_pgs.items||{})[pool_id];
             const pool_cfg = this.state.config.pools[pool_id];
             if (!this.validate_pool_cfg(pool_id, pool_cfg, false))
             {
-                continue next_pool;
+                continue;
             }
             for (let pg_num = 1; pg_num <= pool_cfg.pg_count; pg_num++)
             {
@@ -680,18 +665,17 @@ class Mon
                     !(this.state.pg.state[pool_id][pg_num].state instanceof Array))
                 {
                     // Unclean
-                    continue next_pool;
+                    return;
                 }
                 let st = this.state.pg.state[pool_id][pg_num].state.join(',');
                 if (st != 'active' && st != 'active,left_on_dead' && st != 'left_on_dead,active')
                 {
                     // Unclean
-                    continue next_pool;
+                    return;
                 }
             }
-            new_clean_pgs.items[pool_id] = this.state.config.pgs.items[pool_id];
         }
-        this.state.history.last_clean_pgs = new_clean_pgs;
+        this.state.history.last_clean_pgs = JSON.parse(JSON.stringify(this.state.config.pgs));
         await this.etcd_call('/kv/txn', {
             success: [ { requestPut: {
                 key: b64(this.etcd_prefix+'/history/last_clean_pgs'),
@@ -1378,14 +1362,16 @@ class Mon
     // This is required for multiple change events to trigger at most 1 recheck in 1s
     schedule_recheck()
     {
-        if (!this.recheck_timer)
+        if (this.recheck_timer)
         {
-            this.recheck_timer = setTimeout(() =>
-            {
-                this.recheck_timer = null;
-                this.recheck_pgs().catch(this.die);
-            }, this.config.mon_change_timeout || 1000);
+            clearTimeout(this.recheck_timer);
+            this.recheck_timer = null;
         }
+        this.recheck_timer = setTimeout(() =>
+        {
+            this.recheck_timer = null;
+            this.recheck_pgs().catch(this.die);
+        }, this.config.mon_change_timeout || 1000);
     }
 
     sum_op_stats(timestamp, prev_stats)
@@ -1452,24 +1438,8 @@ class Mon
     sum_object_counts()
     {
         const object_counts = { object: 0n, clean: 0n, misplaced: 0n, degraded: 0n, incomplete: 0n };
-        const object_bytes = { object: 0n, clean: 0n, misplaced: 0n, degraded: 0n, incomplete: 0n };
         for (const pool_id in this.state.pg.stats)
         {
-            let object_size = 0;
-            for (const osd_num of this.state.pg.stats[pool_id].write_osd_set||[])
-            {
-                if (osd_num && this.state.osd.stats[osd_num] && this.state.osd.stats[osd_num].block_size)
-                {
-                    object_size = this.state.osd.stats[osd_num].block_size;
-                    break;
-                }
-            }
-            if (!object_size)
-            {
-                object_size = (this.state.config.pools[pool_id]||{}).block_size ||
-                    this.config.block_size || 131072;
-            }
-            object_size = BigInt(object_size);
             for (const pg_num in this.state.pg.stats[pool_id])
             {
                 const st = this.state.pg.stats[pool_id][pg_num];
@@ -1480,13 +1450,12 @@ class Mon
                         if (st[k+'_count'])
                         {
                             object_counts[k] += BigInt(st[k+'_count']);
-                            object_bytes[k] += BigInt(st[k+'_count']) * object_size;
                         }
                     }
                 }
             }
         }
-        return { object_counts, object_bytes };
+        return object_counts;
     }
 
     sum_inode_stats(prev_stats, timestamp, prev_timestamp)
@@ -1599,7 +1568,7 @@ class Mon
     {
         const txn = [];
         const timestamp = Date.now();
-        const { object_counts, object_bytes } = this.sum_object_counts();
+        const object_counts = this.sum_object_counts();
         let stats = this.sum_op_stats(timestamp, this.prev_stats);
         let inode_stats = this.sum_inode_stats(
             this.prev_stats ? this.prev_stats.inode_stats : null,
@@ -1607,7 +1576,6 @@ class Mon
         );
         this.prev_stats = { timestamp, ...stats, inode_stats };
         stats.object_counts = object_counts;
-        stats.object_bytes = object_bytes;
         stats = this.serialize_bigints(stats);
         inode_stats = this.serialize_bigints(inode_stats);
         txn.push({ requestPut: { key: b64(this.etcd_prefix+'/stats'), value: b64(JSON.stringify(stats)) } });
@@ -1695,7 +1663,6 @@ class Mon
             // Do not clear these to null
             kv.value = kv.value || {};
         }
-        const old = cur[key_parts[key_parts.length-1]];
         cur[key_parts[key_parts.length-1]] = kv.value;
         if (key === 'config/global')
         {
@@ -1720,12 +1687,7 @@ class Mon
         }
         else if (key_parts[0] === 'osd' && key_parts[1] === 'stats')
         {
-            // Recheck OSD tree on OSD addition/deletion
-            if ((!old) != (!kv.value) || old && kv.value && old.size != kv.value.size)
-            {
-                this.schedule_recheck();
-            }
-            // Recheck PGs <osd_out_time> after last OSD statistics report
+            // Recheck PGs <osd_out_time> later
             this.schedule_next_recheck_at(
                 !this.state.osd.stats[key[2]] ? 0 : this.state.osd.stats[key[2]].time+this.config.osd_out_time
             );

mon/vitastor-mon.service

@@ -1,18 +0,0 @@
-[Unit]
-Description=Vitastor monitor
-After=network-online.target local-fs.target time-sync.target
-Wants=network-online.target local-fs.target time-sync.target
-
-[Service]
-Restart=always
-ExecStart=node /usr/lib/vitastor/mon/mon-main.js
-WorkingDirectory=/
-User=vitastor
-PrivateTmp=false
-TasksMax=infinity
-Restart=always
-StartLimitInterval=0
-RestartSec=10
-
-[Install]
-WantedBy=vitastor.target

mon/vitastor-osd@.service

@@ -1,22 +0,0 @@
-[Unit]
-Description=Vitastor object storage daemon osd.%i
-After=network-online.target local-fs.target time-sync.target
-Wants=network-online.target local-fs.target time-sync.target
-PartOf=vitastor.target
-
-[Service]
-LimitNOFILE=1048576
-LimitNPROC=1048576
-LimitMEMLOCK=infinity
-ExecStart=bash -c 'exec vitastor-disk exec-osd /dev/vitastor/osd%i-data >>/var/log/vitastor/osd%i.log 2>&1'
-ExecStartPre=+vitastor-disk pre-exec /dev/vitastor/osd%i-data
-WorkingDirectory=/
-User=vitastor
-PrivateTmp=false
-TasksMax=infinity
-Restart=always
-StartLimitInterval=0
-RestartSec=10
-
-[Install]
-WantedBy=vitastor.target

mon/vitastor.target

@@ -1,4 +0,0 @@
-[Unit]
-Description=vitastor target
-[Install]
-WantedBy=multi-user.target

patches/cinder-vitastor.py

@@ -50,7 +50,7 @@ from cinder.volume import configuration
 from cinder.volume import driver
 from cinder.volume import volume_utils
 
-VERSION = '0.8.3'
+VERSION = '0.7.1'
 
 LOG = logging.getLogger(__name__)
 

patches/pve-qemu-6.2-vitastor.patch

@@ -1,169 +0,0 @@
-Index: qemu/block/meson.build
-===================================================================
---- qemu.orig/block/meson.build
-+++ qemu/block/meson.build
-@@ -91,6 +91,7 @@ foreach m : [
-   [libnfs, 'nfs', files('nfs.c')],
-   [libssh, 'ssh', files('ssh.c')],
-   [rbd, 'rbd', files('rbd.c')],
-+  [vitastor, 'vitastor', files('vitastor.c')],
- ]
-   if m[0].found()
-     module_ss = ss.source_set()
-Index: qemu/meson.build
-===================================================================
---- qemu.orig/meson.build
-+++ qemu/meson.build
-@@ -838,6 +838,26 @@ if not get_option('rbd').auto() or have_
-   endif
- endif
- 
-+vitastor = not_found
-+if not get_option('vitastor').auto() or have_block
-+  libvitastor_client = cc.find_library('vitastor_client', has_headers: ['vitastor_c.h'],
-+    required: get_option('vitastor'), kwargs: static_kwargs)
-+  if libvitastor_client.found()
-+    if cc.links('''
-+      #include <vitastor_c.h>
-+      int main(void) {
-+        vitastor_c_create_qemu(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
-+        return 0;
-+      }''', dependencies: libvitastor_client)
-+      vitastor = declare_dependency(dependencies: libvitastor_client)
-+    elif get_option('vitastor').enabled()
-+      error('could not link libvitastor_client')
-+    else
-+      warning('could not link libvitastor_client, disabling')
-+    endif
-+  endif
-+endif
-+
- glusterfs = not_found
- glusterfs_ftruncate_has_stat = false
- glusterfs_iocb_has_stat = false
-@@ -1459,6 +1479,7 @@ config_host_data.set('CONFIG_LINUX_AIO',
- config_host_data.set('CONFIG_LINUX_IO_URING', linux_io_uring.found())
- config_host_data.set('CONFIG_LIBPMEM', libpmem.found())
- config_host_data.set('CONFIG_RBD', rbd.found())
-+config_host_data.set('CONFIG_VITASTOR', vitastor.found())
- config_host_data.set('CONFIG_SDL', sdl.found())
- config_host_data.set('CONFIG_SDL_IMAGE', sdl_image.found())
- config_host_data.set('CONFIG_SECCOMP', seccomp.found())
-@@ -3424,6 +3445,7 @@ if spice_protocol.found()
-   summary_info += {'  spice server support': spice}
- endif
- summary_info += {'rbd support':       rbd}
-+summary_info += {'vitastor support':  vitastor}
- summary_info += {'xfsctl support':    config_host.has_key('CONFIG_XFS')}
- summary_info += {'smartcard support': cacard}
- summary_info += {'U2F support':       u2f}
-Index: qemu/meson_options.txt
-===================================================================
---- qemu.orig/meson_options.txt
-+++ qemu/meson_options.txt
-@@ -121,6 +121,8 @@ option('lzo', type : 'feature', value :
-        description: 'lzo compression support')
- option('rbd', type : 'feature', value : 'auto',
-        description: 'Ceph block device driver')
-+option('vitastor', type : 'feature', value : 'auto',
-+       description: 'Vitastor block device driver')
- option('gtk', type : 'feature', value : 'auto',
-        description: 'GTK+ user interface')
- option('sdl', type : 'feature', value : 'auto',
-Index: qemu/qapi/block-core.json
-===================================================================
---- qemu.orig/qapi/block-core.json
-+++ qemu/qapi/block-core.json
-@@ -3179,7 +3179,7 @@
-             'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
-             { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
-             'pbs',
--            'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
-+            'ssh', 'throttle', 'vdi', 'vhdx', 'vitastor', 'vmdk', 'vpc', 'vvfat' ] }
- 
- ##
- # @BlockdevOptionsFile:
-@@ -4125,6 +4125,28 @@
-             '*server': ['InetSocketAddressBase'] } }
- 
- ##
-+# @BlockdevOptionsVitastor:
-+#
-+# Driver specific block device options for vitastor
-+#
-+# @image:       Image name
-+# @inode:       Inode number
-+# @pool:        Pool ID
-+# @size:        Desired image size in bytes
-+# @config-path: Path to Vitastor configuration
-+# @etcd-host:   etcd connection address(es)
-+# @etcd-prefix: etcd key/value prefix
-+##
-+{ 'struct': 'BlockdevOptionsVitastor',
-+  'data': { '*inode': 'uint64',
-+            '*pool': 'uint64',
-+            '*size': 'uint64',
-+            '*image': 'str',
-+            '*config-path': 'str',
-+            '*etcd-host': 'str',
-+            '*etcd-prefix': 'str' } }
-+
-+##
- # @ReplicationMode:
- #
- # An enumeration of replication modes.
-@@ -4520,6 +4542,7 @@
-       'throttle':   'BlockdevOptionsThrottle',
-       'vdi':        'BlockdevOptionsGenericFormat',
-       'vhdx':       'BlockdevOptionsGenericFormat',
-+      'vitastor':   'BlockdevOptionsVitastor',
-       'vmdk':       'BlockdevOptionsGenericCOWFormat',
-       'vpc':        'BlockdevOptionsGenericFormat',
-       'vvfat':      'BlockdevOptionsVVFAT'
-@@ -4910,6 +4933,17 @@
-             '*encrypt' :        'RbdEncryptionCreateOptions' } }
- 
- ##
-+# @BlockdevCreateOptionsVitastor:
-+#
-+# Driver specific image creation options for Vitastor.
-+#
-+# @size: Size of the virtual disk in bytes
-+##
-+{ 'struct': 'BlockdevCreateOptionsVitastor',
-+  'data': { 'location':         'BlockdevOptionsVitastor',
-+            'size':             'size' } }
-+
-+##
- # @BlockdevVmdkSubformat:
- #
- # Subformat options for VMDK images
-@@ -5108,6 +5142,7 @@
-       'ssh':            'BlockdevCreateOptionsSsh',
-       'vdi':            'BlockdevCreateOptionsVdi',
-       'vhdx':           'BlockdevCreateOptionsVhdx',
-+      'vitastor':       'BlockdevCreateOptionsVitastor',
-       'vmdk':           'BlockdevCreateOptionsVmdk',
-       'vpc':            'BlockdevCreateOptionsVpc'
-   } }
-Index: qemu/scripts/ci/org.centos/stream/8/x86_64/configure
-===================================================================
---- qemu.orig/scripts/ci/org.centos/stream/8/x86_64/configure
-+++ qemu/scripts/ci/org.centos/stream/8/x86_64/configure
-@@ -31,7 +31,7 @@
- --with-git=meson \
- --with-git-submodules=update \
- --target-list="x86_64-softmmu" \
----block-drv-rw-whitelist="qcow2,raw,file,host_device,nbd,iscsi,rbd,blkdebug,luks,null-co,nvme,copy-on-read,throttle,gluster" \
-+--block-drv-rw-whitelist="qcow2,raw,file,host_device,nbd,iscsi,rbd,vitastor,blkdebug,luks,null-co,nvme,copy-on-read,throttle,gluster" \
- --audio-drv-list="" \
- --block-drv-ro-whitelist="vmdk,vhdx,vpc,https,ssh" \
- --with-coroutine=ucontext \
-@@ -183,6 +183,7 @@
- --enable-opengl \
- --enable-pie \
- --enable-rbd \
-+--enable-vitastor \
- --enable-rdma \
- --enable-seccomp \
- --enable-snappy \

patches/pve-qemu-7.1-vitastor.patch

@@ -1,169 +0,0 @@
-Index: qemu/block/meson.build
-===================================================================
---- qemu.orig/block/meson.build
-+++ qemu/block/meson.build
-@@ -111,6 +111,7 @@ foreach m : [
-   [libnfs, 'nfs', files('nfs.c')],
-   [libssh, 'ssh', files('ssh.c')],
-   [rbd, 'rbd', files('rbd.c')],
-+  [vitastor, 'vitastor', files('vitastor.c')],
- ]
-   if m[0].found()
-     module_ss = ss.source_set()
-Index: qemu/meson.build
-===================================================================
---- qemu.orig/meson.build
-+++ qemu/meson.build
-@@ -967,6 +967,26 @@ if not get_option('rbd').auto() or have_
-   endif
- endif
- 
-+vitastor = not_found
-+if not get_option('vitastor').auto() or have_block
-+  libvitastor_client = cc.find_library('vitastor_client', has_headers: ['vitastor_c.h'],
-+    required: get_option('vitastor'), kwargs: static_kwargs)
-+  if libvitastor_client.found()
-+    if cc.links('''
-+      #include <vitastor_c.h>
-+      int main(void) {
-+        vitastor_c_create_qemu(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
-+        return 0;
-+      }''', dependencies: libvitastor_client)
-+      vitastor = declare_dependency(dependencies: libvitastor_client)
-+    elif get_option('vitastor').enabled()
-+      error('could not link libvitastor_client')
-+    else
-+      warning('could not link libvitastor_client, disabling')
-+    endif
-+  endif
-+endif
-+
- glusterfs = not_found
- glusterfs_ftruncate_has_stat = false
- glusterfs_iocb_has_stat = false
-@@ -1802,6 +1822,7 @@ config_host_data.set('CONFIG_NUMA', numa
- config_host_data.set('CONFIG_OPENGL', opengl.found())
- config_host_data.set('CONFIG_PROFILER', get_option('profiler'))
- config_host_data.set('CONFIG_RBD', rbd.found())
-+config_host_data.set('CONFIG_VITASTOR', vitastor.found())
- config_host_data.set('CONFIG_RDMA', rdma.found())
- config_host_data.set('CONFIG_SDL', sdl.found())
- config_host_data.set('CONFIG_SDL_IMAGE', sdl_image.found())
-@@ -3965,6 +3986,7 @@ if spice_protocol.found()
-   summary_info += {'  spice server support': spice}
- endif
- summary_info += {'rbd support':       rbd}
-+summary_info += {'vitastor support':  vitastor}
- summary_info += {'smartcard support': cacard}
- summary_info += {'U2F support':       u2f}
- summary_info += {'libusb':            libusb}
-Index: qemu/meson_options.txt
-===================================================================
---- qemu.orig/meson_options.txt
-+++ qemu/meson_options.txt
-@@ -167,6 +167,8 @@ option('lzo', type : 'feature', value :
-        description: 'lzo compression support')
- option('rbd', type : 'feature', value : 'auto',
-        description: 'Ceph block device driver')
-+option('vitastor', type : 'feature', value : 'auto',
-+       description: 'Vitastor block device driver')
- option('opengl', type : 'feature', value : 'auto',
-        description: 'OpenGL support')
- option('rdma', type : 'feature', value : 'auto',
-Index: qemu/qapi/block-core.json
-===================================================================
---- qemu.orig/qapi/block-core.json
-+++ qemu/qapi/block-core.json
-@@ -3209,7 +3209,7 @@
-             'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
-             { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
-             'pbs',
--            'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
-+            'ssh', 'throttle', 'vdi', 'vhdx', 'vitastor', 'vmdk', 'vpc', 'vvfat' ] }
- 
- ##
- # @BlockdevOptionsFile:
-@@ -4149,6 +4149,28 @@
-             '*server': ['InetSocketAddressBase'] } }
- 
- ##
-+# @BlockdevOptionsVitastor:
-+#
-+# Driver specific block device options for vitastor
-+#
-+# @image:       Image name
-+# @inode:       Inode number
-+# @pool:        Pool ID
-+# @size:        Desired image size in bytes
-+# @config-path: Path to Vitastor configuration
-+# @etcd-host:   etcd connection address(es)
-+# @etcd-prefix: etcd key/value prefix
-+##
-+{ 'struct': 'BlockdevOptionsVitastor',
-+  'data': { '*inode': 'uint64',
-+            '*pool': 'uint64',
-+            '*size': 'uint64',
-+            '*image': 'str',
-+            '*config-path': 'str',
-+            '*etcd-host': 'str',
-+            '*etcd-prefix': 'str' } }
-+
-+##
- # @ReplicationMode:
- #
- # An enumeration of replication modes.
-@@ -4593,6 +4615,7 @@
-       'throttle':   'BlockdevOptionsThrottle',
-       'vdi':        'BlockdevOptionsGenericFormat',
-       'vhdx':       'BlockdevOptionsGenericFormat',
-+      'vitastor':   'BlockdevOptionsVitastor',
-       'vmdk':       'BlockdevOptionsGenericCOWFormat',
-       'vpc':        'BlockdevOptionsGenericFormat',
-       'vvfat':      'BlockdevOptionsVVFAT'
-@@ -4985,6 +5008,17 @@
-             '*encrypt' :        'RbdEncryptionCreateOptions' } }
- 
- ##
-+# @BlockdevCreateOptionsVitastor:
-+#
-+# Driver specific image creation options for Vitastor.
-+#
-+# @size: Size of the virtual disk in bytes
-+##
-+{ 'struct': 'BlockdevCreateOptionsVitastor',
-+  'data': { 'location':         'BlockdevOptionsVitastor',
-+            'size':             'size' } }
-+
-+##
- # @BlockdevVmdkSubformat:
- #
- # Subformat options for VMDK images
-@@ -5182,6 +5216,7 @@
-       'ssh':            'BlockdevCreateOptionsSsh',
-       'vdi':            'BlockdevCreateOptionsVdi',
-       'vhdx':           'BlockdevCreateOptionsVhdx',
-+      'vitastor':       'BlockdevCreateOptionsVitastor',
-       'vmdk':           'BlockdevCreateOptionsVmdk',
-       'vpc':            'BlockdevCreateOptionsVpc'
-   } }
-Index: qemu/scripts/ci/org.centos/stream/8/x86_64/configure
-===================================================================
---- qemu.orig/scripts/ci/org.centos/stream/8/x86_64/configure
-+++ qemu/scripts/ci/org.centos/stream/8/x86_64/configure
-@@ -31,7 +31,7 @@
- --with-git=meson \
- --with-git-submodules=update \
- --target-list="x86_64-softmmu" \
----block-drv-rw-whitelist="qcow2,raw,file,host_device,nbd,iscsi,rbd,blkdebug,luks,null-co,nvme,copy-on-read,throttle,gluster" \
-+--block-drv-rw-whitelist="qcow2,raw,file,host_device,nbd,iscsi,rbd,vitastor,blkdebug,luks,null-co,nvme,copy-on-read,throttle,gluster" \
- --audio-drv-list="" \
- --block-drv-ro-whitelist="vmdk,vhdx,vpc,https,ssh" \
- --with-coroutine=ucontext \
-@@ -179,6 +179,7 @@
- --enable-opengl \
- --enable-pie \
- --enable-rbd \
-+--enable-vitastor \
- --enable-rdma \
- --enable-seccomp \
- --enable-snappy \

patches/qemu-make-patches.sh

@@ -9,7 +9,7 @@ for i in "$DIR"/qemu-*-vitastor.patch "$DIR"/pve-qemu-*-vitastor.patch; do
         echo '===================================================================' >> $i
         echo '--- /dev/null' >> $i
         echo '+++ a/block/vitastor.c' >> $i
-        echo '@@ -0,0 +1,'$(wc -l "$DIR"/../src/qemu_driver.c | cut -d ' ' -f 1)' @@' >> $i
+        echo '@@ -0,0 +1,'$(wc -l "$DIR"/../src/qemu_driver.c)' @@' >> $i
         cat "$DIR"/../src/qemu_driver.c | sed 's/^/+/' >> $i
     fi
 done

rpm/build-tarball.sh

@@ -25,4 +25,4 @@ rm fio
 mv fio-copy fio
 FIO=`rpm -qi fio | perl -e 'while(<>) { /^Epoch[\s:]+(\S+)/ && print "$1:"; /^Version[\s:]+(\S+)/ && print $1; /^Release[\s:]+(\S+)/ && print "-$1"; }'`
 perl -i -pe 's/(Requires:\s*fio)([^\n]+)?/$1 = '$FIO'/' $VITASTOR/rpm/vitastor-el$EL.spec
-tar --transform 's#^#vitastor-0.8.3/#' --exclude 'rpm/*.rpm' -czf $VITASTOR/../vitastor-0.8.3$(rpm --eval '%dist').tar.gz *
+tar --transform 's#^#vitastor-0.7.1/#' --exclude 'rpm/*.rpm' -czf $VITASTOR/../vitastor-0.7.1$(rpm --eval '%dist').tar.gz *

rpm/qemu-kvm-4.2-el7.spec.patch

@@ -58,7 +58,7 @@
 +BuildRequires: gperftools-devel
 +BuildRequires: libusbx-devel >= 1.0.21
  %if %{have_usbredir}
- BuildRequires: usbredir-devel >= 0.8.2
+ BuildRequires: usbredir-devel >= 0.7.1
  %endif
 @@ -856,12 +861,13 @@ BuildRequires: virglrenderer-devel
  # For smartcard NSS support

rpm/vitastor-el7.Dockerfile

@@ -35,7 +35,7 @@ ADD . /root/vitastor
 RUN set -e; \
     cd /root/vitastor/rpm; \
     sh build-tarball.sh; \
-    cp /root/vitastor-0.8.3.el7.tar.gz ~/rpmbuild/SOURCES; \
+    cp /root/vitastor-0.7.1.el7.tar.gz ~/rpmbuild/SOURCES; \
     cp vitastor-el7.spec ~/rpmbuild/SPECS/vitastor.spec; \
     cd ~/rpmbuild/SPECS/; \
     rpmbuild -ba vitastor.spec; \

rpm/vitastor-el7.spec

@@ -1,11 +1,11 @@
 Name:           vitastor
-Version:        0.8.3
+Version:        0.7.1
 Release:        1%{?dist}
 Summary:        Vitastor, a fast software-defined clustered block storage
 
 License:        Vitastor Network Public License 1.1
 URL:            https://vitastor.io/
-Source0:        vitastor-0.8.3.el7.tar.gz
+Source0:        vitastor-0.7.1.el7.tar.gz
 
 BuildRequires:  liburing-devel >= 0.6
 BuildRequires:  gperftools-devel
@@ -36,8 +36,6 @@ Requires:       libJerasure2
 Requires:       libisa-l
 Requires:       liburing >= 0.6
 Requires:       vitastor-client = %{version}-%{release}
-Requires:       util-linux
-Requires:       parted
 
 
 %description -n vitastor-osd
@@ -104,11 +102,8 @@ cd mon
 npm install
 cd ..
 mkdir -p %buildroot/usr/lib/vitastor
+cp mon/make-osd.sh %buildroot/usr/lib/vitastor
 cp -r mon %buildroot/usr/lib/vitastor
-mkdir -p %buildroot/lib/systemd/system
-cp mon/vitastor.target mon/vitastor-mon.service mon/vitastor-osd@.service %buildroot/lib/systemd/system
-mkdir -p %buildroot/lib/udev/rules.d
-cp mon/90-vitastor.rules %buildroot/lib/udev/rules.d
 
 
 %files
@@ -119,27 +114,10 @@ cp mon/90-vitastor.rules %buildroot/lib/udev/rules.d
 %_bindir/vitastor-osd
 %_bindir/vitastor-disk
 %_bindir/vitastor-dump-journal
-/lib/systemd/system/vitastor-osd@.service
-/lib/systemd/system/vitastor.target
-/lib/udev/rules.d/90-vitastor.rules
-
-
-%pre -n vitastor-osd
-groupadd -r -f vitastor 2>/dev/null ||:
-useradd -r -g vitastor -s /sbin/nologin -c "Vitastor daemons" -M -d /nonexistent vitastor 2>/dev/null ||:
-install -o vitastor -g vitastor -d /var/log/vitastor
-mkdir -p /etc/vitastor
 
 
 %files -n vitastor-mon
 /usr/lib/vitastor/mon
-/lib/systemd/system/vitastor-mon.service
-
-
-%pre -n vitastor-mon
-groupadd -r -f vitastor 2>/dev/null ||:
-useradd -r -g vitastor -s /sbin/nologin -c "Vitastor daemons" -M -d /nonexistent vitastor 2>/dev/null ||:
-mkdir -p /etc/vitastor
 
 
 %files -n vitastor-client
@@ -150,6 +128,7 @@ mkdir -p /etc/vitastor
 %_bindir/vita
 %_libdir/libvitastor_blk.so*
 %_libdir/libvitastor_client.so*
+/usr/lib/vitastor/make-osd.sh
 
 
 %files -n vitastor-client-devel

rpm/vitastor-el8.Dockerfile

@@ -35,7 +35,7 @@ ADD . /root/vitastor
 RUN set -e; \
     cd /root/vitastor/rpm; \
     sh build-tarball.sh; \
-    cp /root/vitastor-0.8.3.el8.tar.gz ~/rpmbuild/SOURCES; \
+    cp /root/vitastor-0.7.1.el8.tar.gz ~/rpmbuild/SOURCES; \
     cp vitastor-el8.spec ~/rpmbuild/SPECS/vitastor.spec; \
     cd ~/rpmbuild/SPECS/; \
     rpmbuild -ba vitastor.spec; \

rpm/vitastor-el8.spec

@@ -1,11 +1,11 @@
 Name:           vitastor
-Version:        0.8.3
+Version:        0.7.1
 Release:        1%{?dist}
 Summary:        Vitastor, a fast software-defined clustered block storage
 
 License:        Vitastor Network Public License 1.1
 URL:            https://vitastor.io/
-Source0:        vitastor-0.8.3.el8.tar.gz
+Source0:        vitastor-0.7.1.el8.tar.gz
 
 BuildRequires:  liburing-devel >= 0.6
 BuildRequires:  gperftools-devel
@@ -35,8 +35,6 @@ Requires:       libJerasure2
 Requires:       libisa-l
 Requires:       liburing >= 0.6
 Requires:       vitastor-client = %{version}-%{release}
-Requires:       util-linux
-Requires:       parted
 
 
 %description -n vitastor-osd
@@ -101,11 +99,8 @@ cd mon
 npm install
 cd ..
 mkdir -p %buildroot/usr/lib/vitastor
+cp mon/make-osd.sh %buildroot/usr/lib/vitastor
 cp -r mon %buildroot/usr/lib/vitastor
-mkdir -p %buildroot/lib/systemd/system
-cp mon/vitastor.target mon/vitastor-mon.service mon/vitastor-osd@.service %buildroot/lib/systemd/system
-mkdir -p %buildroot/lib/udev/rules.d
-cp mon/90-vitastor.rules %buildroot/lib/udev/rules.d
 
 
 %files
@@ -116,27 +111,10 @@ cp mon/90-vitastor.rules %buildroot/lib/udev/rules.d
 %_bindir/vitastor-osd
 %_bindir/vitastor-disk
 %_bindir/vitastor-dump-journal
-/lib/systemd/system/vitastor-osd@.service
-/lib/systemd/system/vitastor.target
-/lib/udev/rules.d/90-vitastor.rules
-
-
-%pre -n vitastor-osd
-groupadd -r -f vitastor 2>/dev/null ||:
-useradd -r -g vitastor -s /sbin/nologin -c "Vitastor daemons" -M -d /nonexistent vitastor 2>/dev/null ||:
-install -o vitastor -g vitastor -d /var/log/vitastor
-mkdir -p /etc/vitastor
 
 
 %files -n vitastor-mon
 /usr/lib/vitastor/mon
-/lib/systemd/system/vitastor-mon.service
-
-
-%pre -n vitastor-mon
-groupadd -r -f vitastor 2>/dev/null ||:
-useradd -r -g vitastor -s /sbin/nologin -c "Vitastor daemons" -M -d /nonexistent vitastor 2>/dev/null ||:
-mkdir -p /etc/vitastor
 
 
 %files -n vitastor-client
@@ -147,6 +125,7 @@ mkdir -p /etc/vitastor
 %_bindir/vita
 %_libdir/libvitastor_blk.so*
 %_libdir/libvitastor_client.so*
+/usr/lib/vitastor/make-osd.sh
 
 
 %files -n vitastor-client-devel

src/CMakeLists.txt

@@ -15,7 +15,7 @@ if("${CMAKE_INSTALL_PREFIX}" MATCHES "^/usr/local/?$")
 	set(CMAKE_INSTALL_RPATH "${CMAKE_INSTALL_PREFIX}/${CMAKE_INSTALL_LIBDIR}")
 endif()
 
-add_definitions(-DVERSION="0.8.3")
+add_definitions(-DVERSION="0.7.1")
 add_definitions(-Wall -Wno-sign-compare -Wno-comment -Wno-parentheses -Wno-pointer-arith -fdiagnostics-color=always -I ${CMAKE_SOURCE_DIR}/src)
 if (${WITH_ASAN})
 	add_definitions(-fsanitize=address -fno-omit-frame-pointer)
@@ -64,7 +64,7 @@ include_directories(
 
 # libvitastor_blk.so
 add_library(vitastor_blk SHARED
-	allocator.cpp blockstore.cpp blockstore_impl.cpp blockstore_disk.cpp blockstore_init.cpp blockstore_open.cpp blockstore_journal.cpp blockstore_read.cpp
+	allocator.cpp blockstore.cpp blockstore_impl.cpp blockstore_init.cpp blockstore_open.cpp blockstore_journal.cpp blockstore_read.cpp
 	blockstore_write.cpp blockstore_sync.cpp blockstore_stable.cpp blockstore_rollback.cpp blockstore_flush.cpp crc32c.c ringloop.cpp
 )
 target_link_libraries(vitastor_blk
@@ -94,7 +94,7 @@ endif (IBVERBS_LIBRARIES)
 add_library(vitastor_common STATIC
 	epoll_manager.cpp etcd_state_client.cpp messenger.cpp addr_util.cpp
 	msgr_stop.cpp msgr_op.cpp msgr_send.cpp msgr_receive.cpp ringloop.cpp ../json11/json11.cpp
-	http_client.cpp osd_ops.cpp pg_states.cpp timerfd_manager.cpp str_util.cpp ${MSGR_RDMA}
+	http_client.cpp osd_ops.cpp pg_states.cpp timerfd_manager.cpp base64.cpp ${MSGR_RDMA}
 )
 target_compile_options(vitastor_common PUBLIC -fPIC)
 
@@ -131,6 +131,7 @@ add_library(vitastor_client SHARED
 	vitastor_c.cpp
 	cli_common.cpp
 	cli_alloc_osd.cpp
+	cli_simple_offsets.cpp
 	cli_status.cpp
 	cli_df.cpp
 	cli_ls.cpp
@@ -140,7 +141,6 @@ add_library(vitastor_client SHARED
 	cli_merge.cpp
 	cli_rm_data.cpp
 	cli_rm.cpp
-	cli_rm_osd.cpp
 )
 set_target_properties(vitastor_client PROPERTIES PUBLIC_HEADER "vitastor_c.h")
 target_link_libraries(vitastor_client
@@ -195,9 +195,7 @@ configure_file(vitastor.pc.in vitastor.pc @ONLY)
 
 # vitastor-disk
 add_executable(vitastor-disk
-	disk_tool.cpp disk_simple_offsets.cpp
-	disk_tool_journal.cpp disk_tool_meta.cpp disk_tool_prepare.cpp disk_tool_resize.cpp disk_tool_udev.cpp disk_tool_utils.cpp disk_tool_upgrade.cpp
-	crc32c.c str_util.cpp ../json11/json11.cpp rw_blocking.cpp allocator.cpp ringloop.cpp blockstore_disk.cpp
+	disk_tool.cpp crc32c.c rw_blocking.cpp allocator.cpp ringloop.cpp
 )
 target_link_libraries(vitastor-disk
 	tcmalloc_minimal
@@ -235,16 +233,9 @@ add_executable(osd_test osd_test.cpp rw_blocking.cpp addr_util.cpp)
 target_link_libraries(osd_test tcmalloc_minimal)
 
 # osd_rmw_test
-# FIXME: Move to tests
 add_executable(osd_rmw_test osd_rmw_test.cpp allocator.cpp)
 target_link_libraries(osd_rmw_test Jerasure ${ISAL_LIBRARIES} tcmalloc_minimal)
 
-if (ISAL_LIBRARIES)
-	add_executable(osd_rmw_test_je osd_rmw_test.cpp allocator.cpp)
-	target_compile_definitions(osd_rmw_test_je PUBLIC -DNO_ISAL)
-	target_link_libraries(osd_rmw_test_je Jerasure tcmalloc_minimal)
-endif (ISAL_LIBRARIES)
-
 # stub_uring_osd
 add_executable(stub_uring_osd
 	stub_uring_osd.cpp
@@ -271,14 +262,6 @@ target_link_libraries(test_cas
 	vitastor_client
 )
 
-# test_crc32
-add_executable(test_crc32
-	test_crc32.cpp
-)
-target_link_libraries(test_crc32
-	vitastor_blk
-)
-
 # test_cluster_client
 add_executable(test_cluster_client
 	test_cluster_client.cpp

src/base64.cpp

@@ -0,0 +1,55 @@
+// Copyright (c) Vitaliy Filippov, 2019+
+// License: VNPL-1.1 (see README.md for details)
+
+#include "base64.h"
+
+std::string base64_encode(const std::string &in)
+{
+    std::string out;
+    unsigned val = 0;
+    int valb = -6;
+    for (unsigned char c: in)
+    {
+        val = (val << 8) + c;
+        valb += 8;
+        while (valb >= 0)
+        {
+            out.push_back("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"[(val>>valb) & 0x3F]);
+            valb -= 6;
+        }
+    }
+    if (valb > -6)
+        out.push_back("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"[((val<<8)>>(valb+8)) & 0x3F]);
+    while (out.size() % 4)
+        out.push_back('=');
+    return out;
+}
+
+static char T[256] = { 0 };
+
+std::string base64_decode(const std::string &in)
+{
+    std::string out;
+    if (T[0] == 0)
+    {
+        for (int i = 0; i < 256; i++)
+            T[i] = -1;
+        for (int i = 0; i < 64; i++)
+            T[(unsigned char)("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"[i])] = i;
+    }
+    unsigned val = 0;
+    int valb = -8;
+    for (unsigned char c: in)
+    {
+        if (T[c] == -1)
+            break;
+        val = (val<<6) + T[c];
+        valb += 6;
+        if (valb >= 0)
+        {
+            out.push_back(char((val >> valb) & 0xFF));
+            valb -= 8;
+        }
+    }
+    return out;
+}

src/base64.h

@@ -0,0 +1,8 @@
+// Copyright (c) Vitaliy Filippov, 2019+
+// License: VNPL-1.1 (see README.md for details)
+
+#pragma once
+#include <string>
+
+std::string base64_encode(const std::string &in);
+std::string base64_decode(const std::string &in);

src/blockstore.h

@@ -11,6 +11,7 @@
 
 #include <string>
 #include <map>
+#include <unordered_map>
 #include <functional>
 
 #include "object_id.h"
@@ -154,7 +155,7 @@ struct blockstore_op_t
     uint8_t private_data[BS_OP_PRIVATE_DATA_SIZE];
 };
 
-typedef std::map<std::string, std::string> blockstore_config_t;
+typedef std::unordered_map<std::string, std::string> blockstore_config_t;
 
 class blockstore_impl_t;
 

src/blockstore_disk.cpp

@@ -1,323 +0,0 @@
-// Copyright (c) Vitaliy Filippov, 2019+
-// License: VNPL-1.1 (see README.md for details)
-
-#include <sys/file.h>
-
-#include <stdexcept>
-
-#include "blockstore_impl.h"
-#include "blockstore_disk.h"
-#include "str_util.h"
-
-static uint32_t is_power_of_two(uint64_t value)
-{
-    uint32_t l = 0;
-    while (value > 1)
-    {
-        if (value & 1)
-        {
-            return 64;
-        }
-        value = value >> 1;
-        l++;
-    }
-    return l;
-}
-
-void blockstore_disk_t::parse_config(std::map<std::string, std::string> & config)
-{
-    // Parse
-    if (config["disable_device_lock"] == "true" || config["disable_device_lock"] == "1" || config["disable_device_lock"] == "yes")
-    {
-        disable_flock = true;
-    }
-    cfg_journal_size = parse_size(config["journal_size"]);
-    data_device = config["data_device"];
-    data_offset = parse_size(config["data_offset"]);
-    cfg_data_size = parse_size(config["data_size"]);
-    meta_device = config["meta_device"];
-    meta_offset = parse_size(config["meta_offset"]);
-    data_block_size = parse_size(config["block_size"]);
-    journal_device = config["journal_device"];
-    journal_offset = parse_size(config["journal_offset"]);
-    disk_alignment = strtoull(config["disk_alignment"].c_str(), NULL, 10);
-    journal_block_size = strtoull(config["journal_block_size"].c_str(), NULL, 10);
-    meta_block_size = strtoull(config["meta_block_size"].c_str(), NULL, 10);
-    bitmap_granularity = strtoull(config["bitmap_granularity"].c_str(), NULL, 10);
-    // Validate
-    if (!data_block_size)
-    {
-        data_block_size = (1 << DEFAULT_DATA_BLOCK_ORDER);
-    }
-    if ((block_order = is_power_of_two(data_block_size)) >= 64 || data_block_size < MIN_DATA_BLOCK_SIZE || data_block_size >= MAX_DATA_BLOCK_SIZE)
-    {
-        throw std::runtime_error("Bad block size");
-    }
-    if (!disk_alignment)
-    {
-        disk_alignment = 4096;
-    }
-    else if (disk_alignment % DIRECT_IO_ALIGNMENT)
-    {
-        throw std::runtime_error("disk_alignment must be a multiple of "+std::to_string(DIRECT_IO_ALIGNMENT));
-    }
-    if (!journal_block_size)
-    {
-        journal_block_size = 4096;
-    }
-    else if (journal_block_size % DIRECT_IO_ALIGNMENT)
-    {
-        throw std::runtime_error("journal_block_size must be a multiple of "+std::to_string(DIRECT_IO_ALIGNMENT));
-    }
-    if (!meta_block_size)
-    {
-        meta_block_size = 4096;
-    }
-    else if (meta_block_size % DIRECT_IO_ALIGNMENT)
-    {
-        throw std::runtime_error("meta_block_size must be a multiple of "+std::to_string(DIRECT_IO_ALIGNMENT));
-    }
-    if (data_offset % disk_alignment)
-    {
-        throw std::runtime_error("data_offset must be a multiple of disk_alignment = "+std::to_string(disk_alignment));
-    }
-    if (!bitmap_granularity)
-    {
-        bitmap_granularity = DEFAULT_BITMAP_GRANULARITY;
-    }
-    else if (bitmap_granularity % disk_alignment)
-    {
-        throw std::runtime_error("Sparse write tracking granularity must be a multiple of disk_alignment = "+std::to_string(disk_alignment));
-    }
-    if (data_block_size % bitmap_granularity)
-    {
-        throw std::runtime_error("Block size must be a multiple of sparse write tracking granularity");
-    }
-    if (meta_device == "")
-    {
-        meta_device = data_device;
-    }
-    if (journal_device == "")
-    {
-        journal_device = meta_device;
-    }
-    if (meta_offset % meta_block_size)
-    {
-        throw std::runtime_error("meta_offset must be a multiple of meta_block_size = "+std::to_string(meta_block_size));
-    }
-    if (journal_offset % journal_block_size)
-    {
-        throw std::runtime_error("journal_offset must be a multiple of journal_block_size = "+std::to_string(journal_block_size));
-    }
-    clean_entry_bitmap_size = data_block_size / bitmap_granularity / 8;
-    clean_entry_size = sizeof(clean_disk_entry) + 2*clean_entry_bitmap_size;
-}
-
-void blockstore_disk_t::calc_lengths(bool skip_meta_check)
-{
-    // data
-    data_len = data_device_size - data_offset;
-    if (data_fd == meta_fd && data_offset < meta_offset)
-    {
-        data_len = meta_offset - data_offset;
-    }
-    if (data_fd == journal_fd && data_offset < journal_offset)
-    {
-        data_len = data_len < journal_offset-data_offset
-            ? data_len : journal_offset-data_offset;
-    }
-    if (cfg_data_size != 0)
-    {
-        if (data_len < cfg_data_size)
-        {
-            throw std::runtime_error("Data area ("+std::to_string(data_len)+
-                " bytes) is smaller than configured size ("+std::to_string(cfg_data_size)+" bytes)");
-        }
-        data_len = cfg_data_size;
-    }
-    // meta
-    uint64_t meta_area_size = (meta_fd == data_fd ? data_device_size : meta_device_size) - meta_offset;
-    if (meta_fd == data_fd && meta_offset <= data_offset)
-    {
-        meta_area_size = data_offset - meta_offset;
-    }
-    if (meta_fd == journal_fd && meta_offset <= journal_offset)
-    {
-        meta_area_size = meta_area_size < journal_offset-meta_offset
-            ? meta_area_size : journal_offset-meta_offset;
-    }
-    // journal
-    journal_len = (journal_fd == data_fd ? data_device_size : (journal_fd == meta_fd ? meta_device_size : journal_device_size)) - journal_offset;
-    if (journal_fd == data_fd && journal_offset <= data_offset)
-    {
-        journal_len = data_offset - journal_offset;
-    }
-    if (journal_fd == meta_fd && journal_offset <= meta_offset)
-    {
-        journal_len = journal_len < meta_offset-journal_offset
-            ? journal_len : meta_offset-journal_offset;
-    }
-    // required metadata size
-    block_count = data_len / data_block_size;
-    meta_len = (1 + (block_count - 1 + meta_block_size / clean_entry_size) / (meta_block_size / clean_entry_size)) * meta_block_size;
-    if (!skip_meta_check && meta_area_size < meta_len)
-    {
-        throw std::runtime_error("Metadata area is too small, need at least "+std::to_string(meta_len)+" bytes");
-    }
-    // requested journal size
-    if (!skip_meta_check && cfg_journal_size > journal_len)
-    {
-        throw std::runtime_error("Requested journal_size is too large");
-    }
-    else if (cfg_journal_size > 0)
-    {
-        journal_len = cfg_journal_size;
-    }
-    if (journal_len < MIN_JOURNAL_SIZE)
-    {
-        throw std::runtime_error("Journal is too small, need at least "+std::to_string(MIN_JOURNAL_SIZE)+" bytes");
-    }
-}
-
-// FIXME: Move to utils
-static void check_size(int fd, uint64_t *size, uint64_t *sectsize, std::string name)
-{
-    int sect;
-    struct stat st;
-    if (fstat(fd, &st) < 0)
-    {
-        throw std::runtime_error("Failed to stat "+name);
-    }
-    if (S_ISREG(st.st_mode))
-    {
-        *size = st.st_size;
-        if (sectsize)
-        {
-            *sectsize = st.st_blksize;
-        }
-    }
-    else if (S_ISBLK(st.st_mode))
-    {
-        if (ioctl(fd, BLKGETSIZE64, size) < 0 ||
-            ioctl(fd, BLKSSZGET, &sect) < 0)
-        {
-            throw std::runtime_error("Failed to get "+name+" size or block size: "+strerror(errno));
-        }
-        if (sectsize)
-        {
-            *sectsize = sect;
-        }
-    }
-    else
-    {
-        throw std::runtime_error(name+" is neither a file nor a block device");
-    }
-}
-
-void blockstore_disk_t::open_data()
-{
-    data_fd = open(data_device.c_str(), O_DIRECT|O_RDWR);
-    if (data_fd == -1)
-    {
-        throw std::runtime_error("Failed to open data device "+data_device+": "+std::string(strerror(errno)));
-    }
-    check_size(data_fd, &data_device_size, &data_device_sect, "data device");
-    if (disk_alignment % data_device_sect)
-    {
-        throw std::runtime_error(
-            "disk_alignment ("+std::to_string(disk_alignment)+
-            ") is not a multiple of data device sector size ("+std::to_string(data_device_sect)+")"
-        );
-    }
-    if (data_offset >= data_device_size)
-    {
-        throw std::runtime_error("data_offset exceeds device size = "+std::to_string(data_device_size));
-    }
-    if (!disable_flock && flock(data_fd, LOCK_EX|LOCK_NB) != 0)
-    {
-        throw std::runtime_error(std::string("Failed to lock data device: ") + strerror(errno));
-    }
-}
-
-void blockstore_disk_t::open_meta()
-{
-    if (meta_device != data_device)
-    {
-        meta_fd = open(meta_device.c_str(), O_DIRECT|O_RDWR);
-        if (meta_fd == -1)
-        {
-            throw std::runtime_error("Failed to open metadata device "+meta_device+": "+std::string(strerror(errno)));
-        }
-        check_size(meta_fd, &meta_device_size, &meta_device_sect, "metadata device");
-        if (meta_offset >= meta_device_size)
-        {
-            throw std::runtime_error("meta_offset exceeds device size = "+std::to_string(meta_device_size));
-        }
-        if (!disable_flock && flock(meta_fd, LOCK_EX|LOCK_NB) != 0)
-        {
-            throw std::runtime_error(std::string("Failed to lock metadata device: ") + strerror(errno));
-        }
-    }
-    else
-    {
-        meta_fd = data_fd;
-        meta_device_sect = data_device_sect;
-        meta_device_size = 0;
-        if (meta_offset >= data_device_size)
-        {
-            throw std::runtime_error("meta_offset exceeds device size = "+std::to_string(data_device_size));
-        }
-    }
-    if (meta_block_size % meta_device_sect)
-    {
-        throw std::runtime_error(
-            "meta_block_size ("+std::to_string(meta_block_size)+
-            ") is not a multiple of data device sector size ("+std::to_string(meta_device_sect)+")"
-        );
-    }
-}
-
-void blockstore_disk_t::open_journal()
-{
-    if (journal_device != meta_device)
-    {
-        journal_fd = open(journal_device.c_str(), O_DIRECT|O_RDWR);
-        if (journal_fd == -1)
-        {
-            throw std::runtime_error("Failed to open journal device "+journal_device+": "+std::string(strerror(errno)));
-        }
-        check_size(journal_fd, &journal_device_size, &journal_device_sect, "journal device");
-        if (!disable_flock && flock(journal_fd, LOCK_EX|LOCK_NB) != 0)
-        {
-            throw std::runtime_error(std::string("Failed to lock journal device: ") + strerror(errno));
-        }
-    }
-    else
-    {
-        journal_fd = meta_fd;
-        journal_device_sect = meta_device_sect;
-        journal_device_size = 0;
-        if (journal_offset >= data_device_size)
-        {
-            throw std::runtime_error("journal_offset exceeds device size");
-        }
-    }
-    if (journal_block_size % journal_device_sect)
-    {
-        throw std::runtime_error(
-            "journal_block_size ("+std::to_string(journal_block_size)+
-            ") is not a multiple of journal device sector size ("+std::to_string(journal_device_sect)+")"
-        );
-    }
-}
-
-void blockstore_disk_t::close_all()
-{
-    if (data_fd >= 0)
-        close(data_fd);
-    if (meta_fd >= 0 && meta_fd != data_fd)
-        close(meta_fd);
-    if (journal_fd >= 0 && journal_fd != meta_fd)
-        close(journal_fd);
-    data_fd = meta_fd = journal_fd = -1;
-}

src/blockstore_disk.h

@@ -1,42 +0,0 @@
-// Copyright (c) Vitaliy Filippov, 2019+
-// License: VNPL-1.1 (see README.md for details)
-
-#pragma once
-
-#include <stdint.h>
-
-#include <string>
-#include <map>
-
-struct blockstore_disk_t
-{
-    std::string data_device, meta_device, journal_device;
-    uint32_t data_block_size;
-    uint64_t cfg_journal_size, cfg_data_size;
-    // Required write alignment and journal/metadata/data areas' location alignment
-    uint32_t disk_alignment = 4096;
-    // Journal block size - minimum_io_size of the journal device is the best choice
-    uint64_t journal_block_size = 4096;
-    // Metadata block size - minimum_io_size of the metadata device is the best choice
-    uint64_t meta_block_size = 4096;
-    // Sparse write tracking granularity. 4 KB is a good choice. Must be a multiple of disk_alignment
-    uint64_t bitmap_granularity = 4096;
-    // By default, Blockstore locks all opened devices exclusively. This option can be used to disable locking
-    bool disable_flock = false;
-
-    int meta_fd = -1, data_fd = -1, journal_fd = -1;
-    uint64_t meta_offset, meta_device_sect, meta_device_size, meta_len;
-    uint64_t data_offset, data_device_sect, data_device_size, data_len;
-    uint64_t journal_offset, journal_device_sect, journal_device_size, journal_len;
-
-    uint32_t block_order;
-    uint64_t block_count;
-    uint32_t clean_entry_bitmap_size = 0, clean_entry_size = 0;
-
-    void parse_config(std::map<std::string, std::string> & config);
-    void open_data();
-    void open_meta();
-    void open_journal();
-    void calc_lengths(bool skip_meta_check = false);
-    void close_all();
-};

src/blockstore_flush.cpp

@@ -15,11 +15,11 @@ journal_flusher_t::journal_flusher_t(blockstore_impl_t *bs)
     active_flushers = 0;
     syncing_flushers = 0;
     // FIXME: allow to configure flusher_start_threshold and journal_trim_interval
-    flusher_start_threshold = bs->dsk.journal_block_size / sizeof(journal_entry_stable);
+    flusher_start_threshold = bs->journal_block_size / sizeof(journal_entry_stable);
     journal_trim_interval = 512;
     journal_trim_counter = bs->journal.flush_journal ? 1 : 0;
     trim_wanted = bs->journal.flush_journal ? 1 : 0;
-    journal_superblock = bs->journal.inmemory ? bs->journal.buffer : memalign_or_die(MEM_ALIGNMENT, bs->dsk.journal_block_size);
+    journal_superblock = bs->journal.inmemory ? bs->journal.buffer : memalign_or_die(MEM_ALIGNMENT, bs->journal_block_size);
     co = new journal_flusher_co[max_flusher_count];
     for (int i = 0; i < max_flusher_count; i++)
     {
@@ -35,14 +35,24 @@ journal_flusher_co::journal_flusher_co()
     {
         bs->live = true;
         if (data->res != data->iov.iov_len)
-            bs->disk_error_abort("read operation during flush", data->res, data->iov.iov_len);
+        {
+            throw std::runtime_error(
+                "data read operation failed during flush ("+std::to_string(data->res)+" != "+std::to_string(data->iov.iov_len)+
+                "). can't continue, sorry :-("
+            );
+        }
         wait_count--;
     };
     simple_callback_w = [this](ring_data_t* data)
     {
         bs->live = true;
         if (data->res != data->iov.iov_len)
-            bs->disk_error_abort("write operation during flush", data->res, data->iov.iov_len);
+        {
+            throw std::runtime_error(
+                "write operation failed ("+std::to_string(data->res)+" != "+std::to_string(data->iov.iov_len)+
+                "). state "+std::to_string(wait_state)+". in-memory state is corrupted. AAAAAAAaaaaaaaaa!!!111"
+            );
+        }
         wait_count--;
     };
 }
@@ -77,7 +87,7 @@ void journal_flusher_t::loop()
             cur_flusher_count--;
         }
     }
-    for (int i = 0; (active_flushers > 0 || dequeuing || trim_wanted > 0) && i < cur_flusher_count; i++)
+    for (int i = 0; (active_flushers > 0 || dequeuing) && i < cur_flusher_count; i++)
         co[i].loop();
 }
 
@@ -162,8 +172,7 @@ void journal_flusher_t::mark_trim_possible()
     if (trim_wanted > 0)
     {
         dequeuing = true;
-        if (!journal_trim_counter)
-            journal_trim_counter = journal_trim_interval;
+        journal_trim_counter++;
         bs->ringloop->wakeup();
     }
 }
@@ -297,8 +306,6 @@ bool journal_flusher_co::loop()
         goto resume_20;
     else if (wait_state == 21)
         goto resume_21;
-    else if (wait_state == 22)
-        goto resume_22;
 resume_0:
     if (flusher->flush_queue.size() < flusher->min_flusher_count && !flusher->trim_wanted ||
         !flusher->flush_queue.size() || !flusher->dequeuing)
@@ -479,38 +486,31 @@ resume_1:
             bs->ringloop->wakeup();
         }
         // Reads completed, submit writes and set bitmap bits
-        if (bs->dsk.clean_entry_bitmap_size)
+        if (bs->clean_entry_bitmap_size)
         {
             new_clean_bitmap = (bs->inmemory_meta
-                ? (uint8_t*)meta_new.buf + meta_new.pos*bs->dsk.clean_entry_size + sizeof(clean_disk_entry)
-                : (uint8_t*)bs->clean_bitmap + (clean_loc >> bs->dsk.block_order)*(2*bs->dsk.clean_entry_bitmap_size));
+                ? (uint8_t*)meta_new.buf + meta_new.pos*bs->clean_entry_size + sizeof(clean_disk_entry)
+                : (uint8_t*)bs->clean_bitmap + (clean_loc >> bs->block_order)*(2*bs->clean_entry_bitmap_size));
             if (clean_init_bitmap)
             {
-                memset(new_clean_bitmap, 0, bs->dsk.clean_entry_bitmap_size);
-                bitmap_set(new_clean_bitmap, clean_bitmap_offset, clean_bitmap_len, bs->dsk.bitmap_granularity);
+                memset(new_clean_bitmap, 0, bs->clean_entry_bitmap_size);
+                bitmap_set(new_clean_bitmap, clean_bitmap_offset, clean_bitmap_len, bs->bitmap_granularity);
             }
         }
         for (it = v.begin(); it != v.end(); it++)
         {
             if (new_clean_bitmap)
             {
-                bitmap_set(new_clean_bitmap, it->offset, it->len, bs->dsk.bitmap_granularity);
+                bitmap_set(new_clean_bitmap, it->offset, it->len, bs->bitmap_granularity);
             }
             await_sqe(4);
             data->iov = (struct iovec){ it->buf, (size_t)it->len };
             data->callback = simple_callback_w;
             my_uring_prep_writev(
-                sqe, bs->dsk.data_fd, &data->iov, 1, bs->dsk.data_offset + clean_loc + it->offset
+                sqe, bs->data_fd, &data->iov, 1, bs->data_offset + clean_loc + it->offset
             );
             wait_count++;
         }
-        // Wait for data writes before fsyncing it
-    resume_22:
-        if (wait_count > 0)
-        {
-            wait_state = 22;
-            return false;
-        }
         // Sync data before writing metadata
     resume_16:
     resume_17:
@@ -521,7 +521,7 @@ resume_1:
             return false;
         }
     resume_5:
-        // Submit metadata writes, but only when data is written and fsynced
+        // And metadata writes, but only after data writes complete
         if (!bs->inmemory_meta && meta_new.it->second.state == 0 || wait_count > 0)
         {
             // metadata sector is still being read or data is still being written, wait for it
@@ -536,35 +536,35 @@ resume_1:
                 return false;
             }
             // zero out old metadata entry
-            memset((uint8_t*)meta_old.buf + meta_old.pos*bs->dsk.clean_entry_size, 0, bs->dsk.clean_entry_size);
+            memset((uint8_t*)meta_old.buf + meta_old.pos*bs->clean_entry_size, 0, bs->clean_entry_size);
             await_sqe(15);
-            data->iov = (struct iovec){ meta_old.buf, bs->dsk.meta_block_size };
+            data->iov = (struct iovec){ meta_old.buf, bs->meta_block_size };
             data->callback = simple_callback_w;
             my_uring_prep_writev(
-                sqe, bs->dsk.meta_fd, &data->iov, 1, bs->dsk.meta_offset + bs->dsk.meta_block_size + meta_old.sector
+                sqe, bs->meta_fd, &data->iov, 1, bs->meta_offset + meta_old.sector
             );
             wait_count++;
         }
         if (has_delete)
         {
-            clean_disk_entry *new_entry = (clean_disk_entry*)((uint8_t*)meta_new.buf + meta_new.pos*bs->dsk.clean_entry_size);
+            clean_disk_entry *new_entry = (clean_disk_entry*)((uint8_t*)meta_new.buf + meta_new.pos*bs->clean_entry_size);
             if (new_entry->oid.inode != 0 && new_entry->oid != cur.oid)
             {
                 printf("Fatal error (metadata corruption or bug): tried to delete metadata entry %lu (%lx:%lx v%lu) while deleting %lx:%lx\n",
-                    clean_loc >> bs->dsk.block_order, new_entry->oid.inode, new_entry->oid.stripe,
+                    clean_loc >> bs->block_order, new_entry->oid.inode, new_entry->oid.stripe,
                     new_entry->version, cur.oid.inode, cur.oid.stripe);
                 exit(1);
             }
             // zero out new metadata entry
-            memset((uint8_t*)meta_new.buf + meta_new.pos*bs->dsk.clean_entry_size, 0, bs->dsk.clean_entry_size);
+            memset((uint8_t*)meta_new.buf + meta_new.pos*bs->clean_entry_size, 0, bs->clean_entry_size);
         }
         else
         {
-            clean_disk_entry *new_entry = (clean_disk_entry*)((uint8_t*)meta_new.buf + meta_new.pos*bs->dsk.clean_entry_size);
+            clean_disk_entry *new_entry = (clean_disk_entry*)((uint8_t*)meta_new.buf + meta_new.pos*bs->clean_entry_size);
             if (new_entry->oid.inode != 0 && new_entry->oid != cur.oid)
             {
                 printf("Fatal error (metadata corruption or bug): tried to overwrite non-zero metadata entry %lu (%lx:%lx v%lu) with %lx:%lx v%lu\n",
-                    clean_loc >> bs->dsk.block_order, new_entry->oid.inode, new_entry->oid.stripe, new_entry->version,
+                    clean_loc >> bs->block_order, new_entry->oid.inode, new_entry->oid.stripe, new_entry->version,
                     cur.oid.inode, cur.oid.stripe, cur.version);
                 exit(1);
             }
@@ -572,20 +572,20 @@ resume_1:
             new_entry->version = cur.version;
             if (!bs->inmemory_meta)
             {
-                memcpy(&new_entry->bitmap, new_clean_bitmap, bs->dsk.clean_entry_bitmap_size);
+                memcpy(&new_entry->bitmap, new_clean_bitmap, bs->clean_entry_bitmap_size);
             }
             // copy latest external bitmap/attributes
-            if (bs->dsk.clean_entry_bitmap_size)
+            if (bs->clean_entry_bitmap_size)
             {
-                void *bmp_ptr = bs->dsk.clean_entry_bitmap_size > sizeof(void*) ? dirty_end->second.bitmap : &dirty_end->second.bitmap;
-                memcpy((uint8_t*)(new_entry+1) + bs->dsk.clean_entry_bitmap_size, bmp_ptr, bs->dsk.clean_entry_bitmap_size);
+                void *bmp_ptr = bs->clean_entry_bitmap_size > sizeof(void*) ? dirty_end->second.bitmap : &dirty_end->second.bitmap;
+                memcpy((uint8_t*)(new_entry+1) + bs->clean_entry_bitmap_size, bmp_ptr, bs->clean_entry_bitmap_size);
             }
         }
         await_sqe(6);
-        data->iov = (struct iovec){ meta_new.buf, bs->dsk.meta_block_size };
+        data->iov = (struct iovec){ meta_new.buf, bs->meta_block_size };
         data->callback = simple_callback_w;
         my_uring_prep_writev(
-            sqe, bs->dsk.meta_fd, &data->iov, 1, bs->dsk.meta_offset + bs->dsk.meta_block_size + meta_new.sector
+            sqe, bs->meta_fd, &data->iov, 1, bs->meta_offset + meta_new.sector
         );
         wait_count++;
     resume_7:
@@ -615,12 +615,7 @@ resume_1:
         }
         for (it = v.begin(); it != v.end(); it++)
         {
-            // Free it if it's not taken from the journal
-            if (it->buf && (!bs->journal.inmemory || it->buf < bs->journal.buffer ||
-                it->buf >= (uint8_t*)bs->journal.buffer + bs->journal.len))
-            {
-                free(it->buf);
-            }
+            free(it->buf);
         }
         v.clear();
         // And sync metadata (in batches - not per each operation!)
@@ -674,9 +669,9 @@ resume_1:
                     .version = JOURNAL_VERSION,
                 };
                 ((journal_entry_start*)flusher->journal_superblock)->crc32 = je_crc32((journal_entry*)flusher->journal_superblock);
-                data->iov = (struct iovec){ flusher->journal_superblock, bs->dsk.journal_block_size };
+                data->iov = (struct iovec){ flusher->journal_superblock, bs->journal_block_size };
                 data->callback = simple_callback_w;
-                my_uring_prep_writev(sqe, bs->dsk.journal_fd, &data->iov, 1, bs->journal.offset);
+                my_uring_prep_writev(sqe, bs->journal.fd, &data->iov, 1, bs->journal.offset);
                 wait_count++;
             resume_13:
                 if (wait_count > 0)
@@ -687,7 +682,7 @@ resume_1:
                 if (!bs->disable_journal_fsync)
                 {
                     await_sqe(20);
-                    my_uring_prep_fsync(sqe, bs->dsk.journal_fd, IORING_FSYNC_DATASYNC);
+                    my_uring_prep_fsync(sqe, bs->journal.fd, IORING_FSYNC_DATASYNC);
                     data->iov = { 0 };
                     data->callback = simple_callback_w;
                 resume_21:
@@ -765,22 +760,21 @@ bool journal_flusher_co::scan_dirty(int wait_base)
                     {
                         submit_offset = dirty_it->second.location + offset - dirty_it->second.offset;
                         submit_len = it == v.end() || it->offset >= end_offset ? end_offset-offset : it->offset-offset;
-                        it = v.insert(it, (copy_buffer_t){ .offset = offset, .len = submit_len });
+                        it = v.insert(it, (copy_buffer_t){ .offset = offset, .len = submit_len, .buf = memalign_or_die(MEM_ALIGNMENT, submit_len) });
                         copy_count++;
                         if (bs->journal.inmemory)
                         {
-                            // Take it from memory, don't copy it
-                            it->buf = (uint8_t*)bs->journal.buffer + submit_offset;
+                            // Take it from memory
+                            memcpy(it->buf, (uint8_t*)bs->journal.buffer + submit_offset, submit_len);
                         }
                         else
                         {
                             // Read it from disk
-                            it->buf = memalign_or_die(MEM_ALIGNMENT, submit_len);
                             await_sqe(0);
                             data->iov = (struct iovec){ it->buf, (size_t)submit_len };
                             data->callback = simple_callback_r;
                             my_uring_prep_readv(
-                                sqe, bs->dsk.journal_fd, &data->iov, 1, bs->journal.offset + submit_offset
+                                sqe, bs->journal.fd, &data->iov, 1, bs->journal.offset + submit_offset
                             );
                             wait_count++;
                         }
@@ -831,8 +825,8 @@ bool journal_flusher_co::modify_meta_read(uint64_t meta_loc, flusher_meta_write_
     // And yet another option is to use LSM trees for metadata, but it sophisticates everything a lot,
     // so I'll avoid it as long as I can.
     wr.submitted = false;
-    wr.sector = ((meta_loc >> bs->dsk.block_order) / (bs->dsk.meta_block_size / bs->dsk.clean_entry_size)) * bs->dsk.meta_block_size;
-    wr.pos = ((meta_loc >> bs->dsk.block_order) % (bs->dsk.meta_block_size / bs->dsk.clean_entry_size));
+    wr.sector = ((meta_loc >> bs->block_order) / (bs->meta_block_size / bs->clean_entry_size)) * bs->meta_block_size;
+    wr.pos = ((meta_loc >> bs->block_order) % (bs->meta_block_size / bs->clean_entry_size));
     if (bs->inmemory_meta)
     {
         wr.buf = (uint8_t*)bs->metadata_buffer + wr.sector;
@@ -842,20 +836,20 @@ bool journal_flusher_co::modify_meta_read(uint64_t meta_loc, flusher_meta_write_
     if (wr.it == flusher->meta_sectors.end())
     {
         // Not in memory yet, read it
-        wr.buf = memalign_or_die(MEM_ALIGNMENT, bs->dsk.meta_block_size);
+        wr.buf = memalign_or_die(MEM_ALIGNMENT, bs->meta_block_size);
         wr.it = flusher->meta_sectors.emplace(wr.sector, (meta_sector_t){
             .offset = wr.sector,
-            .len = bs->dsk.meta_block_size,
+            .len = bs->meta_block_size,
             .state = 0, // 0 = not read yet
             .buf = wr.buf,
             .usage_count = 1,
         }).first;
         await_sqe(0);
-        data->iov = (struct iovec){ wr.it->second.buf, bs->dsk.meta_block_size };
+        data->iov = (struct iovec){ wr.it->second.buf, bs->meta_block_size };
         data->callback = simple_callback_r;
         wr.submitted = true;
         my_uring_prep_readv(
-            sqe, bs->dsk.meta_fd, &data->iov, 1, bs->dsk.meta_offset + bs->dsk.meta_block_size + wr.sector
+            sqe, bs->meta_fd, &data->iov, 1, bs->meta_offset + wr.sector
         );
         wait_count++;
     }
@@ -873,11 +867,11 @@ void journal_flusher_co::update_clean_db()
     {
 #ifdef BLOCKSTORE_DEBUG
         printf("Free block %lu from %lx:%lx v%lu (new location is %lu)\n",
-            old_clean_loc >> bs->dsk.block_order,
+            old_clean_loc >> bs->block_order,
             cur.oid.inode, cur.oid.stripe, cur.version,
-            clean_loc >> bs->dsk.block_order);
+            clean_loc >> bs->block_order);
 #endif
-        bs->data_alloc->set(old_clean_loc >> bs->dsk.block_order, false);
+        bs->data_alloc->set(old_clean_loc >> bs->block_order, false);
     }
     auto & clean_db = bs->clean_db_shard(cur.oid);
     if (has_delete)
@@ -886,10 +880,10 @@ void journal_flusher_co::update_clean_db()
         clean_db.erase(clean_it);
 #ifdef BLOCKSTORE_DEBUG
         printf("Free block %lu from %lx:%lx v%lu (delete)\n",
-            clean_loc >> bs->dsk.block_order,
+            clean_loc >> bs->block_order,
             cur.oid.inode, cur.oid.stripe, cur.version);
 #endif
-        bs->data_alloc->set(clean_loc >> bs->dsk.block_order, false);
+        bs->data_alloc->set(clean_loc >> bs->block_order, false);
         clean_loc = UINT64_MAX;
     }
     else
@@ -938,7 +932,7 @@ bool journal_flusher_co::fsync_batch(bool fsync_meta, int wait_base)
                 await_sqe(0);
                 data->iov = { 0 };
                 data->callback = simple_callback_w;
-                my_uring_prep_fsync(sqe, fsync_meta ? bs->dsk.meta_fd : bs->dsk.data_fd, IORING_FSYNC_DATASYNC);
+                my_uring_prep_fsync(sqe, fsync_meta ? bs->meta_fd : bs->data_fd, IORING_FSYNC_DATASYNC);
                 cur_sync->state = 1;
                 wait_count++;
             resume_2:

src/blockstore_impl.cpp

@@ -11,19 +11,25 @@ blockstore_impl_t::blockstore_impl_t(blockstore_config_t & config, ring_loop_t *
     ring_consumer.loop = [this]() { loop(); };
     ringloop->register_consumer(&ring_consumer);
     initialized = 0;
+    data_fd = meta_fd = journal.fd = -1;
     parse_config(config);
-    zero_object = (uint8_t*)memalign_or_die(MEM_ALIGNMENT, dsk.data_block_size);
+    zero_object = (uint8_t*)memalign_or_die(MEM_ALIGNMENT, data_block_size);
     try
     {
-        dsk.open_data();
-        dsk.open_meta();
-        dsk.open_journal();
+        open_data();
+        open_meta();
+        open_journal();
         calc_lengths();
-        data_alloc = new allocator(dsk.block_count);
+        data_alloc = new allocator(block_count);
     }
     catch (std::exception & e)
     {
-        dsk.close_all();
+        if (data_fd >= 0)
+            close(data_fd);
+        if (meta_fd >= 0 && meta_fd != data_fd)
+            close(meta_fd);
+        if (journal.fd >= 0 && journal.fd != meta_fd)
+            close(journal.fd);
         throw;
     }
     flusher = new journal_flusher_t(this);
@@ -35,7 +41,12 @@ blockstore_impl_t::~blockstore_impl_t()
     delete flusher;
     free(zero_object);
     ringloop->unregister_consumer(&ring_consumer);
-    dsk.close_all();
+    if (data_fd >= 0)
+        close(data_fd);
+    if (meta_fd >= 0 && meta_fd != data_fd)
+        close(meta_fd);
+    if (journal.fd >= 0 && journal.fd != meta_fd)
+        close(journal.fd);
     if (metadata_buffer)
         free(metadata_buffer);
     if (clean_bitmap)
@@ -107,7 +118,7 @@ void blockstore_impl_t::loop()
         // has_writes == 0 - no writes before the current queue item
         // has_writes == 1 - some writes in progress
         // has_writes == 2 - tried to submit some writes, but failed
-        int has_writes = 0, op_idx = 0, new_idx = 0;
+        int has_writes = 0, op_idx = 0, new_idx = 0, done_lists = 0;
         for (; op_idx < submit_queue.size(); op_idx++, new_idx++)
         {
             auto op = submit_queue[op_idx];
@@ -188,12 +199,16 @@ void blockstore_impl_t::loop()
             else if (op->opcode == BS_OP_LIST)
             {
                 // LIST doesn't have to be blocked by previous modifications
-                process_list(op);
-                wr_st = 2;
+                // But don't do a lot of LISTs at once, because they're blocking and potentially slow
+                if (single_tick_list_limit <= 0 || done_lists < single_tick_list_limit)
+                {
+                    process_list(op);
+                    done_lists++;
+                    wr_st = 2;
+                }
             }
             if (wr_st == 2)
             {
-                submit_queue[op_idx] = NULL;
                 new_idx--;
             }
             if (wr_st == 0)
@@ -302,6 +317,17 @@ void blockstore_impl_t::check_wait(blockstore_op_t *op)
             // do not submit
 #ifdef BLOCKSTORE_DEBUG
             printf("Still waiting for a journal buffer\n");
+#endif
+            return;
+        }
+        PRIV(op)->wait_for = 0;
+    }
+    else if (PRIV(op)->wait_for == WAIT_FREE)
+    {
+        if (!data_alloc->get_free_count() && flusher->is_active())
+        {
+#ifdef BLOCKSTORE_DEBUG
+            printf("Still waiting for free space on the data device\n");
 #endif
             return;
         }
@@ -317,9 +343,9 @@ void blockstore_impl_t::enqueue_op(blockstore_op_t *op)
 {
     if (op->opcode < BS_OP_MIN || op->opcode > BS_OP_MAX ||
         ((op->opcode == BS_OP_READ || op->opcode == BS_OP_WRITE || op->opcode == BS_OP_WRITE_STABLE) && (
-            op->offset >= dsk.data_block_size ||
-            op->len > dsk.data_block_size-op->offset ||
-            (op->len % dsk.disk_alignment)
+            op->offset >= data_block_size ||
+            op->len > data_block_size-op->offset ||
+            (op->len % disk_alignment)
         )) ||
         readonly && op->opcode != BS_OP_READ && op->opcode != BS_OP_LIST)
     {
@@ -583,7 +609,7 @@ void blockstore_impl_t::process_list(blockstore_op_t *op)
                         replace_stable(dirty_it->first.oid, 0, clean_stable_count, stable_count, stable);
                     }
                 }
-                else if (IS_STABLE(dirty_it->second.state) || (dirty_it->second.state & BS_ST_INSTANT))
+                else if (IS_STABLE(dirty_it->second.state))
                 {
                     // First try to replace a clean stable version in the first part of the list
                     if (!replace_stable(dirty_it->first.oid, dirty_it->first.version, 0, clean_stable_count, stable))
@@ -672,16 +698,3 @@ void blockstore_impl_t::dump_diagnostics()
     journal.dump_diagnostics();
     flusher->dump_diagnostics();
 }
-
-void blockstore_impl_t::disk_error_abort(const char *op, int retval, int expected)
-{
-    if (retval == -EAGAIN)
-    {
-        fprintf(stderr, "EAGAIN error received from a disk %s during flush."
-            " It must never happen with io_uring and indicates a kernel bug."
-            " Please upgrade your kernel. Aborting.\n", op);
-        exit(1);
-    }
-    fprintf(stderr, "Disk %s failed: result is %d, expected %d. Can't continue, sorry :-(\n", op, retval, expected);
-    exit(1);
-}

src/blockstore_impl.h

@@ -4,7 +4,6 @@
 #pragma once
 
 #include "blockstore.h"
-#include "blockstore_disk.h"
 
 #include <sys/types.h>
 #include <sys/ioctl.h>
@@ -18,7 +17,6 @@
 #include <list>
 #include <deque>
 #include <new>
-#include <unordered_map>
 
 #include "cpp-btree/btree_map.h"
 
@@ -160,11 +158,12 @@ struct __attribute__((__packed__)) dirty_entry
 #define WAIT_JOURNAL 3
 // Suspend operation until the next journal sector buffer is free
 #define WAIT_JOURNAL_BUFFER 4
+// Suspend operation until there is some free space on the data device
+#define WAIT_FREE 5
 
 struct fulfill_read_t
 {
     uint64_t offset, len;
-    uint64_t journal_sector; // sector+1 if used and !journal.inmemory, otherwise 0
 };
 
 #define PRIV(op) ((blockstore_op_private_t*)(op)->private_data)
@@ -218,10 +217,23 @@ struct pool_shard_settings_t
 
 class blockstore_impl_t
 {
-    blockstore_disk_t dsk;
-
     /******* OPTIONS *******/
+    std::string data_device, meta_device, journal_device;
+    uint32_t data_block_size;
+    uint64_t meta_offset;
+    uint64_t data_offset;
+    uint64_t cfg_journal_size, cfg_data_size;
+    // Required write alignment and journal/metadata/data areas' location alignment
+    uint32_t disk_alignment = 4096;
+    // Journal block size - minimum_io_size of the journal device is the best choice
+    uint64_t journal_block_size = 4096;
+    // Metadata block size - minimum_io_size of the metadata device is the best choice
+    uint64_t meta_block_size = 4096;
+    // Sparse write tracking granularity. 4 KB is a good choice. Must be a multiple of disk_alignment
+    uint64_t bitmap_granularity = 4096;
     bool readonly = false;
+    // By default, Blockstore locks all opened devices exclusively. This option can be used to disable locking
+    bool disable_flock = false;
     // It is safe to disable fsync() if drive write cache is writethrough
     bool disable_data_fsync = false, disable_meta_fsync = false, disable_journal_fsync = false;
     // Enable if you want every operation to be executed with an "implicit fsync"
@@ -240,6 +252,8 @@ class blockstore_impl_t
     int throttle_target_parallelism = 1;
     // Minimum difference in microseconds between target and real execution times to throttle the response
     int throttle_threshold_us = 50;
+    // Maximum number of LIST operations to be processed between
+    int single_tick_list_limit = 1;
     /******* END OF OPTIONS *******/
 
     struct ring_consumer_t ring_consumer;
@@ -254,6 +268,16 @@ class blockstore_impl_t
     allocator *data_alloc = NULL;
     uint8_t *zero_object;
 
+    uint32_t block_order;
+    uint64_t block_count;
+    uint32_t clean_entry_bitmap_size = 0, clean_entry_size = 0;
+
+    int meta_fd;
+    int data_fd;
+    uint64_t meta_device_size, meta_len;
+    uint64_t data_device_size, data_len;
+    uint64_t data_device_sect, meta_device_sect, journal_device_sect;
+
     void *metadata_buffer = NULL;
 
     struct journal_t journal;
@@ -290,7 +314,6 @@ class blockstore_impl_t
     // Journaling
     void prepare_journal_sector_write(int sector, blockstore_op_t *op);
     void handle_journal_write(ring_data_t *data, uint64_t flush_id);
-    void disk_error_abort(const char *op, int retval, int expected);
 
     // Asynchronous init
     int initialized;
@@ -303,7 +326,7 @@ class blockstore_impl_t
     // Read
     int dequeue_read(blockstore_op_t *read_op);
     int fulfill_read(blockstore_op_t *read_op, uint64_t &fulfilled, uint32_t item_start, uint32_t item_end,
-        uint32_t item_state, uint64_t item_version, uint64_t item_location, uint64_t journal_sector);
+        uint32_t item_state, uint64_t item_version, uint64_t item_location);
     int fulfill_read_push(blockstore_op_t *op, void *buf, uint64_t offset, uint64_t len,
         uint32_t item_state, uint64_t item_version);
     void handle_read_event(ring_data_t *data, blockstore_op_t *op);
@@ -371,9 +394,9 @@ public:
     // Print diagnostics to stdout
     void dump_diagnostics();
 
-    inline uint32_t get_block_size() { return dsk.data_block_size; }
-    inline uint64_t get_block_count() { return dsk.block_count; }
+    inline uint32_t get_block_size() { return data_block_size; }
+    inline uint64_t get_block_count() { return block_count; }
     inline uint64_t get_free_block_count() { return data_alloc->get_free_count(); }
-    inline uint32_t get_bitmap_granularity() { return dsk.disk_alignment; }
-    inline uint64_t get_journal_size() { return dsk.journal_len; }
+    inline uint32_t get_bitmap_granularity() { return disk_alignment; }
+    inline uint64_t get_journal_size() { return journal.len; }
 };

src/blockstore_init.cpp

@@ -3,11 +3,6 @@
 
 #include "blockstore_impl.h"
 
-#define INIT_META_EMPTY 0
-#define INIT_META_READING 1
-#define INIT_META_READ_DONE 2
-#define INIT_META_WRITING 3
-
 #define GET_SQE() \
     sqe = bs->get_sqe();\
     if (!sqe)\
@@ -27,33 +22,32 @@ blockstore_init_meta::blockstore_init_meta(blockstore_impl_t *bs)
     this->bs = bs;
 }
 
-void blockstore_init_meta::handle_event(ring_data_t *data, int buf_num)
+void blockstore_init_meta::handle_event(ring_data_t *data)
 {
     if (data->res < 0)
     {
         throw std::runtime_error(
-            std::string("read metadata failed at offset ") + std::to_string(bufs[buf_num].offset) +
+            std::string("read metadata failed at offset ") + std::to_string(metadata_read) +
             std::string(": ") + strerror(-data->res)
         );
     }
-    if (buf_num >= 0)
-    {
-        bufs[buf_num].state = (bufs[buf_num].state == INIT_META_READING
-            ? INIT_META_READ_DONE
-            : INIT_META_EMPTY);
-    }
-    submitted--;
-    bs->ringloop->wakeup();
+    prev_done = data->res > 0 ? submitted : 0;
+    done_len = data->res;
+    done_pos = metadata_read;
+    metadata_read += data->res;
+    submitted = 0;
 }
 
 int blockstore_init_meta::loop()
 {
-    if (wait_state == 1)      goto resume_1;
-    else if (wait_state == 2) goto resume_2;
-    else if (wait_state == 3) goto resume_3;
-    else if (wait_state == 4) goto resume_4;
-    else if (wait_state == 5) goto resume_5;
-    else if (wait_state == 6) goto resume_6;
+    if (wait_state == 1)
+        goto resume_1;
+    else if (wait_state == 2)
+        goto resume_2;
+    else if (wait_state == 3)
+        goto resume_3;
+    else if (wait_state == 4)
+        goto resume_4;
     printf("Reading blockstore metadata\n");
     if (bs->inmemory_meta)
         metadata_buffer = bs->metadata_buffer;
@@ -63,27 +57,27 @@ int blockstore_init_meta::loop()
         throw std::runtime_error("Failed to allocate metadata read buffer");
     // Read superblock
     GET_SQE();
-    data->iov = { metadata_buffer, bs->dsk.meta_block_size };
-    data->callback = [this](ring_data_t *data) { handle_event(data, -1); };
-    my_uring_prep_readv(sqe, bs->dsk.meta_fd, &data->iov, 1, bs->dsk.meta_offset);
+    data->iov = { metadata_buffer, bs->meta_block_size };
+    data->callback = [this](ring_data_t *data) { handle_event(data); };
+    my_uring_prep_readv(sqe, bs->meta_fd, &data->iov, 1, bs->meta_offset);
     bs->ringloop->submit();
-    submitted++;
+    submitted = 1;
 resume_1:
-    if (submitted > 0)
+    if (submitted)
     {
         wait_state = 1;
         return 1;
     }
-    if (iszero((uint64_t*)metadata_buffer, bs->dsk.meta_block_size / sizeof(uint64_t)))
+    if (iszero((uint64_t*)metadata_buffer, bs->meta_block_size / sizeof(uint64_t)))
     {
         {
             blockstore_meta_header_v1_t *hdr = (blockstore_meta_header_v1_t *)metadata_buffer;
             hdr->zero = 0;
             hdr->magic = BLOCKSTORE_META_MAGIC_V1;
             hdr->version = BLOCKSTORE_META_VERSION_V1;
-            hdr->meta_block_size = bs->dsk.meta_block_size;
-            hdr->data_block_size = bs->dsk.data_block_size;
-            hdr->bitmap_granularity = bs->dsk.bitmap_granularity;
+            hdr->meta_block_size = bs->meta_block_size;
+            hdr->data_block_size = bs->data_block_size;
+            hdr->bitmap_granularity = bs->bitmap_granularity;
         }
         if (bs->readonly)
         {
@@ -93,11 +87,11 @@ resume_1:
         {
             printf("Initializing metadata area\n");
             GET_SQE();
-            data->iov = (struct iovec){ metadata_buffer, bs->dsk.meta_block_size };
-            data->callback = [this](ring_data_t *data) { handle_event(data, -1); };
-            my_uring_prep_writev(sqe, bs->dsk.meta_fd, &data->iov, 1, bs->dsk.meta_offset);
+            data->iov = (struct iovec){ metadata_buffer, bs->meta_block_size };
+            data->callback = [this](ring_data_t *data) { handle_event(data); };
+            my_uring_prep_writev(sqe, bs->meta_fd, &data->iov, 1, bs->meta_offset);
             bs->ringloop->submit();
-            submitted++;
+            submitted = 1;
         resume_3:
             if (submitted > 0)
             {
@@ -121,132 +115,80 @@ resume_1:
             );
             exit(1);
         }
-        if (hdr->meta_block_size != bs->dsk.meta_block_size ||
-            hdr->data_block_size != bs->dsk.data_block_size ||
-            hdr->bitmap_granularity != bs->dsk.bitmap_granularity)
+        if (hdr->meta_block_size != bs->meta_block_size ||
+            hdr->data_block_size != bs->data_block_size ||
+            hdr->bitmap_granularity != bs->bitmap_granularity)
         {
             printf(
                 "Configuration stored in metadata superblock"
                 " (meta_block_size=%u, data_block_size=%u, bitmap_granularity=%u)"
                 " differs from OSD configuration (%lu/%u/%lu).\n",
                 hdr->meta_block_size, hdr->data_block_size, hdr->bitmap_granularity,
-                bs->dsk.meta_block_size, bs->dsk.data_block_size, bs->dsk.bitmap_granularity
+                bs->meta_block_size, bs->data_block_size, bs->bitmap_granularity
             );
             exit(1);
         }
     }
     // Skip superblock
-    md_offset = bs->dsk.meta_block_size;
-    next_offset = md_offset;
-    entries_per_block = bs->dsk.meta_block_size / bs->dsk.clean_entry_size;
+    bs->meta_offset += bs->meta_block_size;
+    bs->meta_len -= bs->meta_block_size;
+    prev_done = 0;
+    done_len = 0;
+    done_pos = 0;
+    metadata_read = 0;
     // Read the rest of the metadata
-resume_2:
-    if (next_offset < bs->dsk.meta_len && submitted == 0)
+    while (1)
     {
-        // Submit one read
-        for (int i = 0; i < 2; i++)
+    resume_2:
+        if (submitted)
         {
-            if (!bufs[i].state)
-            {
-                bufs[i].buf = (uint8_t*)metadata_buffer + (bs->inmemory_meta
-                    ? next_offset-md_offset
-                    : i*bs->metadata_buf_size);
-                bufs[i].offset = next_offset;
-                bufs[i].size = bs->dsk.meta_len-next_offset > bs->metadata_buf_size
-                    ? bs->metadata_buf_size : bs->dsk.meta_len-next_offset;
-                bufs[i].state = INIT_META_READING;
-                submitted++;
-                next_offset += bufs[i].size;
-                GET_SQE();
-                data->iov = { bufs[i].buf, bufs[i].size };
-                data->callback = [this, i](ring_data_t *data) { handle_event(data, i); };
-                if (!zero_on_init)
-                    my_uring_prep_readv(sqe, bs->dsk.meta_fd, &data->iov, 1, bs->dsk.meta_offset + bufs[i].offset);
-                else
-                {
-                    // Fill metadata with zeroes
-                    memset(data->iov.iov_base, 0, data->iov.iov_len);
-                    my_uring_prep_writev(sqe, bs->dsk.meta_fd, &data->iov, 1, bs->dsk.meta_offset + bufs[i].offset);
-                }
-                bs->ringloop->submit();
-                break;
-            }
+            wait_state = 2;
+            return 1;
         }
-    }
-    for (int i = 0; i < 2; i++)
-    {
-        if (bufs[i].state == INIT_META_READ_DONE)
+        if (metadata_read < bs->meta_len)
         {
-            // Handle result
-            bool changed = false;
-            for (uint64_t sector = 0; sector < bufs[i].size; sector += bs->dsk.meta_block_size)
-            {
-                // handle <count> entries
-                if (handle_meta_block(bufs[i].buf + sector, entries_per_block,
-                    ((bufs[i].offset + sector - md_offset) / bs->dsk.meta_block_size) * entries_per_block))
-                    changed = true;
-            }
-            if (changed && !bs->inmemory_meta && !bs->readonly)
-            {
-                // write the modified buffer back
-                GET_SQE();
-                data->iov = { bufs[i].buf, bufs[i].size };
-                data->callback = [this, i](ring_data_t *data) { handle_event(data, i); };
-                my_uring_prep_writev(sqe, bs->dsk.meta_fd, &data->iov, 1, bs->dsk.meta_offset + bufs[i].offset);
-                bufs[i].state = INIT_META_WRITING;
-                submitted++;
-            }
+            GET_SQE();
+            data->iov = {
+                (uint8_t*)metadata_buffer + (bs->inmemory_meta
+                    ? metadata_read
+                    : (prev == 1 ? bs->metadata_buf_size : 0)),
+                bs->meta_len - metadata_read > bs->metadata_buf_size ? bs->metadata_buf_size : bs->meta_len - metadata_read,
+            };
+            data->callback = [this](ring_data_t *data) { handle_event(data); };
+            if (!zero_on_init)
+                my_uring_prep_readv(sqe, bs->meta_fd, &data->iov, 1, bs->meta_offset + metadata_read);
             else
             {
-                bufs[i].state = 0;
+                // Fill metadata with zeroes
+                memset(data->iov.iov_base, 0, data->iov.iov_len);
+                my_uring_prep_writev(sqe, bs->meta_fd, &data->iov, 1, bs->meta_offset + metadata_read);
             }
-            bs->ringloop->wakeup();
+            bs->ringloop->submit();
+            submitted = (prev == 1 ? 2 : 1);
+            prev = submitted;
         }
-    }
-    if (submitted > 0)
-    {
-        wait_state = 2;
-        return 1;
-    }
-    if (entries_to_zero.size() && !bs->inmemory_meta && !bs->readonly)
-    {
-        // we have to zero out additional entries
-        for (i = 0; i < entries_to_zero.size(); )
+        if (prev_done)
         {
-            next_offset = entries_to_zero[i]/entries_per_block;
-            for (j = i; j < entries_to_zero.size() && entries_to_zero[j]/entries_per_block == next_offset; j++) {}
-            GET_SQE();
-            data->iov = { metadata_buffer, bs->dsk.meta_block_size };
-            data->callback = [this](ring_data_t *data) { handle_event(data, -1); };
-            my_uring_prep_readv(sqe, bs->dsk.meta_fd, &data->iov, 1, bs->dsk.meta_offset + (1+next_offset)*bs->dsk.meta_block_size);
-            submitted++;
-resume_5:
-            if (submitted > 0)
+            void *done_buf = bs->inmemory_meta
+                ? ((uint8_t*)metadata_buffer + done_pos)
+                : ((uint8_t*)metadata_buffer + (prev_done == 2 ? bs->metadata_buf_size : 0));
+            unsigned count = bs->meta_block_size / bs->clean_entry_size;
+            for (int sector = 0; sector < done_len; sector += bs->meta_block_size)
             {
-                wait_state = 5;
-                return 1;
-            }
-            for (; i < j; i++)
-            {
-                uint64_t pos = (entries_to_zero[i] % entries_per_block);
-                memset((uint8_t*)metadata_buffer + pos*bs->dsk.clean_entry_size, 0, bs->dsk.clean_entry_size);
-            }
-            GET_SQE();
-            data->iov = { metadata_buffer, bs->dsk.meta_block_size };
-            data->callback = [this](ring_data_t *data) { handle_event(data, -1); };
-            my_uring_prep_writev(sqe, bs->dsk.meta_fd, &data->iov, 1, bs->dsk.meta_offset + (1+next_offset)*bs->dsk.meta_block_size);
-            submitted++;
-resume_6:
-            if (submitted > 0)
-            {
-                wait_state = 6;
-                return 1;
+                // handle <count> entries
+                handle_entries((uint8_t*)done_buf + sector, count, bs->block_order);
+                done_cnt += count;
             }
+            prev_done = 0;
+            done_len = 0;
+        }
+        if (!submitted)
+        {
+            break;
         }
-        entries_to_zero.clear();
     }
     // metadata read finished
-    printf("Metadata entries loaded: %lu, free blocks: %lu / %lu\n", entries_loaded, bs->data_alloc->get_free_count(), bs->dsk.block_count);
+    printf("Metadata entries loaded: %lu, free blocks: %lu / %lu\n", entries_loaded, bs->data_alloc->get_free_count(), bs->block_count);
     if (!bs->inmemory_meta)
     {
         free(metadata_buffer);
@@ -255,10 +197,10 @@ resume_6:
     if (zero_on_init && !bs->disable_meta_fsync)
     {
         GET_SQE();
-        my_uring_prep_fsync(sqe, bs->dsk.meta_fd, IORING_FSYNC_DATASYNC);
+        my_uring_prep_fsync(sqe, bs->meta_fd, IORING_FSYNC_DATASYNC);
         data->iov = { 0 };
-        data->callback = [this](ring_data_t *data) { handle_event(data, -1); };
-        submitted++;
+        data->callback = [this](ring_data_t *data) { handle_event(data); };
+        submitted = 1;
         bs->ringloop->submit();
     resume_4:
         if (submitted > 0)
@@ -270,18 +212,14 @@ resume_6:
     return 0;
 }
 
-bool blockstore_init_meta::handle_meta_block(uint8_t *buf, uint64_t entries_per_block, uint64_t done_cnt)
+void blockstore_init_meta::handle_entries(void* entries, unsigned count, int block_order)
 {
-    bool updated = false;
-    uint64_t max_i = entries_per_block;
-    if (max_i > bs->dsk.block_count-done_cnt)
-        max_i = bs->dsk.block_count-done_cnt;
-    for (uint64_t i = 0; i < max_i; i++)
+    for (unsigned i = 0; i < count; i++)
     {
-        clean_disk_entry *entry = (clean_disk_entry*)(buf + i*bs->dsk.clean_entry_size);
-        if (!bs->inmemory_meta && bs->dsk.clean_entry_bitmap_size)
+        clean_disk_entry *entry = (clean_disk_entry*)((uint8_t*)entries + i*bs->clean_entry_size);
+        if (!bs->inmemory_meta && bs->clean_entry_bitmap_size)
         {
-            memcpy(bs->clean_bitmap + (done_cnt+i)*2*bs->dsk.clean_entry_bitmap_size, &entry->bitmap, 2*bs->dsk.clean_entry_bitmap_size);
+            memcpy(bs->clean_bitmap + (done_cnt+i)*2*bs->clean_entry_bitmap_size, &entry->bitmap, 2*bs->clean_entry_bitmap_size);
         }
         if (entry->oid.inode > 0)
         {
@@ -292,39 +230,17 @@ bool blockstore_init_meta::handle_meta_block(uint8_t *buf, uint64_t entries_per_
                 if (clean_it != clean_db.end())
                 {
                     // free the previous block
-                    // here we have to zero out the previous entry because otherwise we'll hit
-                    // "tried to overwrite non-zero metadata entry" later
-                    uint64_t old_clean_loc = clean_it->second.location >> bs->dsk.block_order;
-                    if (bs->inmemory_meta)
-                    {
-                        uint64_t sector = (old_clean_loc / entries_per_block) * bs->dsk.meta_block_size;
-                        uint64_t pos = (old_clean_loc % entries_per_block);
-                        clean_disk_entry *old_entry = (clean_disk_entry*)((uint8_t*)bs->metadata_buffer + sector + pos*bs->dsk.clean_entry_size);
-                        memset(old_entry, 0, bs->dsk.clean_entry_size);
-                    }
-                    else if (old_clean_loc >= done_cnt)
-                    {
-                        updated = true;
-                        uint64_t sector = ((old_clean_loc - done_cnt) / entries_per_block) * bs->dsk.meta_block_size;
-                        uint64_t pos = (old_clean_loc % entries_per_block);
-                        clean_disk_entry *old_entry = (clean_disk_entry*)(buf + sector + pos*bs->dsk.clean_entry_size);
-                        memset(old_entry, 0, bs->dsk.clean_entry_size);
-                    }
-                    else
-                    {
-                        entries_to_zero.push_back(clean_it->second.location >> bs->dsk.block_order);
-                    }
 #ifdef BLOCKSTORE_DEBUG
                     printf("Free block %lu from %lx:%lx v%lu (new location is %lu)\n",
-                        old_clean_loc,
+                        clean_it->second.location >> block_order,
                         clean_it->first.inode, clean_it->first.stripe, clean_it->second.version,
                         done_cnt+i);
 #endif
-                    bs->data_alloc->set(old_clean_loc, false);
+                    bs->data_alloc->set(clean_it->second.location >> block_order, false);
                 }
                 else
                 {
-                    bs->inode_space_stats[entry->oid.inode] += bs->dsk.data_block_size;
+                    bs->inode_space_stats[entry->oid.inode] += bs->data_block_size;
                 }
                 entries_loaded++;
 #ifdef BLOCKSTORE_DEBUG
@@ -333,21 +249,17 @@ bool blockstore_init_meta::handle_meta_block(uint8_t *buf, uint64_t entries_per_
                 bs->data_alloc->set(done_cnt+i, true);
                 clean_db[entry->oid] = (struct clean_entry){
                     .version = entry->version,
-                    .location = (done_cnt+i) << bs->dsk.block_order,
+                    .location = (done_cnt+i) << block_order,
                 };
             }
             else
             {
-                // here we also have to zero out the entry
-                updated = true;
-                memset(entry, 0, bs->dsk.clean_entry_size);
 #ifdef BLOCKSTORE_DEBUG
                 printf("Old clean entry %lu: %lx:%lx v%lu\n", done_cnt+i, entry->oid.inode, entry->oid.stripe, entry->version);
 #endif
             }
         }
     }
-    return updated;
 }
 
 blockstore_init_journal::blockstore_init_journal(blockstore_impl_t *bs)
@@ -416,7 +328,7 @@ int blockstore_init_journal::loop()
     data = ((ring_data_t*)sqe->user_data);
     data->iov = { submitted_buf, bs->journal.block_size };
     data->callback = simple_callback;
-    my_uring_prep_readv(sqe, bs->dsk.journal_fd, &data->iov, 1, bs->journal.offset);
+    my_uring_prep_readv(sqe, bs->journal.fd, &data->iov, 1, bs->journal.offset);
     bs->ringloop->submit();
     wait_count = 1;
 resume_1:
@@ -455,7 +367,7 @@ resume_1:
             GET_SQE();
             data->iov = (struct iovec){ submitted_buf, 2*bs->journal.block_size };
             data->callback = simple_callback;
-            my_uring_prep_writev(sqe, bs->dsk.journal_fd, &data->iov, 1, bs->journal.offset);
+            my_uring_prep_writev(sqe, bs->journal.fd, &data->iov, 1, bs->journal.offset);
             wait_count++;
             bs->ringloop->submit();
         resume_6:
@@ -467,7 +379,7 @@ resume_1:
             if (!bs->disable_journal_fsync)
             {
                 GET_SQE();
-                my_uring_prep_fsync(sqe, bs->dsk.journal_fd, IORING_FSYNC_DATASYNC);
+                my_uring_prep_fsync(sqe, bs->journal.fd, IORING_FSYNC_DATASYNC);
                 data->iov = { 0 };
                 data->callback = simple_callback;
                 wait_count++;
@@ -536,7 +448,7 @@ resume_1:
                     end - journal_pos < JOURNAL_BUFFER_SIZE ? end - journal_pos : JOURNAL_BUFFER_SIZE,
                 };
                 data->callback = [this](ring_data_t *data1) { handle_event(data1); };
-                my_uring_prep_readv(sqe, bs->dsk.journal_fd, &data->iov, 1, bs->journal.offset + journal_pos);
+                my_uring_prep_readv(sqe, bs->journal.fd, &data->iov, 1, bs->journal.offset + journal_pos);
                 bs->ringloop->submit();
             }
             while (done.size() > 0)
@@ -551,7 +463,7 @@ resume_1:
                         GET_SQE();
                         data->iov = { init_write_buf, bs->journal.block_size };
                         data->callback = simple_callback;
-                        my_uring_prep_writev(sqe, bs->dsk.journal_fd, &data->iov, 1, bs->journal.offset + init_write_sector);
+                        my_uring_prep_writev(sqe, bs->journal.fd, &data->iov, 1, bs->journal.offset + init_write_sector);
                         wait_count++;
                         bs->ringloop->submit();
                     resume_7:
@@ -565,7 +477,7 @@ resume_1:
                             GET_SQE();
                             data->iov = { 0 };
                             data->callback = simple_callback;
-                            my_uring_prep_fsync(sqe, bs->dsk.journal_fd, IORING_FSYNC_DATASYNC);
+                            my_uring_prep_fsync(sqe, bs->journal.fd, IORING_FSYNC_DATASYNC);
                             wait_count++;
                             bs->ringloop->submit();
                         }
@@ -632,7 +544,7 @@ resume_1:
             ? bs->journal.len-bs->journal.block_size - (bs->journal.next_free-bs->journal.used_start)
             : bs->journal.used_start - bs->journal.next_free),
         bs->journal.used_start, bs->journal.next_free,
-        bs->data_alloc->get_free_count(), bs->dsk.block_count
+        bs->data_alloc->get_free_count(), bs->block_count
     );
     bs->journal.crc32_last = crc32_last;
     return 0;
@@ -757,9 +669,9 @@ int blockstore_init_journal::handle_journal_part(void *buf, uint64_t done_pos, u
                     };
                     void *bmp = NULL;
                     void *bmp_from = (uint8_t*)je + sizeof(journal_entry_small_write);
-                    if (bs->dsk.clean_entry_bitmap_size <= sizeof(void*))
+                    if (bs->clean_entry_bitmap_size <= sizeof(void*))
                     {
-                        memcpy(&bmp, bmp_from, bs->dsk.clean_entry_bitmap_size);
+                        memcpy(&bmp, bmp_from, bs->clean_entry_bitmap_size);
                     }
                     else
                     {
@@ -767,8 +679,8 @@ int blockstore_init_journal::handle_journal_part(void *buf, uint64_t done_pos, u
                         // allocations for entry bitmaps. This can only be fixed by using
                         // a patched map with dynamic entry size, but not the btree_map,
                         // because it doesn't keep iterators valid all the time.
-                        bmp = malloc_or_die(bs->dsk.clean_entry_bitmap_size);
-                        memcpy(bmp, bmp_from, bs->dsk.clean_entry_bitmap_size);
+                        bmp = malloc_or_die(bs->clean_entry_bitmap_size);
+                        memcpy(bmp, bmp_from, bs->clean_entry_bitmap_size);
                     }
                     bs->dirty_db.emplace(ov, (dirty_entry){
                         .state = (BS_ST_SMALL_WRITE | BS_ST_SYNCED),
@@ -800,7 +712,7 @@ int blockstore_init_journal::handle_journal_part(void *buf, uint64_t done_pos, u
                 printf(
                     "je_big_write%s oid=%lx:%lx ver=%lu loc=%lu\n",
                     je->type == JE_BIG_WRITE_INSTANT ? "_instant" : "",
-                    je->big_write.oid.inode, je->big_write.oid.stripe, je->big_write.version, je->big_write.location >> bs->dsk.block_order
+                    je->big_write.oid.inode, je->big_write.oid.stripe, je->big_write.version, je->big_write.location >> bs->block_order
                 );
 #endif
                 auto dirty_it = bs->dirty_db.upper_bound((obj_ver_id){
@@ -838,9 +750,9 @@ int blockstore_init_journal::handle_journal_part(void *buf, uint64_t done_pos, u
                     };
                     void *bmp = NULL;
                     void *bmp_from = (uint8_t*)je + sizeof(journal_entry_big_write);
-                    if (bs->dsk.clean_entry_bitmap_size <= sizeof(void*))
+                    if (bs->clean_entry_bitmap_size <= sizeof(void*))
                     {
-                        memcpy(&bmp, bmp_from, bs->dsk.clean_entry_bitmap_size);
+                        memcpy(&bmp, bmp_from, bs->clean_entry_bitmap_size);
                     }
                     else
                     {
@@ -848,8 +760,8 @@ int blockstore_init_journal::handle_journal_part(void *buf, uint64_t done_pos, u
                         // allocations for entry bitmaps. This can only be fixed by using
                         // a patched map with dynamic entry size, but not the btree_map,
                         // because it doesn't keep iterators valid all the time.
-                        bmp = malloc_or_die(bs->dsk.clean_entry_bitmap_size);
-                        memcpy(bmp, bmp_from, bs->dsk.clean_entry_bitmap_size);
+                        bmp = malloc_or_die(bs->clean_entry_bitmap_size);
+                        memcpy(bmp, bmp_from, bs->clean_entry_bitmap_size);
                     }
                     auto dirty_it = bs->dirty_db.emplace(ov, (dirty_entry){
                         .state = (BS_ST_BIG_WRITE | BS_ST_SYNCED),
@@ -860,7 +772,7 @@ int blockstore_init_journal::handle_journal_part(void *buf, uint64_t done_pos, u
                         .journal_sector = proc_pos,
                         .bitmap = bmp,
                     }).first;
-                    if (bs->data_alloc->get(je->big_write.location >> bs->dsk.block_order))
+                    if (bs->data_alloc->get(je->big_write.location >> bs->block_order))
                     {
                         // This is probably a big_write that's already flushed and freed, but it may
                         // also indicate a bug. So we remember such entries and recheck them afterwards.
@@ -873,11 +785,11 @@ int blockstore_init_journal::handle_journal_part(void *buf, uint64_t done_pos, u
 #ifdef BLOCKSTORE_DEBUG
                         printf(
                             "Allocate block (journal) %lu: %lx:%lx v%lu\n",
-                            je->big_write.location >> bs->dsk.block_order,
+                            je->big_write.location >> bs->block_order,
                             ov.oid.inode, ov.oid.stripe, ov.version
                         );
 #endif
-                        bs->data_alloc->set(je->big_write.location >> bs->dsk.block_order, true);
+                        bs->data_alloc->set(je->big_write.location >> bs->block_order, true);
                     }
                     bs->journal.used_sectors[proc_pos]++;
 #ifdef BLOCKSTORE_DEBUG
@@ -1001,8 +913,8 @@ void blockstore_init_journal::erase_dirty_object(blockstore_dirty_db_t::iterator
     if (exists && clean_loc == UINT64_MAX)
     {
         auto & sp = bs->inode_space_stats[oid.inode];
-        if (sp > bs->dsk.data_block_size)
-            sp -= bs->dsk.data_block_size;
+        if (sp > bs->data_block_size)
+            sp -= bs->data_block_size;
         else
             bs->inode_space_stats.erase(oid.inode);
     }

src/blockstore_init.h

@@ -3,32 +3,20 @@
 
 #pragma once
 
-struct blockstore_init_meta_buf
-{
-    uint8_t *buf = NULL;
-    uint64_t size = 0;
-    uint64_t offset = 0;
-    int state = 0;
-};
-
 class blockstore_init_meta
 {
     blockstore_impl_t *bs;
     int wait_state = 0;
     bool zero_on_init = false;
     void *metadata_buffer = NULL;
-    blockstore_init_meta_buf bufs[2] = {};
-    int submitted = 0;
+    uint64_t metadata_read = 0;
+    int prev = 0, prev_done = 0, done_len = 0, submitted = 0;
+    uint64_t done_cnt = 0, done_pos = 0;
+    uint64_t entries_loaded = 0;
     struct io_uring_sqe *sqe;
     struct ring_data_t *data;
-    uint64_t md_offset = 0;
-    uint64_t next_offset = 0;
-    uint64_t entries_loaded = 0;
-    unsigned entries_per_block = 0;
-    int i = 0, j = 0;
-    std::vector<uint64_t> entries_to_zero;
-    bool handle_meta_block(uint8_t *buf, uint64_t count, uint64_t done_cnt);
-    void handle_event(ring_data_t *data, int buf_num);
+    void handle_entries(void *entries, unsigned count, int block_order);
+    void handle_event(ring_data_t *data);
 public:
     blockstore_init_meta(blockstore_impl_t *bs);
     int loop();

src/blockstore_journal.cpp

@@ -175,7 +175,7 @@ void blockstore_impl_t::prepare_journal_sector_write(int cur_sector, blockstore_
         };
         data->callback = [this, flush_id = journal.submit_id](ring_data_t *data) { handle_journal_write(data, flush_id); };
         my_uring_prep_writev(
-            sqe, dsk.journal_fd, &data->iov, 1, journal.offset + journal.sector_info[cur_sector].offset
+            sqe, journal.fd, &data->iov, 1, journal.offset + journal.sector_info[cur_sector].offset
         );
     }
     journal.sector_info[cur_sector].dirty = false;
@@ -198,7 +198,10 @@ void blockstore_impl_t::handle_journal_write(ring_data_t *data, uint64_t flush_i
     if (data->res != data->iov.iov_len)
     {
         // FIXME: our state becomes corrupted after a write error. maybe do something better than just die
-        disk_error_abort("journal write", data->res, data->iov.iov_len);
+        throw std::runtime_error(
+            "journal write failed ("+std::to_string(data->res)+" != "+std::to_string(data->iov.iov_len)+
+            "). in-memory state is corrupted. AAAAAAAaaaaaaaaa!!!111"
+        );
     }
     auto fl_it = journal.flushing_ops.upper_bound((pending_journaling_t){ .flush_id = flush_id });
     if (fl_it != journal.flushing_ops.end() && fl_it->flush_id == flush_id)

src/blockstore_journal.h

@@ -10,13 +10,11 @@
 #define JOURNAL_MAGIC 0x4A33
 #define JOURNAL_VERSION 1
 #define JOURNAL_BUFFER_SIZE 4*1024*1024
-#define JOURNAL_ENTRY_HEADER_SIZE 16
 
 // We reserve some extra space for future stabilize requests during writes
 // FIXME: This value should be dynamic i.e. Blockstore ideally shouldn't allow
 // writing more than can be stabilized afterwards
 #define JOURNAL_STABILIZE_RESERVATION 65536
-#define JOURNAL_INSTANT_RESERVATION 131072
 
 // Journal entries
 // Journal entries are linked to each other by their crc32 value
@@ -166,6 +164,7 @@ inline bool operator < (const pending_journaling_t & a, const pending_journaling
 struct journal_t
 {
     int fd;
+    uint64_t device_size;
     bool inmemory = false;
     bool flush_journal = false;
     void *buffer = NULL;

src/blockstore_open.cpp

@@ -4,10 +4,23 @@
 #include <sys/file.h>
 #include "blockstore_impl.h"
 
+static uint32_t is_power_of_two(uint64_t value)
+{
+    uint32_t l = 0;
+    while (value > 1)
+    {
+        if (value & 1)
+        {
+            return 64;
+        }
+        value = value >> 1;
+        l++;
+    }
+    return l;
+}
+
 void blockstore_impl_t::parse_config(blockstore_config_t & config)
 {
-    // Common disk options
-    dsk.parse_config(config);
     // Parse
     if (config["readonly"] == "true" || config["readonly"] == "1" || config["readonly"] == "yes")
     {
@@ -25,6 +38,10 @@ void blockstore_impl_t::parse_config(blockstore_config_t & config)
     {
         disable_journal_fsync = true;
     }
+    if (config["disable_device_lock"] == "true" || config["disable_device_lock"] == "1" || config["disable_device_lock"] == "yes")
+    {
+        disable_flock = true;
+    }
     if (config["flush_journal"] == "true" || config["flush_journal"] == "1" || config["flush_journal"] == "yes")
     {
         // Only flush journal and exit
@@ -39,11 +56,24 @@ void blockstore_impl_t::parse_config(blockstore_config_t & config)
         immediate_commit = IMMEDIATE_SMALL;
     }
     metadata_buf_size = strtoull(config["meta_buf_size"].c_str(), NULL, 10);
+    cfg_journal_size = strtoull(config["journal_size"].c_str(), NULL, 10);
+    data_device = config["data_device"];
+    data_offset = strtoull(config["data_offset"].c_str(), NULL, 10);
+    cfg_data_size = strtoull(config["data_size"].c_str(), NULL, 10);
+    meta_device = config["meta_device"];
+    meta_offset = strtoull(config["meta_offset"].c_str(), NULL, 10);
+    data_block_size = strtoull(config["block_size"].c_str(), NULL, 10);
     inmemory_meta = config["inmemory_metadata"] != "false";
+    journal_device = config["journal_device"];
+    journal.offset = strtoull(config["journal_offset"].c_str(), NULL, 10);
     journal.sector_count = strtoull(config["journal_sector_buffer_count"].c_str(), NULL, 10);
     journal.no_same_sector_overwrites = config["journal_no_same_sector_overwrites"] == "true" ||
         config["journal_no_same_sector_overwrites"] == "1" || config["journal_no_same_sector_overwrites"] == "yes";
     journal.inmemory = config["inmemory_journal"] != "false";
+    disk_alignment = strtoull(config["disk_alignment"].c_str(), NULL, 10);
+    journal_block_size = strtoull(config["journal_block_size"].c_str(), NULL, 10);
+    meta_block_size = strtoull(config["meta_block_size"].c_str(), NULL, 10);
+    bitmap_granularity = strtoull(config["bitmap_granularity"].c_str(), NULL, 10);
     max_flusher_count = strtoull(config["max_flusher_count"].c_str(), NULL, 10);
     if (!max_flusher_count)
         max_flusher_count = strtoull(config["flusher_count"].c_str(), NULL, 10);
@@ -55,6 +85,14 @@ void blockstore_impl_t::parse_config(blockstore_config_t & config)
     throttle_target_parallelism = strtoull(config["throttle_target_parallelism"].c_str(), NULL, 10);
     throttle_threshold_us = strtoull(config["throttle_threshold_us"].c_str(), NULL, 10);
     // Validate
+    if (!data_block_size)
+    {
+        data_block_size = (1 << DEFAULT_DATA_BLOCK_ORDER);
+    }
+    if ((block_order = is_power_of_two(data_block_size)) >= 64 || data_block_size < MIN_DATA_BLOCK_SIZE || data_block_size >= MAX_DATA_BLOCK_SIZE)
+    {
+        throw std::runtime_error("Bad block size");
+    }
     if (!max_flusher_count)
     {
         max_flusher_count = 256;
@@ -67,6 +105,62 @@ void blockstore_impl_t::parse_config(blockstore_config_t & config)
     {
         max_write_iodepth = 128;
     }
+    if (!disk_alignment)
+    {
+        disk_alignment = 4096;
+    }
+    else if (disk_alignment % DIRECT_IO_ALIGNMENT)
+    {
+        throw std::runtime_error("disk_alignment must be a multiple of "+std::to_string(DIRECT_IO_ALIGNMENT));
+    }
+    if (!journal_block_size)
+    {
+        journal_block_size = 4096;
+    }
+    else if (journal_block_size % DIRECT_IO_ALIGNMENT)
+    {
+        throw std::runtime_error("journal_block_size must be a multiple of "+std::to_string(DIRECT_IO_ALIGNMENT));
+    }
+    if (!meta_block_size)
+    {
+        meta_block_size = 4096;
+    }
+    else if (meta_block_size % DIRECT_IO_ALIGNMENT)
+    {
+        throw std::runtime_error("meta_block_size must be a multiple of "+std::to_string(DIRECT_IO_ALIGNMENT));
+    }
+    if (data_offset % disk_alignment)
+    {
+        throw std::runtime_error("data_offset must be a multiple of disk_alignment = "+std::to_string(disk_alignment));
+    }
+    if (!bitmap_granularity)
+    {
+        bitmap_granularity = DEFAULT_BITMAP_GRANULARITY;
+    }
+    else if (bitmap_granularity % disk_alignment)
+    {
+        throw std::runtime_error("Sparse write tracking granularity must be a multiple of disk_alignment = "+std::to_string(disk_alignment));
+    }
+    if (data_block_size % bitmap_granularity)
+    {
+        throw std::runtime_error("Block size must be a multiple of sparse write tracking granularity");
+    }
+    if (journal_device == meta_device || meta_device == "" && journal_device == data_device)
+    {
+        journal_device = "";
+    }
+    if (meta_device == data_device)
+    {
+        meta_device = "";
+    }
+    if (meta_offset % meta_block_size)
+    {
+        throw std::runtime_error("meta_offset must be a multiple of meta_block_size = "+std::to_string(meta_block_size));
+    }
+    if (journal.offset % journal_block_size)
+    {
+        throw std::runtime_error("journal_offset must be a multiple of journal_block_size = "+std::to_string(journal_block_size));
+    }
     if (journal.sector_count < 2)
     {
         journal.sector_count = 32;
@@ -75,11 +169,11 @@ void blockstore_impl_t::parse_config(blockstore_config_t & config)
     {
         metadata_buf_size = 4*1024*1024;
     }
-    if (dsk.meta_device == dsk.data_device)
+    if (meta_device == "")
     {
         disable_meta_fsync = disable_data_fsync;
     }
-    if (dsk.journal_device == dsk.meta_device)
+    if (journal_device == "")
     {
         disable_journal_fsync = disable_meta_fsync;
     }
@@ -108,46 +202,238 @@ void blockstore_impl_t::parse_config(blockstore_config_t & config)
         throttle_threshold_us = 50;
     }
     // init some fields
-    journal.block_size = dsk.journal_block_size;
-    journal.next_free = dsk.journal_block_size;
-    journal.used_start = dsk.journal_block_size;
+    clean_entry_bitmap_size = data_block_size / bitmap_granularity / 8;
+    clean_entry_size = sizeof(clean_disk_entry) + 2*clean_entry_bitmap_size;
+    journal.block_size = journal_block_size;
+    journal.next_free = journal_block_size;
+    journal.used_start = journal_block_size;
     // no free space because sector is initially unmapped
-    journal.in_sector_pos = dsk.journal_block_size;
+    journal.in_sector_pos = journal_block_size;
 }
 
 void blockstore_impl_t::calc_lengths()
 {
-    dsk.calc_lengths();
-    journal.len = dsk.journal_len;
-    journal.block_size = dsk.journal_block_size;
-    journal.offset = dsk.journal_offset;
+    // data
+    data_len = data_device_size - data_offset;
+    if (data_fd == meta_fd && data_offset < meta_offset)
+    {
+        data_len = meta_offset - data_offset;
+    }
+    if (data_fd == journal.fd && data_offset < journal.offset)
+    {
+        data_len = data_len < journal.offset-data_offset
+            ? data_len : journal.offset-data_offset;
+    }
+    if (cfg_data_size != 0)
+    {
+        if (data_len < cfg_data_size)
+        {
+            throw std::runtime_error("Data area ("+std::to_string(data_len)+
+                " bytes) is less than configured size ("+std::to_string(cfg_data_size)+" bytes)");
+        }
+        data_len = cfg_data_size;
+    }
+    // meta
+    uint64_t meta_area_size = (meta_fd == data_fd ? data_device_size : meta_device_size) - meta_offset;
+    if (meta_fd == data_fd && meta_offset <= data_offset)
+    {
+        meta_area_size = data_offset - meta_offset;
+    }
+    if (meta_fd == journal.fd && meta_offset <= journal.offset)
+    {
+        meta_area_size = meta_area_size < journal.offset-meta_offset
+            ? meta_area_size : journal.offset-meta_offset;
+    }
+    // journal
+    journal.len = (journal.fd == data_fd ? data_device_size : (journal.fd == meta_fd ? meta_device_size : journal.device_size)) - journal.offset;
+    if (journal.fd == data_fd && journal.offset <= data_offset)
+    {
+        journal.len = data_offset - journal.offset;
+    }
+    if (journal.fd == meta_fd && journal.offset <= meta_offset)
+    {
+        journal.len = journal.len < meta_offset-journal.offset
+            ? journal.len : meta_offset-journal.offset;
+    }
+    // required metadata size
+    block_count = data_len / data_block_size;
+    meta_len = (1 + (block_count - 1 + meta_block_size / clean_entry_size) / (meta_block_size / clean_entry_size)) * meta_block_size;
+    if (meta_area_size < meta_len)
+    {
+        throw std::runtime_error("Metadata area is too small, need at least "+std::to_string(meta_len)+" bytes");
+    }
     if (inmemory_meta)
     {
-        metadata_buffer = memalign(MEM_ALIGNMENT, dsk.meta_len);
+        metadata_buffer = memalign(MEM_ALIGNMENT, meta_len);
         if (!metadata_buffer)
             throw std::runtime_error("Failed to allocate memory for the metadata");
     }
-    else if (dsk.clean_entry_bitmap_size)
+    else if (clean_entry_bitmap_size)
     {
-        clean_bitmap = (uint8_t*)malloc(dsk.block_count * 2*dsk.clean_entry_bitmap_size);
+        clean_bitmap = (uint8_t*)malloc(block_count * 2*clean_entry_bitmap_size);
         if (!clean_bitmap)
             throw std::runtime_error("Failed to allocate memory for the metadata sparse write bitmap");
     }
+    // requested journal size
+    if (cfg_journal_size > journal.len)
+    {
+        throw std::runtime_error("Requested journal_size is too large");
+    }
+    else if (cfg_journal_size > 0)
+    {
+        journal.len = cfg_journal_size;
+    }
+    if (journal.len < MIN_JOURNAL_SIZE)
+    {
+        throw std::runtime_error("Journal is too small, need at least "+std::to_string(MIN_JOURNAL_SIZE)+" bytes");
+    }
     if (journal.inmemory)
     {
         journal.buffer = memalign(MEM_ALIGNMENT, journal.len);
         if (!journal.buffer)
             throw std::runtime_error("Failed to allocate memory for journal");
     }
+}
+
+static void check_size(int fd, uint64_t *size, uint64_t *sectsize, std::string name)
+{
+    int sect;
+    struct stat st;
+    if (fstat(fd, &st) < 0)
+    {
+        throw std::runtime_error("Failed to stat "+name);
+    }
+    if (S_ISREG(st.st_mode))
+    {
+        *size = st.st_size;
+        if (sectsize)
+        {
+            *sectsize = st.st_blksize;
+        }
+    }
+    else if (S_ISBLK(st.st_mode))
+    {
+        if (ioctl(fd, BLKGETSIZE64, size) < 0 ||
+            ioctl(fd, BLKSSZGET, &sect) < 0)
+        {
+            throw std::runtime_error("Failed to get "+name+" size or block size: "+strerror(errno));
+        }
+        if (sectsize)
+        {
+            *sectsize = sect;
+        }
+    }
     else
     {
-        journal.sector_buf = (uint8_t*)memalign(MEM_ALIGNMENT, journal.sector_count * dsk.journal_block_size);
-        if (!journal.sector_buf)
-            throw std::bad_alloc();
+        throw std::runtime_error(name+" is neither a file nor a block device");
+    }
+}
+
+void blockstore_impl_t::open_data()
+{
+    data_fd = open(data_device.c_str(), O_DIRECT|O_RDWR);
+    if (data_fd == -1)
+    {
+        throw std::runtime_error("Failed to open data device");
+    }
+    check_size(data_fd, &data_device_size, &data_device_sect, "data device");
+    if (disk_alignment % data_device_sect)
+    {
+        throw std::runtime_error(
+            "disk_alignment ("+std::to_string(disk_alignment)+
+            ") is not a multiple of data device sector size ("+std::to_string(data_device_sect)+")"
+        );
+    }
+    if (data_offset >= data_device_size)
+    {
+        throw std::runtime_error("data_offset exceeds device size = "+std::to_string(data_device_size));
+    }
+    if (!disable_flock && flock(data_fd, LOCK_EX|LOCK_NB) != 0)
+    {
+        throw std::runtime_error(std::string("Failed to lock data device: ") + strerror(errno));
+    }
+}
+
+void blockstore_impl_t::open_meta()
+{
+    if (meta_device != "")
+    {
+        meta_offset = 0;
+        meta_fd = open(meta_device.c_str(), O_DIRECT|O_RDWR);
+        if (meta_fd == -1)
+        {
+            throw std::runtime_error("Failed to open metadata device");
+        }
+        check_size(meta_fd, &meta_device_size, &meta_device_sect, "metadata device");
+        if (meta_offset >= meta_device_size)
+        {
+            throw std::runtime_error("meta_offset exceeds device size = "+std::to_string(meta_device_size));
+        }
+        if (!disable_flock && flock(meta_fd, LOCK_EX|LOCK_NB) != 0)
+        {
+            throw std::runtime_error(std::string("Failed to lock metadata device: ") + strerror(errno));
+        }
+    }
+    else
+    {
+        meta_fd = data_fd;
+        meta_device_sect = data_device_sect;
+        meta_device_size = 0;
+        if (meta_offset >= data_device_size)
+        {
+            throw std::runtime_error("meta_offset exceeds device size = "+std::to_string(data_device_size));
+        }
+    }
+    if (meta_block_size % meta_device_sect)
+    {
+        throw std::runtime_error(
+            "meta_block_size ("+std::to_string(meta_block_size)+
+            ") is not a multiple of data device sector size ("+std::to_string(meta_device_sect)+")"
+        );
+    }
+}
+
+void blockstore_impl_t::open_journal()
+{
+    if (journal_device != "")
+    {
+        journal.fd = open(journal_device.c_str(), O_DIRECT|O_RDWR);
+        if (journal.fd == -1)
+        {
+            throw std::runtime_error("Failed to open journal device");
+        }
+        check_size(journal.fd, &journal.device_size, &journal_device_sect, "journal device");
+        if (!disable_flock && flock(journal.fd, LOCK_EX|LOCK_NB) != 0)
+        {
+            throw std::runtime_error(std::string("Failed to lock journal device: ") + strerror(errno));
+        }
+    }
+    else
+    {
+        journal.fd = meta_fd;
+        journal_device_sect = meta_device_sect;
+        journal.device_size = 0;
+        if (journal.offset >= data_device_size)
+        {
+            throw std::runtime_error("journal_offset exceeds device size");
+        }
     }
     journal.sector_info = (journal_sector_info_t*)calloc(journal.sector_count, sizeof(journal_sector_info_t));
     if (!journal.sector_info)
     {
         throw std::bad_alloc();
     }
+    if (!journal.inmemory)
+    {
+        journal.sector_buf = (uint8_t*)memalign(MEM_ALIGNMENT, journal.sector_count * journal_block_size);
+        if (!journal.sector_buf)
+            throw std::bad_alloc();
+    }
+    if (journal_block_size % journal_device_sect)
+    {
+        throw std::runtime_error(
+            "journal_block_size ("+std::to_string(journal_block_size)+
+            ") is not a multiple of journal device sector size ("+std::to_string(journal_device_sect)+")"
+        );
+    }
 }

src/blockstore_read.cpp

@@ -32,9 +32,9 @@ int blockstore_impl_t::fulfill_read_push(blockstore_op_t *op, void *buf, uint64_
     PRIV(op)->pending_ops++;
     my_uring_prep_readv(
         sqe,
-        IS_JOURNAL(item_state) ? dsk.journal_fd : dsk.data_fd,
+        IS_JOURNAL(item_state) ? journal.fd : data_fd,
         &data->iov, 1,
-        (IS_JOURNAL(item_state) ? dsk.journal_offset : dsk.data_offset) + offset
+        (IS_JOURNAL(item_state) ? journal.offset : data_offset) + offset
     );
     data->callback = [this, op](ring_data_t *data) { handle_read_event(data, op); };
     return 1;
@@ -42,7 +42,7 @@ int blockstore_impl_t::fulfill_read_push(blockstore_op_t *op, void *buf, uint64_
 
 // FIXME I've seen a bug here so I want some tests
 int blockstore_impl_t::fulfill_read(blockstore_op_t *read_op, uint64_t &fulfilled, uint32_t item_start, uint32_t item_end,
-    uint32_t item_state, uint64_t item_version, uint64_t item_location, uint64_t journal_sector)
+    uint32_t item_state, uint64_t item_version, uint64_t item_location)
 {
     uint32_t cur_start = item_start;
     if (cur_start < read_op->offset + read_op->len && item_end > read_op->offset)
@@ -72,7 +72,6 @@ int blockstore_impl_t::fulfill_read(blockstore_op_t *read_op, uint64_t &fulfille
                 fulfill_read_t el = {
                     .offset = cur_start,
                     .len = it == PRIV(read_op)->read_vec.end() || it->offset >= item_end ? item_end-cur_start : it->offset-cur_start,
-                    .journal_sector = journal_sector,
                 };
                 it = PRIV(read_op)->read_vec.insert(it, el);
                 if (!fulfill_read_push(read_op,
@@ -98,15 +97,15 @@ endwhile:
 uint8_t* blockstore_impl_t::get_clean_entry_bitmap(uint64_t block_loc, int offset)
 {
     uint8_t *clean_entry_bitmap;
-    uint64_t meta_loc = block_loc >> dsk.block_order;
+    uint64_t meta_loc = block_loc >> block_order;
     if (inmemory_meta)
     {
-        uint64_t sector = (meta_loc / (dsk.meta_block_size / dsk.clean_entry_size)) * dsk.meta_block_size;
-        uint64_t pos = (meta_loc % (dsk.meta_block_size / dsk.clean_entry_size));
-        clean_entry_bitmap = ((uint8_t*)metadata_buffer + sector + pos*dsk.clean_entry_size + sizeof(clean_disk_entry) + offset);
+        uint64_t sector = (meta_loc / (meta_block_size / clean_entry_size)) * meta_block_size;
+        uint64_t pos = (meta_loc % (meta_block_size / clean_entry_size));
+        clean_entry_bitmap = ((uint8_t*)metadata_buffer + sector + pos*clean_entry_size + sizeof(clean_disk_entry) + offset);
     }
     else
-        clean_entry_bitmap = (uint8_t*)(clean_bitmap + meta_loc*2*dsk.clean_entry_bitmap_size + offset);
+        clean_entry_bitmap = (uint8_t*)(clean_bitmap + meta_loc*2*clean_entry_bitmap_size + offset);
     return clean_entry_bitmap;
 }
 
@@ -139,7 +138,7 @@ int blockstore_impl_t::dequeue_read(blockstore_op_t *read_op)
         while (dirty_it->first.oid == read_op->oid)
         {
             dirty_entry& dirty = dirty_it->second;
-            bool version_ok = !IS_IN_FLIGHT(dirty.state) && read_op->version >= dirty_it->first.version;
+            bool version_ok = read_op->version >= dirty_it->first.version;
             if (IS_SYNCED(dirty.state))
             {
                 if (!version_ok && read_op->version != 0)
@@ -153,14 +152,12 @@ int blockstore_impl_t::dequeue_read(blockstore_op_t *read_op)
                     result_version = dirty_it->first.version;
                     if (read_op->bitmap)
                     {
-                        void *bmp_ptr = (dsk.clean_entry_bitmap_size > sizeof(void*) ? dirty_it->second.bitmap : &dirty_it->second.bitmap);
-                        memcpy(read_op->bitmap, bmp_ptr, dsk.clean_entry_bitmap_size);
+                        void *bmp_ptr = (clean_entry_bitmap_size > sizeof(void*) ? dirty_it->second.bitmap : &dirty_it->second.bitmap);
+                        memcpy(read_op->bitmap, bmp_ptr, clean_entry_bitmap_size);
                     }
                 }
-                // If inmemory_journal is false, journal trim will have to wait until the read is completed
                 if (!fulfill_read(read_op, fulfilled, dirty.offset, dirty.offset + dirty.len,
-                    dirty.state, dirty_it->first.version, dirty.location + (IS_JOURNAL(dirty.state) ? 0 : dirty.offset),
-                    (IS_JOURNAL(dirty.state) ? dirty.journal_sector+1 : 0)))
+                    dirty.state, dirty_it->first.version, dirty.location + (IS_JOURNAL(dirty.state) ? 0 : dirty.offset)))
                 {
                     // need to wait. undo added requests, don't dequeue op
                     PRIV(read_op)->read_vec.clear();
@@ -174,23 +171,22 @@ int blockstore_impl_t::dequeue_read(blockstore_op_t *read_op)
             dirty_it--;
         }
     }
-    if (clean_found)
+    if (clean_it != clean_db.end())
     {
         if (!result_version)
         {
             result_version = clean_it->second.version;
             if (read_op->bitmap)
             {
-                void *bmp_ptr = get_clean_entry_bitmap(clean_it->second.location, dsk.clean_entry_bitmap_size);
-                memcpy(read_op->bitmap, bmp_ptr, dsk.clean_entry_bitmap_size);
+                void *bmp_ptr = get_clean_entry_bitmap(clean_it->second.location, clean_entry_bitmap_size);
+                memcpy(read_op->bitmap, bmp_ptr, clean_entry_bitmap_size);
             }
         }
         if (fulfilled < read_op->len)
         {
-            if (!dsk.clean_entry_bitmap_size)
+            if (!clean_entry_bitmap_size)
             {
-                if (!fulfill_read(read_op, fulfilled, 0, dsk.data_block_size,
-                    (BS_ST_BIG_WRITE | BS_ST_STABLE), 0, clean_it->second.location, 0))
+                if (!fulfill_read(read_op, fulfilled, 0, data_block_size, (BS_ST_BIG_WRITE | BS_ST_STABLE), 0, clean_it->second.location))
                 {
                     // need to wait. undo added requests, don't dequeue op
                     PRIV(read_op)->read_vec.clear();
@@ -200,7 +196,7 @@ int blockstore_impl_t::dequeue_read(blockstore_op_t *read_op)
             else
             {
                 uint8_t *clean_entry_bitmap = get_clean_entry_bitmap(clean_it->second.location, 0);
-                uint64_t bmp_start = 0, bmp_end = 0, bmp_size = dsk.data_block_size/dsk.bitmap_granularity;
+                uint64_t bmp_start = 0, bmp_end = 0, bmp_size = data_block_size/bitmap_granularity;
                 while (bmp_start < bmp_size)
                 {
                     while (!(clean_entry_bitmap[bmp_end >> 3] & (1 << (bmp_end & 0x7))) && bmp_end < bmp_size)
@@ -210,8 +206,8 @@ int blockstore_impl_t::dequeue_read(blockstore_op_t *read_op)
                     if (bmp_end > bmp_start)
                     {
                         // fill with zeroes
-                        assert(fulfill_read(read_op, fulfilled, bmp_start * dsk.bitmap_granularity,
-                            bmp_end * dsk.bitmap_granularity, (BS_ST_DELETE | BS_ST_STABLE), 0, 0, 0));
+                        assert(fulfill_read(read_op, fulfilled, bmp_start * bitmap_granularity,
+                            bmp_end * bitmap_granularity, (BS_ST_DELETE | BS_ST_STABLE), 0, 0));
                     }
                     bmp_start = bmp_end;
                     while (clean_entry_bitmap[bmp_end >> 3] & (1 << (bmp_end & 0x7)) && bmp_end < bmp_size)
@@ -220,9 +216,9 @@ int blockstore_impl_t::dequeue_read(blockstore_op_t *read_op)
                     }
                     if (bmp_end > bmp_start)
                     {
-                        if (!fulfill_read(read_op, fulfilled, bmp_start * dsk.bitmap_granularity,
-                            bmp_end * dsk.bitmap_granularity, (BS_ST_BIG_WRITE | BS_ST_STABLE), 0,
-                            clean_it->second.location + bmp_start * dsk.bitmap_granularity, 0))
+                        if (!fulfill_read(read_op, fulfilled, bmp_start * bitmap_granularity,
+                            bmp_end * bitmap_granularity, (BS_ST_BIG_WRITE | BS_ST_STABLE), 0,
+                            clean_it->second.location + bmp_start * bitmap_granularity))
                         {
                             // need to wait. undo added requests, don't dequeue op
                             PRIV(read_op)->read_vec.clear();
@@ -237,7 +233,7 @@ int blockstore_impl_t::dequeue_read(blockstore_op_t *read_op)
     else if (fulfilled < read_op->len)
     {
         // fill remaining parts with zeroes
-        assert(fulfill_read(read_op, fulfilled, 0, dsk.data_block_size, (BS_ST_DELETE | BS_ST_STABLE), 0, 0, 0));
+        assert(fulfill_read(read_op, fulfilled, 0, data_block_size, (BS_ST_DELETE | BS_ST_STABLE), 0, 0));
     }
     assert(fulfilled == read_op->len);
     read_op->version = result_version;
@@ -253,15 +249,6 @@ int blockstore_impl_t::dequeue_read(blockstore_op_t *read_op)
         FINISH_OP(read_op);
         return 2;
     }
-    if (!journal.inmemory)
-    {
-        // Journal trim has to wait until the read is completed - record journal sector usage
-        for (auto & rv: PRIV(read_op)->read_vec)
-        {
-            if (rv.journal_sector)
-                journal.used_sectors[rv.journal_sector-1]++;
-        }
-    }
     read_op->retval = 0;
     return 2;
 }
@@ -277,22 +264,6 @@ void blockstore_impl_t::handle_read_event(ring_data_t *data, blockstore_op_t *op
     }
     if (PRIV(op)->pending_ops == 0)
     {
-        if (!journal.inmemory)
-        {
-            // Release journal sector usage
-            for (auto & rv: PRIV(op)->read_vec)
-            {
-                if (rv.journal_sector)
-                {
-                    auto used = --journal.used_sectors[rv.journal_sector-1];
-                    if (used == 0)
-                    {
-                        journal.used_sectors.erase(rv.journal_sector-1);
-                        flusher->mark_trim_possible();
-                    }
-                }
-            }
-        }
         if (op->retval == 0)
             op->retval = op->len;
         FINISH_OP(op);
@@ -317,8 +288,8 @@ int blockstore_impl_t::read_bitmap(object_id oid, uint64_t target_version, void
                     *result_version = dirty_it->first.version;
                 if (bitmap)
                 {
-                    void *bmp_ptr = (dsk.clean_entry_bitmap_size > sizeof(void*) ? dirty_it->second.bitmap : &dirty_it->second.bitmap);
-                    memcpy(bitmap, bmp_ptr, dsk.clean_entry_bitmap_size);
+                    void *bmp_ptr = (clean_entry_bitmap_size > sizeof(void*) ? dirty_it->second.bitmap : &dirty_it->second.bitmap);
+                    memcpy(bitmap, bmp_ptr, clean_entry_bitmap_size);
                 }
                 return 0;
             }
@@ -335,14 +306,14 @@ int blockstore_impl_t::read_bitmap(object_id oid, uint64_t target_version, void
             *result_version = clean_it->second.version;
         if (bitmap)
         {
-            void *bmp_ptr = get_clean_entry_bitmap(clean_it->second.location, dsk.clean_entry_bitmap_size);
-            memcpy(bitmap, bmp_ptr, dsk.clean_entry_bitmap_size);
+            void *bmp_ptr = get_clean_entry_bitmap(clean_it->second.location, clean_entry_bitmap_size);
+            memcpy(bitmap, bmp_ptr, clean_entry_bitmap_size);
         }
         return 0;
     }
     if (result_version)
         *result_version = 0;
     if (bitmap)
-        memset(bitmap, 0, dsk.clean_entry_bitmap_size);
+        memset(bitmap, 0, clean_entry_bitmap_size);
     return -ENOENT;
 }

src/blockstore_rollback.cpp

@@ -112,7 +112,7 @@ resume_2:
     if (!disable_journal_fsync)
     {
         BS_SUBMIT_GET_SQE(sqe, data);
-        my_uring_prep_fsync(sqe, dsk.journal_fd, IORING_FSYNC_DATASYNC);
+        my_uring_prep_fsync(sqe, journal.fd, IORING_FSYNC_DATASYNC);
         data->iov = { 0 };
         data->callback = [this, op](ring_data_t *data) { handle_write_event(data, op); };
         PRIV(op)->min_flushed_journal_sector = PRIV(op)->max_flushed_journal_sector = 0;
@@ -127,6 +127,7 @@ resume_4:
     {
         mark_rolled_back(*v);
     }
+    flusher->mark_trim_possible();
     // Acknowledge op
     op->retval = 0;
     FINISH_OP(op);
@@ -216,12 +217,12 @@ void blockstore_impl_t::erase_dirty(blockstore_dirty_db_t::iterator dirty_start,
             dirty_it->second.location != UINT64_MAX)
         {
 #ifdef BLOCKSTORE_DEBUG
-            printf("Free block %lu from %lx:%lx v%lu\n", dirty_it->second.location >> dsk.block_order,
+            printf("Free block %lu from %lx:%lx v%lu\n", dirty_it->second.location >> block_order,
                 dirty_it->first.oid.inode, dirty_it->first.oid.stripe, dirty_it->first.version);
 #endif
-            data_alloc->set(dirty_it->second.location >> dsk.block_order, false);
+            data_alloc->set(dirty_it->second.location >> block_order, false);
         }
-        auto used = --journal.used_sectors[dirty_it->second.journal_sector];
+        int used = --journal.used_sectors[dirty_it->second.journal_sector];
 #ifdef BLOCKSTORE_DEBUG
         printf(
             "remove usage of journal offset %08lx by %lx:%lx v%lu (%d refs)\n", dirty_it->second.journal_sector,
@@ -231,9 +232,8 @@ void blockstore_impl_t::erase_dirty(blockstore_dirty_db_t::iterator dirty_start,
         if (used == 0)
         {
             journal.used_sectors.erase(dirty_it->second.journal_sector);
-            flusher->mark_trim_possible();
         }
-        if (dsk.clean_entry_bitmap_size > sizeof(void*))
+        if (clean_entry_bitmap_size > sizeof(void*))
         {
             free(dirty_it->second.bitmap);
             dirty_it->second.bitmap = NULL;

src/blockstore_stable.cpp

@@ -137,7 +137,7 @@ resume_2:
     if (!disable_journal_fsync)
     {
         BS_SUBMIT_GET_SQE(sqe, data);
-        my_uring_prep_fsync(sqe, dsk.journal_fd, IORING_FSYNC_DATASYNC);
+        my_uring_prep_fsync(sqe, journal.fd, IORING_FSYNC_DATASYNC);
         data->iov = { 0 };
         data->callback = [this, op](ring_data_t *data) { handle_write_event(data, op); };
         PRIV(op)->min_flushed_journal_sector = PRIV(op)->max_flushed_journal_sector = 0;
@@ -195,14 +195,14 @@ void blockstore_impl_t::mark_stable(const obj_ver_id & v, bool forget_dirty)
                     }
                     if (!exists)
                     {
-                        inode_space_stats[dirty_it->first.oid.inode] += dsk.data_block_size;
+                        inode_space_stats[dirty_it->first.oid.inode] += data_block_size;
                     }
                 }
                 else if (IS_DELETE(dirty_it->second.state))
                 {
                     auto & sp = inode_space_stats[dirty_it->first.oid.inode];
-                    if (sp > dsk.data_block_size)
-                        sp -= dsk.data_block_size;
+                    if (sp > data_block_size)
+                        sp -= data_block_size;
                     else
                         inode_space_stats.erase(dirty_it->first.oid.inode);
                 }

src/blockstore_sync.cpp

@@ -60,7 +60,7 @@ int blockstore_impl_t::continue_sync(blockstore_op_t *op, bool queue_has_in_prog
         if (!disable_data_fsync)
         {
             BS_SUBMIT_GET_SQE(sqe, data);
-            my_uring_prep_fsync(sqe, dsk.data_fd, IORING_FSYNC_DATASYNC);
+            my_uring_prep_fsync(sqe, data_fd, IORING_FSYNC_DATASYNC);
             data->iov = { 0 };
             data->callback = [this, op](ring_data_t *data) { handle_write_event(data, op); };
             PRIV(op)->min_flushed_journal_sector = PRIV(op)->max_flushed_journal_sector = 0;
@@ -79,7 +79,7 @@ int blockstore_impl_t::continue_sync(blockstore_op_t *op, bool queue_has_in_prog
         // Check space in the journal and journal memory buffers
         blockstore_journal_check_t space_check(this);
         if (!space_check.check_available(op, PRIV(op)->sync_big_writes.size(),
-            sizeof(journal_entry_big_write) + dsk.clean_entry_bitmap_size, JOURNAL_STABILIZE_RESERVATION))
+            sizeof(journal_entry_big_write) + clean_entry_bitmap_size, JOURNAL_STABILIZE_RESERVATION))
         {
             return 0;
         }
@@ -90,7 +90,7 @@ int blockstore_impl_t::continue_sync(blockstore_op_t *op, bool queue_has_in_prog
         int s = 0;
         while (it != PRIV(op)->sync_big_writes.end())
         {
-            if (!journal.entry_fits(sizeof(journal_entry_big_write) + dsk.clean_entry_bitmap_size) &&
+            if (!journal.entry_fits(sizeof(journal_entry_big_write) + clean_entry_bitmap_size) &&
                 journal.sector_info[journal.cur_sector].dirty)
             {
                 prepare_journal_sector_write(journal.cur_sector, op);
@@ -99,7 +99,7 @@ int blockstore_impl_t::continue_sync(blockstore_op_t *op, bool queue_has_in_prog
             auto & dirty_entry = dirty_db.at(*it);
             journal_entry_big_write *je = (journal_entry_big_write*)prefill_single_journal_entry(
                 journal, (dirty_entry.state & BS_ST_INSTANT) ? JE_BIG_WRITE_INSTANT : JE_BIG_WRITE,
-                sizeof(journal_entry_big_write) + dsk.clean_entry_bitmap_size
+                sizeof(journal_entry_big_write) + clean_entry_bitmap_size
             );
             dirty_entry.journal_sector = journal.sector_info[journal.cur_sector].offset;
             journal.used_sectors[journal.sector_info[journal.cur_sector].offset]++;
@@ -115,8 +115,8 @@ int blockstore_impl_t::continue_sync(blockstore_op_t *op, bool queue_has_in_prog
             je->offset = dirty_entry.offset;
             je->len = dirty_entry.len;
             je->location = dirty_entry.location;
-            memcpy((void*)(je+1), (dsk.clean_entry_bitmap_size > sizeof(void*)
-                ? dirty_entry.bitmap : &dirty_entry.bitmap), dsk.clean_entry_bitmap_size);
+            memcpy((void*)(je+1), (clean_entry_bitmap_size > sizeof(void*)
+                ? dirty_entry.bitmap : &dirty_entry.bitmap), clean_entry_bitmap_size);
             je->crc32 = je_crc32((journal_entry*)je);
             journal.crc32_last = je->crc32;
             it++;
@@ -132,7 +132,7 @@ int blockstore_impl_t::continue_sync(blockstore_op_t *op, bool queue_has_in_prog
         if (!disable_journal_fsync)
         {
             BS_SUBMIT_GET_SQE(sqe, data);
-            my_uring_prep_fsync(sqe, dsk.journal_fd, IORING_FSYNC_DATASYNC);
+            my_uring_prep_fsync(sqe, journal.fd, IORING_FSYNC_DATASYNC);
             data->iov = { 0 };
             data->callback = [this, op](ring_data_t *data) { handle_write_event(data, op); };
             PRIV(op)->min_flushed_journal_sector = PRIV(op)->max_flushed_journal_sector = 0;

src/blockstore_write.cpp

@@ -10,9 +10,9 @@ bool blockstore_impl_t::enqueue_write(blockstore_op_t *op)
     bool wait_big = false, wait_del = false;
     void *bmp = NULL;
     uint64_t version = 1;
-    if (!is_del && dsk.clean_entry_bitmap_size > sizeof(void*))
+    if (!is_del && clean_entry_bitmap_size > sizeof(void*))
     {
-        bmp = calloc_or_die(1, dsk.clean_entry_bitmap_size);
+        bmp = calloc_or_die(1, clean_entry_bitmap_size);
     }
     if (dirty_db.size() > 0)
     {
@@ -32,8 +32,8 @@ bool blockstore_impl_t::enqueue_write(blockstore_op_t *op)
                 : ((dirty_it->second.state & BS_ST_WORKFLOW_MASK) == BS_ST_WAIT_BIG);
             if (!is_del && !deleted)
             {
-                if (dsk.clean_entry_bitmap_size > sizeof(void*))
-                    memcpy(bmp, dirty_it->second.bitmap, dsk.clean_entry_bitmap_size);
+                if (clean_entry_bitmap_size > sizeof(void*))
+                    memcpy(bmp, dirty_it->second.bitmap, clean_entry_bitmap_size);
                 else
                     bmp = dirty_it->second.bitmap;
             }
@@ -48,8 +48,8 @@ bool blockstore_impl_t::enqueue_write(blockstore_op_t *op)
             version = clean_it->second.version + 1;
             if (!is_del)
             {
-                void *bmp_ptr = get_clean_entry_bitmap(clean_it->second.location, dsk.clean_entry_bitmap_size);
-                memcpy((dsk.clean_entry_bitmap_size > sizeof(void*) ? bmp : &bmp), bmp_ptr, dsk.clean_entry_bitmap_size);
+                void *bmp_ptr = get_clean_entry_bitmap(clean_it->second.location, clean_entry_bitmap_size);
+                memcpy((clean_entry_bitmap_size > sizeof(void*) ? bmp : &bmp), bmp_ptr, clean_entry_bitmap_size);
             }
         }
         else
@@ -89,18 +89,15 @@ bool blockstore_impl_t::enqueue_write(blockstore_op_t *op)
         else
         {
             // Invalid version requested
-#ifdef BLOCKSTORE_DEBUG
-            printf("Write %lx:%lx v%lu requested, but we already have v%lu\n", op->oid.inode, op->oid.stripe, op->version, version);
-#endif
             op->retval = -EEXIST;
-            if (!is_del && dsk.clean_entry_bitmap_size > sizeof(void*))
+            if (!is_del && clean_entry_bitmap_size > sizeof(void*))
             {
                 free(bmp);
             }
             return false;
         }
     }
-    if (wait_big && !is_del && !deleted && op->len < dsk.data_block_size &&
+    if (wait_big && !is_del && !deleted && op->len < data_block_size &&
         immediate_commit != IMMEDIATE_ALL)
     {
         // Issue an additional sync so that the previous big write can reach the journal
@@ -118,14 +115,14 @@ bool blockstore_impl_t::enqueue_write(blockstore_op_t *op)
     else if (!wait_del)
         printf("Write %lx:%lx v%lu offset=%u len=%u\n", op->oid.inode, op->oid.stripe, op->version, op->offset, op->len);
 #endif
-    // No strict need to add it into dirty_db here except maybe for listings to return
-    // correct data when there are inflight operations in the queue
+    // FIXME No strict need to add it into dirty_db here, it's just left
+    // from the previous implementation where reads waited for writes
     uint32_t state;
     if (is_del)
         state = BS_ST_DELETE | BS_ST_IN_FLIGHT;
     else
     {
-        state = (op->len == dsk.data_block_size || deleted ? BS_ST_BIG_WRITE : BS_ST_SMALL_WRITE);
+        state = (op->len == data_block_size || deleted ? BS_ST_BIG_WRITE : BS_ST_SMALL_WRITE);
         if (state == BS_ST_SMALL_WRITE && throttle_small_writes)
             clock_gettime(CLOCK_REALTIME, &PRIV(op)->tv_begin);
         if (wait_del)
@@ -139,10 +136,10 @@ bool blockstore_impl_t::enqueue_write(blockstore_op_t *op)
         if (op->bitmap)
         {
             // Only allow to overwrite part of the object bitmap respective to the write's offset/len
-            uint8_t *bmp_ptr = (uint8_t*)(dsk.clean_entry_bitmap_size > sizeof(void*) ? bmp : &bmp);
-            uint32_t bit = op->offset/dsk.bitmap_granularity;
-            uint32_t bits_left = op->len/dsk.bitmap_granularity;
-            while (!(bit % 8) && bits_left >= 8)
+            uint8_t *bmp_ptr = (uint8_t*)(clean_entry_bitmap_size > sizeof(void*) ? bmp : &bmp);
+            uint32_t bit = op->offset/bitmap_granularity;
+            uint32_t bits_left = op->len/bitmap_granularity;
+            while (!(bit % 8) && bits_left > 8)
             {
                 // Copy bytes
                 bmp_ptr[bit/8] = ((uint8_t*)op->bitmap)[bit/8];
@@ -178,22 +175,15 @@ void blockstore_impl_t::cancel_all_writes(blockstore_op_t *op, blockstore_dirty_
 {
     while (dirty_it != dirty_db.end() && dirty_it->first.oid == op->oid)
     {
-        if (dsk.clean_entry_bitmap_size > sizeof(void*))
+        if (clean_entry_bitmap_size > sizeof(void*))
             free(dirty_it->second.bitmap);
         dirty_db.erase(dirty_it++);
     }
     bool found = false;
     for (auto other_op: submit_queue)
     {
-        if (!other_op)
-        {
-            // freed operations during submitting are zeroed
-        }
-        else if (other_op == op)
-        {
-            // <op> may be present in queue multiple times due to moving operations in submit_queue
+        if (!found && other_op == op)
             found = true;
-        }
         else if (found && other_op->oid == op->oid &&
             (other_op->opcode == BS_OP_WRITE || other_op->opcode == BS_OP_WRITE_STABLE))
         {
@@ -261,8 +251,7 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
     {
         blockstore_journal_check_t space_check(this);
         if (!space_check.check_available(op, unsynced_big_write_count + 1,
-            sizeof(journal_entry_big_write) + dsk.clean_entry_bitmap_size,
-            (dirty_it->second.state & BS_ST_INSTANT) ? JOURNAL_INSTANT_RESERVATION : JOURNAL_STABILIZE_RESERVATION))
+            sizeof(journal_entry_big_write) + clean_entry_bitmap_size, JOURNAL_STABILIZE_RESERVATION))
         {
             return 0;
         }
@@ -271,28 +260,18 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
         if (loc == UINT64_MAX)
         {
             // no space
+            if (flusher->is_active())
+            {
+                // hope that some space will be available after flush
+                PRIV(op)->wait_for = WAIT_FREE;
+                return 0;
+            }
             cancel_all_writes(op, dirty_it, -ENOSPC);
             return 2;
         }
-        if (inmemory_meta)
-        {
-            // Check once more that metadata entry is zeroed (the reverse means a bug or corruption)
-            uint64_t sector = (loc / (dsk.meta_block_size / dsk.clean_entry_size)) * dsk.meta_block_size;
-            uint64_t pos = (loc % (dsk.meta_block_size / dsk.clean_entry_size));
-            clean_disk_entry *entry = (clean_disk_entry*)((uint8_t*)metadata_buffer + sector + pos*dsk.clean_entry_size);
-            if (entry->oid.inode || entry->oid.stripe || entry->version)
-            {
-                printf(
-                    "Fatal error (metadata corruption or bug): tried to write object %lx:%lx v%lu"
-                    " over a non-zero metadata entry %lu with %lx:%lx v%lu\n", op->oid.inode,
-                    op->oid.stripe, op->version, loc, entry->oid.inode, entry->oid.stripe, entry->version
-                );
-                exit(1);
-            }
-        }
         BS_SUBMIT_GET_SQE(sqe, data);
         write_iodepth++;
-        dirty_it->second.location = loc << dsk.block_order;
+        dirty_it->second.location = loc << block_order;
         dirty_it->second.state = (dirty_it->second.state & ~BS_ST_WORKFLOW_MASK) | BS_ST_SUBMITTED;
 #ifdef BLOCKSTORE_DEBUG
         printf(
@@ -301,9 +280,9 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
         );
 #endif
         data_alloc->set(loc, true);
-        uint64_t stripe_offset = (op->offset % dsk.bitmap_granularity);
-        uint64_t stripe_end = (op->offset + op->len) % dsk.bitmap_granularity;
-        // Zero fill up to dsk.bitmap_granularity
+        uint64_t stripe_offset = (op->offset % bitmap_granularity);
+        uint64_t stripe_end = (op->offset + op->len) % bitmap_granularity;
+        // Zero fill up to bitmap_granularity
         int vcnt = 0;
         if (stripe_offset)
         {
@@ -312,13 +291,13 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
         PRIV(op)->iov_zerofill[vcnt++] = (struct iovec){ op->buf, op->len };
         if (stripe_end)
         {
-            stripe_end = dsk.bitmap_granularity - stripe_end;
+            stripe_end = bitmap_granularity - stripe_end;
             PRIV(op)->iov_zerofill[vcnt++] = (struct iovec){ zero_object, stripe_end };
         }
         data->iov.iov_len = op->len + stripe_offset + stripe_end; // to check it in the callback
         data->callback = [this, op](ring_data_t *data) { handle_write_event(data, op); };
         my_uring_prep_writev(
-            sqe, dsk.data_fd, PRIV(op)->iov_zerofill, vcnt, dsk.data_offset + (loc << dsk.block_order) + op->offset - stripe_offset
+            sqe, data_fd, PRIV(op)->iov_zerofill, vcnt, data_offset + (loc << block_order) + op->offset - stripe_offset
         );
         PRIV(op)->pending_ops = 1;
         PRIV(op)->min_flushed_journal_sector = PRIV(op)->max_flushed_journal_sector = 0;
@@ -340,10 +319,9 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
         blockstore_journal_check_t space_check(this);
         if (unsynced_big_write_count &&
             !space_check.check_available(op, unsynced_big_write_count,
-                sizeof(journal_entry_big_write) + dsk.clean_entry_bitmap_size, 0)
+                sizeof(journal_entry_big_write) + clean_entry_bitmap_size, 0)
             || !space_check.check_available(op, 1,
-                sizeof(journal_entry_small_write) + dsk.clean_entry_bitmap_size,
-                op->len + ((dirty_it->second.state & BS_ST_INSTANT) ? JOURNAL_INSTANT_RESERVATION : JOURNAL_STABILIZE_RESERVATION)))
+                sizeof(journal_entry_small_write) + clean_entry_bitmap_size, op->len + JOURNAL_STABILIZE_RESERVATION))
         {
             return 0;
         }
@@ -351,7 +329,7 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
         BS_SUBMIT_CHECK_SQES(
             // Write current journal sector only if it's dirty and full, or in the immediate_commit mode
             (immediate_commit != IMMEDIATE_NONE ||
-                !journal.entry_fits(sizeof(journal_entry_small_write) + dsk.clean_entry_bitmap_size) ? 1 : 0) +
+                !journal.entry_fits(sizeof(journal_entry_small_write) + clean_entry_bitmap_size) ? 1 : 0) +
             (op->len > 0 ? 1 : 0)
         );
         write_iodepth++;
@@ -359,7 +337,7 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
         auto cb = [this, op](ring_data_t *data) { handle_write_event(data, op); };
         if (immediate_commit == IMMEDIATE_NONE)
         {
-            if (!journal.entry_fits(sizeof(journal_entry_small_write) + dsk.clean_entry_bitmap_size))
+            if (!journal.entry_fits(sizeof(journal_entry_small_write) + clean_entry_bitmap_size))
             {
                 prepare_journal_sector_write(journal.cur_sector, op);
             }
@@ -371,7 +349,7 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
         // Then pre-fill journal entry
         journal_entry_small_write *je = (journal_entry_small_write*)prefill_single_journal_entry(
             journal, op->opcode == BS_OP_WRITE_STABLE ? JE_SMALL_WRITE_INSTANT : JE_SMALL_WRITE,
-            sizeof(journal_entry_small_write) + dsk.clean_entry_bitmap_size
+            sizeof(journal_entry_small_write) + clean_entry_bitmap_size
         );
         dirty_it->second.journal_sector = journal.sector_info[journal.cur_sector].offset;
         journal.used_sectors[journal.sector_info[journal.cur_sector].offset]++;
@@ -383,14 +361,14 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
         );
 #endif
         // Figure out where data will be
-        journal.next_free = (journal.next_free + op->len) <= journal.len ? journal.next_free : dsk.journal_block_size;
+        journal.next_free = (journal.next_free + op->len) <= journal.len ? journal.next_free : journal_block_size;
         je->oid = op->oid;
         je->version = op->version;
         je->offset = op->offset;
         je->len = op->len;
         je->data_offset = journal.next_free;
         je->crc32_data = crc32c(0, op->buf, op->len);
-        memcpy((void*)(je+1), (dsk.clean_entry_bitmap_size > sizeof(void*) ? dirty_it->second.bitmap : &dirty_it->second.bitmap), dsk.clean_entry_bitmap_size);
+        memcpy((void*)(je+1), (clean_entry_bitmap_size > sizeof(void*) ? dirty_it->second.bitmap : &dirty_it->second.bitmap), clean_entry_bitmap_size);
         je->crc32 = je_crc32((journal_entry*)je);
         journal.crc32_last = je->crc32;
         if (immediate_commit != IMMEDIATE_NONE)
@@ -409,7 +387,7 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
             data2->iov = (struct iovec){ op->buf, op->len };
             data2->callback = cb;
             my_uring_prep_writev(
-                sqe2, dsk.journal_fd, &data2->iov, 1, journal.offset + journal.next_free
+                sqe2, journal.fd, &data2->iov, 1, journal.offset + journal.next_free
             );
             PRIV(op)->pending_ops++;
         }
@@ -422,7 +400,7 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
         journal.next_free += op->len;
         if (journal.next_free >= journal.len)
         {
-            journal.next_free = dsk.journal_block_size;
+            journal.next_free = journal_block_size;
         }
         if (!PRIV(op)->pending_ops)
         {
@@ -454,22 +432,15 @@ int blockstore_impl_t::continue_write(blockstore_op_t *op)
 resume_2:
     // Only for the immediate_commit mode: prepare and submit big_write journal entry
     {
+        BS_SUBMIT_CHECK_SQES(1);
         auto dirty_it = dirty_db.find((obj_ver_id){
             .oid = op->oid,
             .version = op->version,
         });
         assert(dirty_it != dirty_db.end());
-        blockstore_journal_check_t space_check(this);
-        if (!space_check.check_available(op, 1,
-            sizeof(journal_entry_big_write) + dsk.clean_entry_bitmap_size,
-            ((dirty_it->second.state & BS_ST_INSTANT) ? JOURNAL_INSTANT_RESERVATION : JOURNAL_STABILIZE_RESERVATION)))
-        {
-            return 0;
-        }
-        BS_SUBMIT_CHECK_SQES(1);
         journal_entry_big_write *je = (journal_entry_big_write*)prefill_single_journal_entry(
             journal, op->opcode == BS_OP_WRITE_STABLE ? JE_BIG_WRITE_INSTANT : JE_BIG_WRITE,
-            sizeof(journal_entry_big_write) + dsk.clean_entry_bitmap_size
+            sizeof(journal_entry_big_write) + clean_entry_bitmap_size
         );
         dirty_it->second.journal_sector = journal.sector_info[journal.cur_sector].offset;
         journal.used_sectors[journal.sector_info[journal.cur_sector].offset]++;
@@ -485,7 +456,7 @@ resume_2:
         je->offset = op->offset;
         je->len = op->len;
         je->location = dirty_it->second.location;
-        memcpy((void*)(je+1), (dsk.clean_entry_bitmap_size > sizeof(void*) ? dirty_it->second.bitmap : &dirty_it->second.bitmap), dsk.clean_entry_bitmap_size);
+        memcpy((void*)(je+1), (clean_entry_bitmap_size > sizeof(void*) ? dirty_it->second.bitmap : &dirty_it->second.bitmap), clean_entry_bitmap_size);
         je->crc32 = je_crc32((journal_entry*)je);
         journal.crc32_last = je->crc32;
         prepare_journal_sector_write(journal.cur_sector, op);
@@ -598,7 +569,10 @@ void blockstore_impl_t::handle_write_event(ring_data_t *data, blockstore_op_t *o
     if (data->res != data->iov.iov_len)
     {
         // FIXME: our state becomes corrupted after a write error. maybe do something better than just die
-        disk_error_abort("data write", data->res, data->iov.iov_len);
+        throw std::runtime_error(
+            "write operation failed ("+std::to_string(data->res)+" != "+std::to_string(data->iov.iov_len)+
+            "). in-memory state is corrupted. AAAAAAAaaaaaaaaa!!!111"
+        );
     }
     PRIV(op)->pending_ops--;
     assert(PRIV(op)->pending_ops >= 0);
@@ -653,14 +627,14 @@ int blockstore_impl_t::dequeue_del(blockstore_op_t *op)
     });
     assert(dirty_it != dirty_db.end());
     blockstore_journal_check_t space_check(this);
-    if (!space_check.check_available(op, 1, sizeof(journal_entry_del), JOURNAL_INSTANT_RESERVATION))
+    if (!space_check.check_available(op, 1, sizeof(journal_entry_del), JOURNAL_STABILIZE_RESERVATION))
     {
         return 0;
     }
     // Write current journal sector only if it's dirty and full, or in the immediate_commit mode
     BS_SUBMIT_CHECK_SQES(
         (immediate_commit != IMMEDIATE_NONE ||
-            (dsk.journal_block_size - journal.in_sector_pos) < sizeof(journal_entry_del) &&
+            (journal_block_size - journal.in_sector_pos) < sizeof(journal_entry_del) &&
             journal.sector_info[journal.cur_sector].dirty) ? 1 : 0
     );
     if (write_iodepth >= max_write_iodepth)
@@ -671,7 +645,7 @@ int blockstore_impl_t::dequeue_del(blockstore_op_t *op)
     // Prepare journal sector write
     if (immediate_commit == IMMEDIATE_NONE)
     {
-        if ((dsk.journal_block_size - journal.in_sector_pos) < sizeof(journal_entry_del) &&
+        if ((journal_block_size - journal.in_sector_pos) < sizeof(journal_entry_del) &&
             journal.sector_info[journal.cur_sector].dirty)
         {
             prepare_journal_sector_write(journal.cur_sector, op);

src/cli.cpp

@@ -12,87 +12,11 @@
 #include "epoll_manager.h"
 #include "cluster_client.h"
 #include "pg_states.h"
-#include "str_util.h"
+#include "base64.h"
 
 static const char *exe_name = NULL;
 
-static const char* help_text =
-    "Vitastor command-line tool\n"
-    "(c) Vitaliy Filippov, 2019+ (VNPL-1.1)\n"
-    "\n"
-    "COMMANDS:\n"
-    "\n"
-    "vitastor-cli status\n"
-    "  Show cluster status\n"
-    "\n"
-    "vitastor-cli df\n"
-    "  Show pool space statistics\n"
-    "\n"
-    "vitastor-cli ls [-l] [-p POOL] [--sort FIELD] [-r] [-n N] [<glob> ...]\n"
-    "  List images (only matching <glob> patterns if passed).\n"
-    "  -p|--pool POOL  Filter images by pool ID or name\n"
-    "  -l|--long       Also report allocated size and I/O statistics\n"
-    "  --del           Also include delete operation statistics\n"
-    "  --sort FIELD    Sort by specified field (name, size, used_size, <read|write|delete>_<iops|bps|lat|queue>)\n"
-    "  -r|--reverse    Sort in descending order\n"
-    "  -n|--count N    Only list first N items\n"
-    "\n"
-    "vitastor-cli create -s|--size <size> [-p|--pool <id|name>] [--parent <parent_name>[@<snapshot>]] <name>\n"
-    "  Create an image. You may use K/M/G/T suffixes for <size>. If --parent is specified,\n"
-    "  a copy-on-write image clone is created. Parent must be a snapshot (readonly image).\n"
-    "  Pool must be specified if there is more than one pool.\n"
-    "\n"
-    "vitastor-cli create --snapshot <snapshot> [-p|--pool <id|name>] <image>\n"
-    "vitastor-cli snap-create [-p|--pool <id|name>] <image>@<snapshot>\n"
-    "  Create a snapshot of image <name>. May be used live if only a single writer is active.\n"
-    "\n"
-    "vitastor-cli modify <name> [--rename <new-name>] [--resize <size>] [--readonly | --readwrite] [-f|--force]\n"
-    "  Rename, resize image or change its readonly status. Images with children can't be made read-write.\n"
-    "  If the new size is smaller than the old size, extra data will be purged.\n"
-    "  You should resize file system in the image, if present, before shrinking it.\n"
-    "  -f|--force  Proceed with shrinking or setting readwrite flag even if the image has children.\n"
-    "\n"
-    "vitastor-cli rm <from> [<to>] [--writers-stopped]\n"
-    "  Remove <from> or all layers between <from> and <to> (<to> must be a child of <from>),\n"
-    "  rebasing all their children accordingly. --writers-stopped allows merging to be a bit\n"
-    "  more effective in case of a single 'slim' read-write child and 'fat' removed parent:\n"
-    "  the child is merged into parent and parent is renamed to child in that case.\n"
-    "  In other cases parent layers are always merged into children.\n"
-    "\n"
-    "vitastor-cli flatten <layer>\n"
-    "  Flatten a layer, i.e. merge data and detach it from parents.\n"
-    "\n"
-    "vitastor-cli rm-data --pool <pool> --inode <inode> [--wait-list] [--min-offset <offset>]\n"
-    "  Remove inode data without changing metadata.\n"
-    "  --wait-list   Retrieve full objects listings before starting to remove objects.\n"
-    "                Requires more memory, but allows to show correct removal progress.\n"
-    "  --min-offset  Purge only data starting with specified offset.\n"
-    "\n"
-    "vitastor-cli merge-data <from> <to> [--target <target>]\n"
-    "  Merge layer data without changing metadata. Merge <from>..<to> to <target>.\n"
-    "  <to> must be a child of <from> and <target> may be one of the layers between\n"
-    "  <from> and <to>, including <from> and <to>.\n"
-    "\n"
-    "vitastor-cli alloc-osd\n"
-    "  Allocate a new OSD number and reserve it by creating empty /osd/stats/<n> key.\n"
-    "\n"
-    "vitastor-cli rm-osd [--force] [--allow-data-loss] [--dry-run] <osd_id> [osd_id...]\n"
-    "  Remove metadata and configuration for specified OSD(s) from etcd.\n"
-    "  Refuses to remove OSDs with data without --force and --allow-data-loss.\n"
-    "  With --dry-run only checks if deletion is possible without data loss and\n"
-    "  redundancy degradation.\n"
-    "\n"
-    "Use vitastor-cli --help <command> for command details or vitastor-cli --help --all for all details.\n"
-    "\n"
-    "GLOBAL OPTIONS:\n"
-    "  --etcd_address <etcd_address>\n"
-    "  --iodepth N         Send N operations in parallel to each OSD when possible (default 32)\n"
-    "  --parallel_osds M   Work with M osds in parallel when possible (default 4)\n"
-    "  --progress 1|0      Report progress (default 1)\n"
-    "  --cas 1|0           Use CAS writes for flatten, merge, rm (default is decide automatically)\n"
-    "  --no-color          Disable colored output\n"
-    "  --json              JSON output\n"
-;
+static void help();
 
 static json11::Json::object parse_args(int narg, const char *args[])
 {
@@ -101,47 +25,42 @@ static json11::Json::object parse_args(int narg, const char *args[])
     cfg["progress"] = "1";
     for (int i = 1; i < narg; i++)
     {
-        if (args[i][0] == '-' && args[i][1] == 'h' && args[i][2] == 0)
+        if (!strcmp(args[i], "-h") || !strcmp(args[i], "--help"))
         {
-            cfg["help"] = "1";
+            help();
         }
-        else if (args[i][0] == '-' && args[i][1] == 'l' && args[i][2] == 0)
+        else if (args[i][0] == '-' && args[i][1] == 'l')
         {
             cfg["long"] = "1";
         }
-        else if (args[i][0] == '-' && args[i][1] == 'n' && args[i][2] == 0)
+        else if (args[i][0] == '-' && args[i][1] == 'n')
         {
             cfg["count"] = args[++i];
         }
-        else if (args[i][0] == '-' && args[i][1] == 'p' && args[i][2] == 0)
+        else if (args[i][0] == '-' && args[i][1] == 'p')
         {
             cfg["pool"] = args[++i];
         }
-        else if (args[i][0] == '-' && args[i][1] == 's' && args[i][2] == 0)
+        else if (args[i][0] == '-' && args[i][1] == 's')
         {
             cfg["size"] = args[++i];
         }
-        else if (args[i][0] == '-' && args[i][1] == 'r' && args[i][2] == 0)
+        else if (args[i][0] == '-' && args[i][1] == 'r')
         {
             cfg["reverse"] = "1";
         }
-        else if (args[i][0] == '-' && args[i][1] == 'f' && args[i][2] == 0)
+        else if (args[i][0] == '-' && args[i][1] == 'f')
         {
             cfg["force"] = "1";
         }
         else if (args[i][0] == '-' && args[i][1] == '-')
         {
             const char *opt = args[i]+2;
-            cfg[opt] = i == narg-1 || !strcmp(opt, "json") ||
-                !strcmp(opt, "wait-list") || !strcmp(opt, "wait_list") ||
-                !strcmp(opt, "long") || !strcmp(opt, "del") ||
-                !strcmp(opt, "no-color") || !strcmp(opt, "no_color") ||
+            cfg[opt] = i == narg-1 || !strcmp(opt, "json") || !strcmp(opt, "wait-list") ||
+                !strcmp(opt, "long") || !strcmp(opt, "del") || !strcmp(opt, "no-color") ||
                 !strcmp(opt, "readonly") || !strcmp(opt, "readwrite") ||
                 !strcmp(opt, "force") || !strcmp(opt, "reverse") ||
-                !strcmp(opt, "allow-data-loss") || !strcmp(opt, "allow_data_loss") ||
-                !strcmp(opt, "dry-run") || !strcmp(opt, "dry_run") ||
-                !strcmp(opt, "help") || !strcmp(opt, "all") ||
-                (!strcmp(opt, "writers-stopped") || !strcmp(opt, "writers_stopped")) && strcmp("1", args[i+1]) != 0
+                !strcmp(opt, "writers-stopped") && strcmp("1", args[i+1]) != 0
                 ? "1" : args[++i];
         }
         else
@@ -157,17 +76,98 @@ static json11::Json::object parse_args(int narg, const char *args[])
             cmd.push_back("rm-data");
         }
     }
-    if (!cmd.size() || cfg["help"].bool_value())
-    {
-        print_help(help_text, "vitastor-cli", cmd.size() ? cmd[0].string_value() : "", cfg["all"].bool_value());
-    }
     cfg["command"] = cmd;
     return cfg;
 }
 
+static void help()
+{
+    printf(
+        "Vitastor command-line tool\n"
+        "(c) Vitaliy Filippov, 2019+ (VNPL-1.1)\n"
+        "\n"
+        "USAGE:\n"
+        "%s status\n"
+        "  Show cluster status\n"
+        "\n"
+        "%s df\n"
+        "  Show pool space statistics\n"
+        "\n"
+        "%s ls [-l] [-p POOL] [--sort FIELD] [-r] [-n N] [<glob> ...]\n"
+        "  List images (only matching <glob> patterns if passed).\n"
+        "  -p|--pool POOL  Filter images by pool ID or name\n"
+        "  -l|--long       Also report allocated size and I/O statistics\n"
+        "  --del           Also include delete operation statistics\n"
+        "  --sort FIELD    Sort by specified field (name, size, used_size, <read|write|delete>_<iops|bps|lat|queue>)\n"
+        "  -r|--reverse    Sort in descending order\n"
+        "  -n|--count N    Only list first N items\n"
+        "\n"
+        "%s create -s|--size <size> [-p|--pool <id|name>] [--parent <parent_name>[@<snapshot>]] <name>\n"
+        "  Create an image. You may use K/M/G/T suffixes for <size>. If --parent is specified,\n"
+        "  a copy-on-write image clone is created. Parent must be a snapshot (readonly image).\n"
+        "  Pool must be specified if there is more than one pool.\n"
+        "\n"
+        "%s create --snapshot <snapshot> [-p|--pool <id|name>] <image>\n"
+        "%s snap-create [-p|--pool <id|name>] <image>@<snapshot>\n"
+        "  Create a snapshot of image <name>. May be used live if only a single writer is active.\n"
+        "\n"
+        "%s modify <name> [--rename <new-name>] [--resize <size>] [--readonly | --readwrite] [-f|--force]\n"
+        "  Rename, resize image or change its readonly status. Images with children can't be made read-write.\n"
+        "  If the new size is smaller than the old size, extra data will be purged.\n"
+        "  You should resize file system in the image, if present, before shrinking it.\n"
+        "  -f|--force  Proceed with shrinking or setting readwrite flag even if the image has children.\n"
+        "\n"
+        "%s rm <from> [<to>] [--writers-stopped]\n"
+        "  Remove <from> or all layers between <from> and <to> (<to> must be a child of <from>),\n"
+        "  rebasing all their children accordingly. --writers-stopped allows merging to be a bit\n"
+        "  more effective in case of a single 'slim' read-write child and 'fat' removed parent:\n"
+        "  the child is merged into parent and parent is renamed to child in that case.\n"
+        "  In other cases parent layers are always merged into children.\n"
+        "\n"
+        "%s flatten <layer>\n"
+        "  Flatten a layer, i.e. merge data and detach it from parents.\n"
+        "\n"
+        "%s rm-data --pool <pool> --inode <inode> [--wait-list] [--min-offset <offset>]\n"
+        "  Remove inode data without changing metadata.\n"
+        "  --wait-list   Retrieve full objects listings before starting to remove objects.\n"
+        "                Requires more memory, but allows to show correct removal progress.\n"
+        "  --min-offset  Purge only data starting with specified offset.\n"
+        "\n"
+        "%s merge-data <from> <to> [--target <target>]\n"
+        "  Merge layer data without changing metadata. Merge <from>..<to> to <target>.\n"
+        "  <to> must be a child of <from> and <target> may be one of the layers between\n"
+        "  <from> and <to>, including <from> and <to>.\n"
+        "\n"
+        "%s alloc-osd\n"
+        "  Allocate a new OSD number and reserve it by creating empty /osd/stats/<n> key.\n"
+        "%s simple-offsets <device>\n"
+        "  Calculate offsets for simple&stupid (no superblock) OSD deployment. Options:\n"
+        "  --object_size 128k       Set blockstore block size\n"
+        "  --bitmap_granularity 4k  Set bitmap granularity\n"
+        "  --journal_size 16M       Set journal size\n"
+        "  --device_block_size 4k   Set device block size\n"
+        "  --journal_offset 0       Set journal offset\n"
+        "  --device_size 0          Set device size\n"
+        "  --format text            Result format: json, options, env, or text\n"
+        "\n"
+        "GLOBAL OPTIONS:\n"
+        "  --etcd_address <etcd_address>\n"
+        "  --iodepth N         Send N operations in parallel to each OSD when possible (default 32)\n"
+        "  --parallel_osds M   Work with M osds in parallel when possible (default 4)\n"
+        "  --progress 1|0      Report progress (default 1)\n"
+        "  --cas 1|0           Use CAS writes for flatten, merge, rm (default is decide automatically)\n"
+        "  --no-color          Disable colored output\n"
+        "  --json              JSON output\n"
+        ,
+        exe_name, exe_name, exe_name, exe_name, exe_name, exe_name, exe_name,
+        exe_name, exe_name, exe_name, exe_name, exe_name, exe_name
+    );
+    exit(0);
+}
+
 static int run(cli_tool_t *p, json11::Json::object cfg)
 {
-    cli_result_t result = {};
+    cli_result_t result;
     p->parse_config(cfg);
     json11::Json::array cmd = cfg["command"].array_items();
     cfg.erase("command");
@@ -235,16 +235,6 @@ static int run(cli_tool_t *p, json11::Json::object cfg)
         // Delete inode data
         action_cb = p->start_rm_data(cfg);
     }
-    else if (cmd[0] == "rm-osd")
-    {
-        // Delete OSD metadata from etcd
-        if (cmd.size() > 1)
-        {
-            cmd.erase(cmd.begin(), cmd.begin()+1);
-            cfg["osd_id"] = cmd;
-        }
-        action_cb = p->start_rm_osd(cfg);
-    }
     else if (cmd[0] == "merge-data")
     {
         // Merge layer data without affecting metadata
@@ -281,6 +271,15 @@ static int run(cli_tool_t *p, json11::Json::object cfg)
         // Allocate a new OSD number
         action_cb = p->start_alloc_osd(cfg);
     }
+    else if (cmd[0] == "simple-offsets")
+    {
+        // Calculate offsets for simple & stupid OSD deployment without superblock
+        if (cmd.size() > 1)
+        {
+            cfg["device"] = cmd[1];
+        }
+        action_cb = p->simple_offsets(cfg);
+    }
     else
     {
         result = { .err = EINVAL, .text = "unknown command: "+cmd[0].string_value() };

src/cli.h

@@ -45,7 +45,7 @@ public:
     cli_result_t etcd_err;
     json11::Json etcd_result;
 
-    void parse_config(json11::Json::object & cfg);
+    void parse_config(json11::Json cfg);
 
     void change_parent(inode_t cur, inode_t new_parent, cli_result_t *result);
     inode_config_t* get_inode_cfg(const std::string & name);
@@ -64,8 +64,8 @@ public:
     std::function<bool(cli_result_t &)> start_merge(json11::Json);
     std::function<bool(cli_result_t &)> start_flatten(json11::Json);
     std::function<bool(cli_result_t &)> start_rm(json11::Json);
-    std::function<bool(cli_result_t &)> start_rm_osd(json11::Json cfg);
     std::function<bool(cli_result_t &)> start_alloc_osd(json11::Json cfg);
+    std::function<bool(cli_result_t &)> simple_offsets(json11::Json cfg);
 
     // Should be called like loop_and_wait(start_status(), <completion callback>)
     void loop_and_wait(std::function<bool(cli_result_t &)> loop_cb, std::function<void(const cli_result_t &)> complete_cb);
@@ -73,8 +73,12 @@ public:
     void etcd_txn(json11::Json txn);
 };
 
+uint64_t parse_size(std::string size_str);
+
 std::string print_table(json11::Json items, json11::Json header, bool use_esc);
 
+std::string format_size(uint64_t size, bool nobytes = false);
+
 std::string format_lat(uint64_t lat);
 
 std::string format_q(double depth);

src/cli_alloc_osd.cpp

@@ -4,7 +4,7 @@
 #include <ctype.h>
 #include "cli.h"
 #include "cluster_client.h"
-#include "str_util.h"
+#include "base64.h"
 
 #include <algorithm>
 

src/cli_common.cpp

@@ -1,7 +1,7 @@
 // Copyright (c) Vitaliy Filippov, 2019+
 // License: VNPL-1.1 (see README.md for details)
 
-#include "str_util.h"
+#include "base64.h"
 #include "cluster_client.h"
 #include "cli.h"
 
@@ -100,20 +100,9 @@ inode_config_t* cli_tool_t::get_inode_cfg(const std::string & name)
     return NULL;
 }
 
-void cli_tool_t::parse_config(json11::Json::object & cfg)
+void cli_tool_t::parse_config(json11::Json cfg)
 {
-    for (auto kv_it = cfg.begin(); kv_it != cfg.end();)
-    {
-        // Translate all options with - to _
-        if (kv_it->first.find("-") != std::string::npos)
-        {
-            cfg[str_replace(kv_it->first, "-", "_")] = kv_it->second;
-            cfg.erase(kv_it++);
-        }
-        else
-            kv_it++;
-    }
-    color = !cfg["no_color"].bool_value();
+    color = !cfg["no-color"].bool_value();
     json_output = cfg["json"].bool_value();
     iodepth = cfg["iodepth"].uint64_value();
     if (!iodepth)
@@ -123,7 +112,7 @@ void cli_tool_t::parse_config(json11::Json::object & cfg)
         parallel_osds = 4;
     log_level = cfg["log_level"].int64_value();
     progress = cfg["progress"].uint64_value() ? true : false;
-    list_first = cfg["wait_list"].uint64_value() ? true : false;
+    list_first = cfg["wait-list"].uint64_value() ? true : false;
 }
 
 struct cli_result_looper_t

src/cli_create.cpp

@@ -4,7 +4,7 @@
 #include <ctype.h>
 #include "cli.h"
 #include "cluster_client.h"
-#include "str_util.h"
+#include "base64.h"
 
 // Create an image, snapshot or clone
 //
@@ -507,6 +507,34 @@ resume_3:
     }
 };
 
+uint64_t parse_size(std::string size_str)
+{
+    if (!size_str.length())
+    {
+        return 0;
+    }
+    uint64_t mul = 1;
+    char type_char = tolower(size_str[size_str.length()-1]);
+    if (type_char == 'k' || type_char == 'm' || type_char == 'g' || type_char == 't')
+    {
+        if (type_char == 'k')
+            mul = (uint64_t)1<<10;
+        else if (type_char == 'm')
+            mul = (uint64_t)1<<20;
+        else if (type_char == 'g')
+            mul = (uint64_t)1<<30;
+        else /*if (type_char == 't')*/
+            mul = (uint64_t)1<<40;
+        size_str = size_str.substr(0, size_str.length()-1);
+    }
+    uint64_t size = json11::Json(size_str).uint64_value() * mul;
+    if (size == 0 && size_str != "0" && (size_str != "" || mul != 1))
+    {
+        return UINT64_MAX;
+    }
+    return size;
+}
+
 std::function<bool(cli_result_t &)> cli_tool_t::start_create(json11::Json cfg)
 {
     auto image_creator = new image_creator_t();
@@ -517,7 +545,7 @@ std::function<bool(cli_result_t &)> cli_tool_t::start_create(json11::Json cfg)
     image_creator->force_size = cfg["force_size"].bool_value();
     if (cfg["image_meta"].is_object())
     {
-        image_creator->new_meta = cfg["image_meta"];
+        image_creator->new_meta = cfg["image-meta"];
     }
     if (cfg["snapshot"].string_value() != "")
     {
@@ -526,9 +554,8 @@ std::function<bool(cli_result_t &)> cli_tool_t::start_create(json11::Json cfg)
     image_creator->new_parent = cfg["parent"].string_value();
     if (cfg["size"].string_value() != "")
     {
-        bool ok;
-        image_creator->size = parse_size(cfg["size"].string_value(), &ok);
-        if (!ok)
+        image_creator->size = parse_size(cfg["size"].string_value());
+        if (image_creator->size == UINT64_MAX)
         {
             return [size = cfg["size"].string_value()](cli_result_t & result)
             {

src/cli_df.cpp

@@ -3,7 +3,7 @@
 
 #include "cli.h"
 #include "cluster_client.h"
-#include "str_util.h"
+#include "base64.h"
 
 // List pools with space statistics
 struct pool_lister_t

src/cli_flatten.cpp

@@ -133,7 +133,7 @@ std::function<bool(cli_result_t &)> cli_tool_t::start_flatten(json11::Json cfg)
     auto flattener = new snap_flattener_t();
     flattener->parent = this;
     flattener->target_name = cfg["image"].string_value();
-    flattener->fsync_interval = cfg["fsync_interval"].uint64_value();
+    flattener->fsync_interval = cfg["fsync-interval"].uint64_value();
     if (!flattener->fsync_interval)
         flattener->fsync_interval = 128;
     if (!cfg["cas"].is_null())

src/cli_ls.cpp

@@ -4,7 +4,7 @@
 #include <algorithm>
 #include "cli.h"
 #include "cluster_client.h"
-#include "str_util.h"
+#include "base64.h"
 
 // List existing images
 //
@@ -446,6 +446,33 @@ std::string print_table(json11::Json items, json11::Json header, bool use_esc)
     return str;
 }
 
+static uint64_t size_thresh[] = { (uint64_t)1024*1024*1024*1024, (uint64_t)1024*1024*1024, (uint64_t)1024*1024, 1024, 0 };
+static uint64_t size_thresh_d[] = { (uint64_t)1000000000000, (uint64_t)1000000000, (uint64_t)1000000, (uint64_t)1000, 0 };
+static const int size_thresh_n = sizeof(size_thresh)/sizeof(size_thresh[0]);
+static const char *size_unit = "TGMKB";
+
+std::string format_size(uint64_t size, bool nobytes)
+{
+    uint64_t *thr = nobytes ? size_thresh_d : size_thresh;
+    char buf[256];
+    for (int i = 0; i < size_thresh_n; i++)
+    {
+        if (size >= thr[i] || i >= size_thresh_n-1)
+        {
+            double value = thr[i] ? (double)size/thr[i] : size;
+            int l = snprintf(buf, sizeof(buf), "%.1f", value);
+            assert(l < sizeof(buf)-2);
+            if (buf[l-1] == '0')
+                l -= 2;
+            buf[l] = i == size_thresh_n-1 && nobytes ? 0 : ' ';
+            buf[l+1] = i == size_thresh_n-1 && nobytes ? 0 : size_unit[i];
+            buf[l+2] = 0;
+            break;
+        }
+    }
+    return std::string(buf);
+}
+
 std::string format_lat(uint64_t lat)
 {
     char buf[256];

src/cli_merge.cpp

@@ -47,7 +47,6 @@ struct snap_merger_t
     int state = 0;
     int lists_todo = 0;
     uint64_t target_block_size = 0;
-    uint32_t target_bitmap_granularity = 0;
     btree::safe_btree_set<uint64_t> merge_offsets;
     btree::safe_btree_set<uint64_t>::iterator oit;
     std::map<inode_t, std::vector<uint64_t>> layer_lists;
@@ -102,7 +101,7 @@ struct snap_merger_t
         std::vector<inode_t> chain_list;
         inode_config_t *cur = to_cfg;
         chain_list.push_back(cur->num);
-        layer_block_size[cur->num] = get_block_size(cur->num, NULL);
+        layer_block_size[cur->num] = get_block_size(cur->num);
         while (cur->parent_id != from_cfg->num &&
             cur->parent_id != to_cfg->num &&
             cur->parent_id != 0)
@@ -125,7 +124,7 @@ struct snap_merger_t
             }
             cur = &it->second;
             chain_list.push_back(cur->num);
-            layer_block_size[cur->num] = get_block_size(cur->num, NULL);
+            layer_block_size[cur->num] = get_block_size(cur->num);
         }
         if (cur->parent_id != from_cfg->num)
         {
@@ -134,7 +133,7 @@ struct snap_merger_t
             return;
         }
         chain_list.push_back(from_cfg->num);
-        layer_block_size[from_cfg->num] = get_block_size(from_cfg->num, NULL);
+        layer_block_size[from_cfg->num] = get_block_size(from_cfg->num);
         int i = chain_list.size()-1;
         for (inode_t item: chain_list)
         {
@@ -205,16 +204,14 @@ struct snap_merger_t
                 use_cas ? " online (with CAS)" : "", INODE_NO_POOL(target), INODE_POOL(target)
             );
         }
-        target_block_size = get_block_size(target, &target_bitmap_granularity);
+        target_block_size = get_block_size(target);
     }
 
-    uint64_t get_block_size(inode_t inode, uint32_t *bitmap_granularity)
+    uint64_t get_block_size(inode_t inode)
     {
         auto & pool_cfg = parent->cli->st_cli.pool_config.at(INODE_POOL(inode));
         uint64_t pg_data_size = (pool_cfg.scheme == POOL_SCHEME_REPLICATED ? 1 : pool_cfg.pg_size-pool_cfg.parity_chunks);
-        if (bitmap_granularity)
-            *bitmap_granularity = pool_cfg.bitmap_granularity;
-        return pool_cfg.data_block_size * pg_data_size;
+        return parent->cli->get_bs_block_size() * pg_data_size;
     }
 
     void continue_merge_reent()
@@ -412,7 +409,7 @@ struct snap_merger_t
             }
             else
             {
-                uint64_t bitmap_bytes = target_block_size/target_bitmap_granularity/8;
+                uint64_t bitmap_bytes = target_block_size/parent->cli->get_bs_bitmap_granularity()/8;
                 int i;
                 for (i = 0; i < bitmap_bytes; i++)
                 {
@@ -472,7 +469,7 @@ struct snap_merger_t
     {
         // Write each non-empty range using an individual operation
         // FIXME: Allow to use single write with "holes" (OSDs don't allow it yet)
-        uint32_t gran = target_bitmap_granularity;
+        uint32_t gran = parent->cli->get_bs_bitmap_granularity();
         uint64_t bitmap_size = target_block_size / gran;
         while (rwo->end < bitmap_size && !rwo->error_code)
         {
@@ -631,8 +628,8 @@ std::function<bool(cli_result_t &)> cli_tool_t::start_merge(json11::Json cfg)
     merger->from_name = cfg["from"].string_value();
     merger->to_name = cfg["to"].string_value();
     merger->target_name = cfg["target"].string_value();
-    merger->delete_source = cfg["delete_source"].string_value() != "";
-    merger->fsync_interval = cfg["fsync_interval"].uint64_value();
+    merger->delete_source = cfg["delete-source"].string_value() != "";
+    merger->fsync_interval = cfg["fsync-interval"].uint64_value();
     if (!merger->fsync_interval)
         merger->fsync_interval = 128;
     if (!cfg["cas"].is_null())

src/cli_modify.cpp

@@ -3,7 +3,7 @@
 
 #include "cli.h"
 #include "cluster_client.h"
-#include "str_util.h"
+#include "base64.h"
 
 // Rename, resize image (and purge extra data on shrink) or change its readonly status
 struct image_changer_t
@@ -236,7 +236,7 @@ std::function<bool(cli_result_t &)> cli_tool_t::start_modify(json11::Json cfg)
     changer->force = cfg["force"].bool_value();
     changer->set_readonly = cfg["readonly"].bool_value();
     changer->set_readwrite = cfg["readwrite"].bool_value();
-    changer->fsync_interval = cfg["fsync_interval"].uint64_value();
+    changer->fsync_interval = cfg["fsync-interval"].uint64_value();
     if (!changer->fsync_interval)
         changer->fsync_interval = 128;
     // FIXME Check that the image doesn't have children when shrinking

src/cli_rm.cpp

@@ -4,7 +4,7 @@
 #include <fcntl.h>
 #include "cli.h"
 #include "cluster_client.h"
-#include "str_util.h"
+#include "base64.h"
 
 // Remove layer(s): similar to merge, but alters metadata and processes multiple merge targets
 //
@@ -639,7 +639,7 @@ std::function<bool(cli_result_t &)> cli_tool_t::start_rm(json11::Json cfg)
     snap_remover->parent = this;
     snap_remover->from_name = cfg["from"].string_value();
     snap_remover->to_name = cfg["to"].string_value();
-    snap_remover->fsync_interval = cfg["fsync_interval"].uint64_value();
+    snap_remover->fsync_interval = cfg["fsync-interval"].uint64_value();
     if (!snap_remover->fsync_interval)
         snap_remover->fsync_interval = 128;
     if (!cfg["cas"].is_null())

src/cli_rm_data.cpp

@@ -218,7 +218,7 @@ std::function<bool(cli_result_t &)> cli_tool_t::start_rm_data(json11::Json cfg)
         remover->inode = (remover->inode & (((uint64_t)1 << (64-POOL_ID_BITS)) - 1)) | (((uint64_t)remover->pool_id) << (64-POOL_ID_BITS));
     }
     remover->pool_id = INODE_POOL(remover->inode);
-    remover->min_offset = cfg["min_offset"].uint64_value();
+    remover->min_offset = cfg["min-offset"].uint64_value();
     return [remover](cli_result_t & result)
     {
         remover->loop();

src/cli_rm_osd.cpp

@@ -1,491 +0,0 @@
-// Copyright (c) Vitaliy Filippov, 2019+
-// License: VNPL-1.1 (see README.md for details)
-
-#include <ctype.h>
-#include "cli.h"
-#include "cluster_client.h"
-#include "str_util.h"
-#include "epoll_manager.h"
-
-#include <algorithm>
-
-// Delete OSD metadata from etcd
-struct rm_osd_t
-{
-    cli_tool_t *parent;
-
-    bool dry_run, force_warning, force_dataloss;
-    uint64_t etcd_tx_retry_ms = 500;
-    uint64_t etcd_tx_retries = 10000;
-    std::vector<uint64_t> osd_ids;
-
-    int state = 0;
-    cli_result_t result;
-
-    std::set<uint64_t> to_remove;
-    std::set<uint64_t> to_restart;
-    json11::Json::array pool_effects;
-    json11::Json::array history_updates, history_checks;
-    json11::Json new_pgs, new_clean_pgs;
-    uint64_t new_pgs_mod_rev, new_clean_pgs_mod_rev;
-    uint64_t cur_retry = 0;
-    uint64_t retry_wait = 0;
-    bool is_warning, is_dataloss;
-
-    bool is_done()
-    {
-        return state == 100;
-    }
-
-    void loop()
-    {
-        if (state == 1)
-            goto resume_1;
-        else if (state == 2)
-            goto resume_2;
-        else if (state == 3)
-            goto resume_3;
-        else if (state == 4)
-            goto resume_4;
-        if (!osd_ids.size())
-        {
-            result = (cli_result_t){ .err = EINVAL, .text = "OSD numbers are not specified" };
-            state = 100;
-            return;
-        }
-        for (auto osd_id: osd_ids)
-        {
-            if (!osd_id)
-            {
-                result = (cli_result_t){ .err = EINVAL, .text = "OSD number can't be zero" };
-                state = 100;
-                return;
-            }
-            to_remove.insert(osd_id);
-        }
-        // Check if OSDs are still used in data distribution
-        is_warning = is_dataloss = false;
-        for (auto & pp: parent->cli->st_cli.pool_config)
-        {
-            // Will OSD deletion make pool incomplete / down / degraded?
-            bool pool_incomplete = false, pool_down = false, pool_degraded = false;
-            bool hist_incomplete = false, hist_degraded = false;
-            auto & pool_cfg = pp.second;
-            uint64_t pg_data_size = (pool_cfg.scheme == POOL_SCHEME_REPLICATED ? 1 : pool_cfg.pg_size-pool_cfg.parity_chunks);
-            for (auto & pgp: pool_cfg.pg_config)
-            {
-                auto & pg_cfg = pgp.second;
-                int pg_cursize = 0, pg_rm = 0;
-                for (auto pg_osd: pg_cfg.target_set)
-                {
-                    if (pg_osd != 0)
-                    {
-                        pg_cursize++;
-                        if (to_remove.find(pg_osd) != to_remove.end())
-                            pg_rm++;
-                    }
-                }
-                for (auto & hist_item: pg_cfg.target_history)
-                {
-                    int hist_size = 0, hist_rm = 0;
-                    for (auto & old_osd: hist_item)
-                    {
-                        if (old_osd != 0)
-                        {
-                            hist_size++;
-                            if (to_remove.find(old_osd) != to_remove.end())
-                                hist_rm++;
-                        }
-                    }
-                    if (hist_rm > 0)
-                    {
-                        hist_degraded = true;
-                        if (hist_size-hist_rm == 0)
-                            pool_incomplete = true;
-                        else if (hist_size-hist_rm < pg_data_size)
-                            hist_incomplete = true;
-                    }
-                }
-                if (pg_rm > 0)
-                {
-                    pool_degraded = true;
-                    if (pg_cursize-pg_rm < pg_data_size)
-                        pool_incomplete = true;
-                    else if (pg_cursize-pg_rm < pool_cfg.pg_minsize)
-                        pool_down = true;
-                }
-            }
-            if (pool_incomplete || pool_down || pool_degraded || hist_incomplete || hist_degraded)
-            {
-                pool_effects.push_back(json11::Json::object {
-                    { "pool_id", (uint64_t)pool_cfg.id },
-                    { "pool_name", pool_cfg.name },
-                    { "effect", (pool_incomplete
-                        ? "incomplete"
-                        : (hist_incomplete
-                            ? "has_incomplete"
-                            : (pool_down
-                                ? "offline"
-                                : (pool_degraded
-                                    ? "degraded"
-                                    : (hist_degraded ? "has_degraded" : "?")
-                                )
-                            )
-                        )
-                    ) },
-                });
-                is_warning = true;
-                if (pool_incomplete || hist_incomplete)
-                    is_dataloss = true;
-            }
-        }
-        result.data = json11::Json::object {
-            { "osd_ids", osd_ids },
-            { "pool_errors", pool_effects },
-        };
-        if (is_dataloss || is_warning || dry_run)
-        {
-            std::string error;
-            for (auto & e: pool_effects)
-            {
-                error += "Pool "+e["pool_name"].string_value()+" (ID "+e["pool_id"].as_string()+") will have "+(
-                    e["effect"] == "has_incomplete"
-                        ? std::string("INCOMPLETE objects (DATA LOSS)")
-                        : (e["effect"] == "incomplete"
-                            ? std::string("INCOMPLETE PGs (DATA LOSS)")
-                            : (e["effect"] == "has_degraded"
-                                ? std::string("DEGRADED objects")
-                                : strtoupper(e["effect"].string_value())+" PGs"))
-                )+" after deleting OSD(s).\n";
-            }
-            if (is_dataloss && !force_dataloss && !dry_run)
-                error += "OSDs not deleted. Please move data to other OSDs or bypass this check with --allow-data-loss if you know what you are doing.\n";
-            else if (is_warning && !force_warning && !dry_run)
-                error += "OSDs not deleted. Please move data to other OSDs or bypass this check with --force if you know what you are doing.\n";
-            else if (!is_dataloss && !is_warning && dry_run)
-                error += "OSDs can be deleted without data loss.\n";
-            result.text = error;
-            if (dry_run || is_dataloss && !force_dataloss || is_warning && !force_warning)
-            {
-                result.err = is_dataloss && !force_dataloss || is_warning && !force_warning ? EBUSY : 0;
-                state = 100;
-                return;
-            }
-        }
-        parent->etcd_txn(json11::Json::object { { "success", json11::Json::array {
-            json11::Json::object {
-                { "request_range", json11::Json::object {
-                    { "key", base64_encode(
-                        parent->cli->st_cli.etcd_prefix+"/config/pgs"
-                    ) },
-                } },
-            },
-            json11::Json::object {
-                { "request_range", json11::Json::object {
-                    { "key", base64_encode(
-                        parent->cli->st_cli.etcd_prefix+"/history/last_clean_pgs"
-                    ) },
-                } },
-            },
-        } } });
-    resume_4:
-        state = 4;
-        if (parent->waiting > 0)
-            return;
-        if (parent->etcd_err.err)
-        {
-            result = parent->etcd_err;
-            state = 100;
-            return;
-        }
-        {
-            auto kv = parent->cli->st_cli.parse_etcd_kv(parent->etcd_result["responses"][0]["response_range"]["kvs"][0]);
-            new_pgs = remove_osds_from_pgs(kv);
-            new_pgs_mod_rev = kv.mod_revision;
-            kv = parent->cli->st_cli.parse_etcd_kv(parent->etcd_result["responses"][1]["response_range"]["kvs"][0]);
-            new_clean_pgs = remove_osds_from_pgs(kv);
-            new_clean_pgs_mod_rev = kv.mod_revision;
-        }
-        // Remove keys from etcd
-        {
-            json11::Json::array rm_items, rm_checks;
-            for (auto osd_id: osd_ids)
-            {
-                rm_items.push_back("/config/osd/"+std::to_string(osd_id));
-                rm_items.push_back("/osd/stats/"+std::to_string(osd_id));
-                rm_items.push_back("/osd/state/"+std::to_string(osd_id));
-                rm_items.push_back("/osd/inodestats/"+std::to_string(osd_id));
-                rm_items.push_back("/osd/space/"+std::to_string(osd_id));
-            }
-            for (int i = 0; i < rm_items.size(); i++)
-            {
-                rm_items[i] = json11::Json::object {
-                    { "request_delete_range", json11::Json::object {
-                        { "key", base64_encode(
-                            parent->cli->st_cli.etcd_prefix+rm_items[i].string_value()
-                        ) },
-                    } },
-                };
-            }
-            if (!new_pgs.is_null())
-            {
-                auto pgs_key = base64_encode(parent->cli->st_cli.etcd_prefix+"/config/pgs");
-                rm_items.push_back(json11::Json::object {
-                    { "request_put", json11::Json::object {
-                        { "key", pgs_key },
-                        { "value", base64_encode(new_pgs.dump()) },
-                    } },
-                });
-                rm_checks.push_back(json11::Json::object {
-                    { "target", "MOD" },
-                    { "key", pgs_key },
-                    { "result", "LESS" },
-                    { "mod_revision", new_pgs_mod_rev+1 },
-                });
-            }
-            if (!new_clean_pgs.is_null())
-            {
-                auto pgs_key = base64_encode(parent->cli->st_cli.etcd_prefix+"/history/last_clean_pgs");
-                rm_items.push_back(json11::Json::object {
-                    { "request_put", json11::Json::object {
-                        { "key", pgs_key },
-                        { "value", base64_encode(new_clean_pgs.dump()) },
-                    } },
-                });
-                rm_checks.push_back(json11::Json::object {
-                    { "target", "MOD" },
-                    { "key", pgs_key },
-                    { "result", "LESS" },
-                    { "mod_revision", new_clean_pgs_mod_rev+1 },
-                });
-            }
-            parent->etcd_txn(json11::Json::object { { "success", rm_items }, { "checks", rm_checks } });
-        }
-    resume_1:
-        state = 1;
-        if (parent->waiting > 0)
-            return;
-        if (parent->etcd_err.err)
-        {
-            result = parent->etcd_err;
-            state = 100;
-            return;
-        }
-        // Remove old OSD from PG all_peers to prevent left_on_dead and from
-        // target_history to prevent INCOMPLETE if --allow-data-loss is specified
-        for (auto & rsp: parent->etcd_result["responses"].array_items())
-        {
-            if (rsp["response_delete_range"]["deleted"].uint64_value() > 0)
-            {
-                // Wait for mon_change_timeout before updating PG history, or the monitor's change will likely interfere with ours
-                retry_wait = parent->cli->merged_config["mon_change_timeout"].uint64_value();
-                if (!retry_wait)
-                    retry_wait = 1000;
-                retry_wait += etcd_tx_retry_ms;
-            }
-        }
-        while (1)
-        {
-    resume_2:
-            if (!remove_osds_from_history(2))
-                return;
-    resume_3:
-            state = 3;
-            if (parent->waiting > 0)
-                return;
-            if (parent->etcd_err.err)
-            {
-                result = parent->etcd_err;
-                state = 100;
-                return;
-            }
-            if (parent->etcd_result["succeeded"].bool_value())
-                break;
-            if ((++cur_retry) >= etcd_tx_retries)
-            {
-                result.err = EAGAIN;
-                result.text += "Failed to remove OSDs from PG history due to update conflicts."
-                    " Some PGs may remain left_on_dead or incomplete. Please retry later\n";
-                state = 100;
-                return;
-            }
-            retry_wait = etcd_tx_retry_ms;
-        }
-        std::string ids = "";
-        for (auto osd_id: osd_ids)
-        {
-            ids += (ids.size() ? ", " : "")+std::to_string(osd_id);
-        }
-        ids = (osd_ids.size() > 1 ? "OSDs " : "OSD ")+ids+(osd_ids.size() > 1 ? " are" : " is")+" removed from etcd";
-        state = 100;
-        result.text = (result.text != "" ? ids+"\n"+result.text : ids);
-        result.err = 0;
-    }
-
-    json11::Json remove_osds_from_pgs(const etcd_kv_t & kv)
-    {
-        if (kv.value.is_null())
-        {
-            return kv.value;
-        }
-        json11::Json::object new_pgs;
-        for (auto & pp: kv.value["items"].object_items())
-        {
-            if (pp.second.is_object())
-            {
-                json11::Json::object new_pool;
-                for (auto & pgp: pp.second.object_items())
-                {
-                    json11::Json::array osd_set;
-                    for (auto & osd_json: pgp.second["osd_set"].array_items())
-                    {
-                        uint64_t osd_num = osd_json.uint64_value();
-                        osd_set.push_back(osd_num == 0 || to_remove.find(osd_num) != to_remove.end() ? 0 : osd_num);
-                    }
-                    json11::Json::object new_pg = pgp.second.object_items();
-                    new_pg["osd_set"] = osd_set;
-                    new_pool[pgp.first] = new_pg;
-                }
-                new_pgs[pp.first] = new_pool;
-            }
-            else
-                new_pgs[pp.first] = pp.second;
-        }
-        auto res = kv.value.object_items();
-        res["items"] = new_pgs;
-        return res;
-    }
-
-    bool remove_osds_from_history(int base_state)
-    {
-        if (state == base_state+0)
-            goto resume_0;
-        history_updates.clear();
-        history_checks.clear();
-        for (auto & pp: parent->cli->st_cli.pool_config)
-        {
-            bool update_pg_history = false;
-            auto & pool_cfg = pp.second;
-            for (auto & pgp: pool_cfg.pg_config)
-            {
-                auto pg_num = pgp.first;
-                auto & pg_cfg = pgp.second;
-                for (int i = 0; i < pg_cfg.all_peers.size(); i++)
-                {
-                    if (to_remove.find(pg_cfg.all_peers[i]) != to_remove.end())
-                    {
-                        update_pg_history = true;
-                        pg_cfg.all_peers.erase(pg_cfg.all_peers.begin()+i, pg_cfg.all_peers.begin()+i+1);
-                        i--;
-                    }
-                }
-                for (int i = 0; i < pg_cfg.target_history.size(); i++)
-                {
-                    int hist_size = 0, hist_rm = 0;
-                    for (auto & old_osd: pg_cfg.target_history[i])
-                    {
-                        if (old_osd != 0)
-                        {
-                            hist_size++;
-                            if (to_remove.find(old_osd) != to_remove.end())
-                            {
-                                hist_rm++;
-                                old_osd = 0;
-                            }
-                        }
-                    }
-                    if (hist_rm > 0)
-                    {
-                        if (hist_size-hist_rm == 0)
-                        {
-                            pg_cfg.target_history.erase(pg_cfg.target_history.begin()+i, pg_cfg.target_history.begin()+i+1);
-                            i--;
-                        }
-                        update_pg_history = true;
-                    }
-                }
-                if (update_pg_history)
-                {
-                    std::string history_key = base64_encode(
-                        parent->cli->st_cli.etcd_prefix+"/pg/history/"+
-                        std::to_string(pool_cfg.id)+"/"+std::to_string(pg_num)
-                    );
-                    history_updates.push_back(json11::Json::object {
-                        { "request_put", json11::Json::object {
-                            { "key", history_key },
-                            { "value", base64_encode(json11::Json(json11::Json::object {
-                                { "epoch", pg_cfg.epoch },
-                                { "all_peers", pg_cfg.all_peers },
-                                { "osd_sets", pg_cfg.target_history },
-                            }).dump()) },
-                        } },
-                    });
-                    history_checks.push_back(json11::Json::object {
-                        { "target", "MOD" },
-                        { "key", history_key },
-                        { "result", "LESS" },
-                        { "mod_revision", parent->cli->st_cli.etcd_watch_revision+1 },
-                    });
-                }
-            }
-        }
-        if (history_updates.size())
-        {
-            if (retry_wait)
-            {
-                parent->waiting++;
-                parent->epmgr->tfd->set_timer(retry_wait, false, [this](int timer_id)
-                {
-                    parent->waiting--;
-                    parent->ringloop->wakeup();
-                });
-    resume_0:
-                state = base_state+0;
-                if (parent->waiting > 0)
-                    return false;
-            }
-            parent->etcd_txn(json11::Json::object {
-                { "success", history_updates },
-                { "compare", history_checks },
-            });
-        }
-        else
-            parent->etcd_result = json11::Json::object{ { "succeeded", true } };
-        return true;
-    }
-};
-
-std::function<bool(cli_result_t &)> cli_tool_t::start_rm_osd(json11::Json cfg)
-{
-    auto rm_osd = new rm_osd_t();
-    rm_osd->parent = this;
-    rm_osd->dry_run = cfg["dry_run"].bool_value();
-    rm_osd->force_dataloss = cfg["allow_data_loss"].bool_value();
-    rm_osd->force_warning = rm_osd->force_dataloss || cfg["force"].bool_value();
-    if (!cfg["etcd_tx_retries"].is_null())
-        rm_osd->etcd_tx_retries = cfg["etcd_tx_retries"].uint64_value();
-    if (!cfg["etcd_tx_retry_ms"].is_null())
-    {
-        rm_osd->etcd_tx_retry_ms = cfg["etcd_tx_retry_ms"].uint64_value();
-        if (rm_osd->etcd_tx_retry_ms < 100)
-            rm_osd->etcd_tx_retry_ms = 100;
-    }
-    if (cfg["osd_id"].is_number() || cfg["osd_id"].is_string())
-        rm_osd->osd_ids.push_back(cfg["osd_id"].uint64_value());
-    else
-    {
-        for (auto & id: cfg["osd_id"].array_items())
-            rm_osd->osd_ids.push_back(id.uint64_value());
-    }
-    return [rm_osd](cli_result_t & result)
-    {
-        rm_osd->loop();
-        if (rm_osd->is_done())
-        {
-            result = rm_osd->result;
-            delete rm_osd;
-            return true;
-        }
-        return false;
-    };
-}

src/cli_simple_offsets.cpp

@@ -5,14 +5,13 @@
 #include <sys/ioctl.h>
 #include <ctype.h>
 #include <unistd.h>
+#include "cli.h"
+#include "cluster_client.h"
+#include "base64.h"
 #include <sys/stat.h>
 
-#include "json11/json11.hpp"
-#include "str_util.h"
-#include "blockstore.h"
-
 // Calculate offsets for a block device and print OSD command line parameters
-void disk_tool_simple_offsets(json11::Json cfg, bool json_output)
+std::function<bool(cli_result_t &)> cli_tool_t::simple_offsets(json11::Json cfg)
 {
     std::string device = cfg["device"].string_value();
     if (device == "")
@@ -30,7 +29,7 @@ void disk_tool_simple_offsets(json11::Json cfg, bool json_output)
     if (json_output)
         format = "json";
     if (!object_size)
-        object_size = 1 << DEFAULT_DATA_BLOCK_ORDER;
+        object_size = DEFAULT_BLOCK_SIZE;
     if (!bitmap_granularity)
         bitmap_granularity = DEFAULT_BITMAP_GRANULARITY;
     if (!journal_size)
@@ -146,4 +145,5 @@ void disk_tool_simple_offsets(json11::Json cfg, bool json_output)
             device.c_str(), journal_offset, meta_offset, data_offset
         );
     }
+    return NULL;
 }

src/cli_status.cpp

@@ -3,12 +3,10 @@
 
 #include "cli.h"
 #include "cluster_client.h"
-#include "str_util.h"
+#include "base64.h"
 #include "pg_states.h"
 #include "http_client.h"
 
-static const char *obj_states[] = { "clean", "misplaced", "degraded", "incomplete" };
-
 // Print cluster status:
 // etcd, mon, osd states
 // raw/used space, object states, pool states, pg states
@@ -198,57 +196,21 @@ resume_2:
             }
             pgs_by_state_str += std::to_string(kv.second)+" "+kv.first;
         }
+        uint64_t object_size = parent->cli->get_bs_block_size();
+        std::string more_states;
+        uint64_t obj_n;
+        obj_n = agg_stats["object_counts"]["misplaced"].uint64_value();
+        if (obj_n > 0)
+            more_states += ", "+format_size(obj_n*object_size)+" misplaced";
+        obj_n = agg_stats["object_counts"]["degraded"].uint64_value();
+        if (obj_n > 0)
+            more_states += ", "+format_size(obj_n*object_size)+" degraded";
+        obj_n = agg_stats["object_counts"]["incomplete"].uint64_value();
+        if (obj_n > 0)
+            more_states += ", "+format_size(obj_n*object_size)+" incomplete";
         bool readonly = json_is_true(parent->cli->merged_config["readonly"]);
         bool no_recovery = json_is_true(parent->cli->merged_config["no_recovery"]);
         bool no_rebalance = json_is_true(parent->cli->merged_config["no_rebalance"]);
-        if (parent->json_output)
-        {
-            // JSON output
-            auto json_status = json11::Json::object {
-                { "etcd_alive", etcd_alive },
-                { "etcd_count", (uint64_t)etcd_states.size() },
-                { "etcd_db_size", etcd_db_size },
-                { "mon_count", mon_count },
-                { "mon_master", mon_master },
-                { "osd_up", osd_up },
-                { "osd_count", osd_count },
-                { "total_raw", total_raw },
-                { "free_raw", free_raw },
-                { "down_raw", down_raw },
-                { "free_down_raw", free_down_raw },
-                { "readonly", readonly },
-                { "no_recovery", no_recovery },
-                { "no_rebalance", no_rebalance },
-                { "pool_count", pool_count },
-                { "active_pool_count", pools_active },
-                { "pg_states", pgs_by_state },
-                { "op_stats", agg_stats["op_stats"] },
-                { "recovery_stats", agg_stats["recovery_stats"] },
-                { "object_counts", agg_stats["object_counts"] },
-            };
-            for (int i = 0; i < sizeof(obj_states)/sizeof(obj_states[0]); i++)
-            {
-                std::string str(obj_states[i]);
-                uint64_t obj_n = agg_stats["object_bytes"][str].uint64_value();
-                if (!obj_n)
-                    obj_n = agg_stats["object_counts"][str].uint64_value() * parent->cli->st_cli.global_block_size;
-                json_status[str+"_data"] = obj_n;
-            }
-            printf("%s\n", json11::Json(json_status).dump().c_str());
-            state = 100;
-            return;
-        }
-        std::string more_states;
-        for (int i = 0; i < sizeof(obj_states)/sizeof(obj_states[0]); i++)
-        {
-            std::string str(obj_states[i]);
-            uint64_t obj_n = agg_stats["object_bytes"][str].uint64_value();
-            if (!obj_n)
-                obj_n = agg_stats["object_counts"][str].uint64_value() * parent->cli->st_cli.global_block_size;
-            if (!i || obj_n > 0)
-                more_states += format_size(obj_n)+" "+str+", ";
-        }
-        more_states.resize(more_states.size()-2);
         std::string recovery_io;
         {
             uint64_t deg_bps = agg_stats["recovery_stats"]["degraded"]["bps"].uint64_value();
@@ -270,6 +232,38 @@ resume_2:
             else if (no_rebalance)
                 recovery_io += "    rebalance: disabled\n";
         }
+        if (parent->json_output)
+        {
+            // JSON output
+            printf("%s\n", json11::Json(json11::Json::object {
+                { "etcd_alive", etcd_alive },
+                { "etcd_count", (uint64_t)etcd_states.size() },
+                { "etcd_db_size", etcd_db_size },
+                { "mon_count", mon_count },
+                { "mon_master", mon_master },
+                { "osd_up", osd_up },
+                { "osd_count", osd_count },
+                { "total_raw", total_raw },
+                { "free_raw", free_raw },
+                { "down_raw", down_raw },
+                { "free_down_raw", free_down_raw },
+                { "readonly", readonly },
+                { "no_recovery", no_recovery },
+                { "no_rebalance", no_rebalance },
+                { "clean_data", agg_stats["object_counts"]["clean"].uint64_value() * object_size },
+                { "misplaced_data", agg_stats["object_counts"]["misplaced"].uint64_value() * object_size },
+                { "degraded_data", agg_stats["object_counts"]["degraded"].uint64_value() * object_size },
+                { "incomplete_data", agg_stats["object_counts"]["incomplete"].uint64_value() * object_size },
+                { "pool_count", pool_count },
+                { "active_pool_count", pools_active },
+                { "pg_states", pgs_by_state },
+                { "op_stats", agg_stats["op_stats"] },
+                { "recovery_stats", agg_stats["recovery_stats"] },
+                { "object_counts", agg_stats["object_counts"] },
+            }).dump().c_str());
+            state = 100;
+            return;
+        }
         printf(
             "  cluster:\n"
             "    etcd: %d / %ld up, %s database size\n"
@@ -278,7 +272,7 @@ resume_2:
             "  \n"
             "  data:\n"
             "    raw:   %s used, %s / %s available%s\n"
-            "    state: %s\n"
+            "    state: %s clean%s\n"
             "    pools: %d / %d active\n"
             "    pgs:   %s\n"
             "  \n"
@@ -292,7 +286,7 @@ resume_2:
             format_size(free_raw-free_down_raw).c_str(),
             format_size(total_raw-down_raw).c_str(),
             (down_raw > 0 ? (", "+format_size(down_raw)+" down").c_str() : ""),
-            more_states.c_str(),
+            format_size(agg_stats["object_counts"]["clean"].uint64_value() * object_size).c_str(), more_states.c_str(),
             pools_active, pool_count,
             pgs_by_state_str.c_str(),
             readonly ? " (read-only mode)" : "",

src/cluster_client.cpp

@@ -14,7 +14,6 @@
 #define CACHE_FLUSHING 2
 #define CACHE_REPEATING 3
 #define OP_FLUSH_BUFFER 0x02
-#define OP_IMMEDIATE_COMMIT 0x04
 
 cluster_client_t::cluster_client_t(ring_loop_t *ringloop, timerfd_manager_t *tfd, json11::Json & config)
 {
@@ -128,26 +127,26 @@ void cluster_client_t::calc_wait(cluster_op_t *op)
                 op->prev_wait++;
             }
         }
-        if (!op->prev_wait)
+        if (!op->prev_wait && pgs_loaded)
             continue_rw(op);
     }
     else if (op->opcode == OSD_OP_SYNC)
     {
         for (auto prev = op->prev; prev; prev = prev->prev)
         {
-            if (prev->opcode == OSD_OP_SYNC || prev->opcode == OSD_OP_WRITE && !(prev->flags & OP_IMMEDIATE_COMMIT))
+            if (prev->opcode == OSD_OP_SYNC || prev->opcode == OSD_OP_WRITE)
             {
                 op->prev_wait++;
             }
         }
-        if (!op->prev_wait)
+        if (!op->prev_wait && pgs_loaded)
             continue_sync(op);
     }
     else /* if (op->opcode == OSD_OP_READ || op->opcode == OSD_OP_READ_BITMAP) */
     {
         for (auto prev = op_queue_head; prev && prev != op; prev = prev->next)
         {
-            if (prev->opcode == OSD_OP_WRITE && (prev->flags & OP_FLUSH_BUFFER))
+            if (prev->opcode == OSD_OP_WRITE && prev->flags & OP_FLUSH_BUFFER)
             {
                 op->prev_wait++;
             }
@@ -157,7 +156,7 @@ void cluster_client_t::calc_wait(cluster_op_t *op)
                 break;
             }
         }
-        if (!op->prev_wait)
+        if (!op->prev_wait && pgs_loaded)
             continue_rw(op);
     }
 }
@@ -169,7 +168,7 @@ void cluster_client_t::inc_wait(uint64_t opcode, uint64_t flags, cluster_op_t *n
         while (next)
         {
             auto n2 = next->next;
-            if (next->opcode == OSD_OP_SYNC && !(flags & OP_IMMEDIATE_COMMIT) ||
+            if (next->opcode == OSD_OP_SYNC ||
                 next->opcode == OSD_OP_WRITE && (flags & OP_FLUSH_BUFFER) && !(next->flags & OP_FLUSH_BUFFER) ||
                 (next->opcode == OSD_OP_READ || next->opcode == OSD_OP_READ_BITMAP) && (flags & OP_FLUSH_BUFFER))
             {
@@ -221,11 +220,9 @@ void cluster_client_t::erase_op(cluster_op_t *op)
     if (op_queue_tail == op)
         op_queue_tail = op->prev;
     op->next = op->prev = NULL;
-    if (!(flags & OP_IMMEDIATE_COMMIT))
-        inc_wait(opcode, flags, next, -1);
-    // Call callback at the end to avoid inconsistencies in prev_wait
-    // if the callback adds more operations itself
     std::function<void(cluster_op_t*)>(op->callback)(op);
+    if (!immediate_commit)
+        inc_wait(opcode, flags, next, -1);
 }
 
 void cluster_client_t::continue_ops(bool up_retry)
@@ -265,6 +262,21 @@ restart:
     continuing_ops = 0;
 }
 
+static uint32_t is_power_of_two(uint64_t value)
+{
+    uint32_t l = 0;
+    while (value > 1)
+    {
+        if (value & 1)
+        {
+            return 64;
+        }
+        value = value >> 1;
+        l++;
+    }
+    return l;
+}
+
 void cluster_client_t::on_load_config_hook(json11::Json::object & config)
 {
     this->merged_config = config;
@@ -272,6 +284,24 @@ void cluster_client_t::on_load_config_hook(json11::Json::object & config)
     {
         this->merged_config[kv.first] = kv.second;
     }
+    bs_block_size = config["block_size"].uint64_value();
+    bs_bitmap_granularity = config["bitmap_granularity"].uint64_value();
+    if (!bs_block_size)
+    {
+        bs_block_size = DEFAULT_BLOCK_SIZE;
+    }
+    if (!bs_bitmap_granularity)
+    {
+        bs_bitmap_granularity = DEFAULT_BITMAP_GRANULARITY;
+    }
+    bs_bitmap_size = bs_block_size / bs_bitmap_granularity / 8;
+    uint32_t block_order;
+    if ((block_order = is_power_of_two(bs_block_size)) >= 64 || bs_block_size < MIN_DATA_BLOCK_SIZE || bs_block_size >= MAX_DATA_BLOCK_SIZE)
+    {
+        throw std::runtime_error("Bad block size");
+    }
+    // Cluster-wide immediate_commit mode
+    immediate_commit = (config["immediate_commit"] == "all");
     if (config.find("client_max_dirty_bytes") != config.end())
     {
         client_max_dirty_bytes = config["client_max_dirty_bytes"].uint64_value();
@@ -349,15 +379,9 @@ void cluster_client_t::on_change_hook(std::map<std::string, etcd_kv_t> & changes
     continue_ops();
 }
 
-bool cluster_client_t::get_immediate_commit(uint64_t inode)
+bool cluster_client_t::get_immediate_commit()
 {
-    pool_id_t pool_id = INODE_POOL(inode);
-    if (!pool_id)
-        return true;
-    auto pool_it = st_cli.pool_config.find(pool_id);
-    if (pool_it == st_cli.pool_config.end())
-        return true;
-    return pool_it->second.immediate_commit == IMMEDIATE_ALL;
+    return immediate_commit;
 }
 
 void cluster_client_t::on_change_osd_state_hook(uint64_t peer_osd)
@@ -415,45 +439,9 @@ void cluster_client_t::execute(cluster_op_t *op)
         std::function<void(cluster_op_t*)>(op->callback)(op);
         return;
     }
-    if (!pgs_loaded)
-    {
-        offline_ops.push_back(op);
-        return;
-    }
     op->cur_inode = op->inode;
     op->retval = 0;
-    op->flags = op->flags & OSD_OP_IGNORE_READONLY; // single allowed flag
-    if (op->opcode != OSD_OP_SYNC)
-    {
-        pool_id_t pool_id = INODE_POOL(op->cur_inode);
-        if (!pool_id)
-        {
-            op->retval = -EINVAL;
-            std::function<void(cluster_op_t*)>(op->callback)(op);
-            return;
-        }
-        auto pool_it = st_cli.pool_config.find(pool_id);
-        if (pool_it == st_cli.pool_config.end() || pool_it->second.real_pg_count == 0)
-        {
-            // Pools are loaded, but this one is unknown
-            op->retval = -EINVAL;
-            std::function<void(cluster_op_t*)>(op->callback)(op);
-            return;
-        }
-        // Check alignment
-        if ((op->opcode == OSD_OP_READ || op->opcode == OSD_OP_WRITE) && !op->len ||
-            op->offset % pool_it->second.bitmap_granularity || op->len % pool_it->second.bitmap_granularity)
-        {
-            op->retval = -EINVAL;
-            std::function<void(cluster_op_t*)>(op->callback)(op);
-            return;
-        }
-        if (pool_it->second.immediate_commit == IMMEDIATE_ALL)
-        {
-            op->flags |= OP_IMMEDIATE_COMMIT;
-        }
-    }
-    if (op->opcode == OSD_OP_WRITE && !(op->flags & OP_IMMEDIATE_COMMIT))
+    if (op->opcode == OSD_OP_WRITE && !immediate_commit)
     {
         if (dirty_bytes >= client_max_dirty_bytes || dirty_ops >= client_max_dirty_ops)
         {
@@ -492,9 +480,9 @@ void cluster_client_t::execute(cluster_op_t *op)
     }
     else
         op_queue_tail = op_queue_head = op;
-    if (!(op->flags & OP_IMMEDIATE_COMMIT))
+    if (!immediate_commit)
         calc_wait(op);
-    else
+    else if (pgs_loaded)
     {
         if (op->opcode == OSD_OP_SYNC)
             continue_sync(op);
@@ -622,6 +610,28 @@ int cluster_client_t::continue_rw(cluster_op_t *op)
     else if (op->state == 3)
         goto resume_3;
 resume_0:
+    if ((op->opcode == OSD_OP_READ || op->opcode == OSD_OP_WRITE) && !op->len ||
+        op->offset % bs_bitmap_granularity || op->len % bs_bitmap_granularity)
+    {
+        op->retval = -EINVAL;
+        erase_op(op);
+        return 1;
+    }
+    {
+        pool_id_t pool_id = INODE_POOL(op->cur_inode);
+        if (!pool_id)
+        {
+            op->retval = -EINVAL;
+            erase_op(op);
+            return 1;
+        }
+        if (st_cli.pool_config.find(pool_id) == st_cli.pool_config.end() ||
+            st_cli.pool_config[pool_id].real_pg_count == 0)
+        {
+            // Postpone operations to unknown pools
+            return 0;
+        }
+    }
     if (op->opcode == OSD_OP_WRITE || op->opcode == OSD_OP_DELETE)
     {
         if (!(op->flags & OSD_OP_IGNORE_READONLY))
@@ -634,7 +644,7 @@ resume_0:
                 return 1;
             }
         }
-        if (op->opcode == OSD_OP_WRITE && !(op->flags & OP_IMMEDIATE_COMMIT) && !(op->flags & OP_FLUSH_BUFFER))
+        if (op->opcode == OSD_OP_WRITE && !immediate_commit && !(op->flags & OP_FLUSH_BUFFER))
         {
             copy_write(op, dirty_buffers);
         }
@@ -804,7 +814,7 @@ void cluster_client_t::slice_rw(cluster_op_t *op)
     // Primary OSDs still operate individual stripes, but their size is multiplied by PG minsize in case of EC
     auto & pool_cfg = st_cli.pool_config.at(INODE_POOL(op->cur_inode));
     uint32_t pg_data_size = (pool_cfg.scheme == POOL_SCHEME_REPLICATED ? 1 : pool_cfg.pg_size-pool_cfg.parity_chunks);
-    uint64_t pg_block_size = pool_cfg.data_block_size * pg_data_size;
+    uint64_t pg_block_size = bs_block_size * pg_data_size;
     uint64_t first_stripe = (op->offset / pg_block_size) * pg_block_size;
     uint64_t last_stripe = op->len > 0 ? ((op->offset + op->len - 1) / pg_block_size) * pg_block_size : first_stripe;
     op->retval = 0;
@@ -812,9 +822,9 @@ void cluster_client_t::slice_rw(cluster_op_t *op)
     if (op->opcode == OSD_OP_READ || op->opcode == OSD_OP_READ_BITMAP)
     {
         // Allocate memory for the bitmap
-        unsigned object_bitmap_size = (((op->opcode == OSD_OP_READ_BITMAP ? pg_block_size : op->len) / pool_cfg.bitmap_granularity + 7) / 8);
+        unsigned object_bitmap_size = (((op->opcode == OSD_OP_READ_BITMAP ? pg_block_size : op->len) / bs_bitmap_granularity + 7) / 8);
         object_bitmap_size = (object_bitmap_size < 8 ? 8 : object_bitmap_size);
-        unsigned bitmap_mem = object_bitmap_size + (pool_cfg.data_block_size / pool_cfg.bitmap_granularity / 8 * pg_data_size) * op->parts.size();
+        unsigned bitmap_mem = object_bitmap_size + (bs_bitmap_size * pg_data_size) * op->parts.size();
         if (op->bitmap_buf_size < bitmap_mem)
         {
             op->bitmap_buf = realloc_or_die(op->bitmap_buf, bitmap_mem);
@@ -844,7 +854,7 @@ void cluster_client_t::slice_rw(cluster_op_t *op)
             bool skip_prev = true;
             while (cur < end)
             {
-                unsigned bmp_loc = (cur - op->offset)/pool_cfg.bitmap_granularity;
+                unsigned bmp_loc = (cur - op->offset)/bs_bitmap_granularity;
                 bool skip = (((*((uint8_t*)op->bitmap_buf + bmp_loc/8)) >> (bmp_loc%8)) & 0x1);
                 if (skip_prev != skip)
                 {
@@ -862,7 +872,7 @@ void cluster_client_t::slice_rw(cluster_op_t *op)
                     skip_prev = skip;
                     prev = cur;
                 }
-                cur += pool_cfg.bitmap_granularity;
+                cur += bs_bitmap_granularity;
             }
             assert(cur > prev);
             if (skip_prev)
@@ -894,7 +904,7 @@ bool cluster_client_t::affects_osd(uint64_t inode, uint64_t offset, uint64_t len
 {
     auto & pool_cfg = st_cli.pool_config.at(INODE_POOL(inode));
     uint32_t pg_data_size = (pool_cfg.scheme == POOL_SCHEME_REPLICATED ? 1 : pool_cfg.pg_size-pool_cfg.parity_chunks);
-    uint64_t pg_block_size = pool_cfg.data_block_size * pg_data_size;
+    uint64_t pg_block_size = bs_block_size * pg_data_size;
     uint64_t first_stripe = (offset / pg_block_size) * pg_block_size;
     uint64_t last_stripe = len > 0 ? ((offset + len - 1) / pg_block_size) * pg_block_size : first_stripe;
     for (uint64_t stripe = first_stripe; stripe <= last_stripe; stripe += pg_block_size)
@@ -925,7 +935,7 @@ bool cluster_client_t::try_send(cluster_op_t *op, int i)
             part->osd_num = primary_osd;
             part->flags |= PART_SENT;
             op->inflight_count++;
-            uint64_t pg_bitmap_size = (pool_cfg.data_block_size / pool_cfg.bitmap_granularity / 8) * (
+            uint64_t pg_bitmap_size = bs_bitmap_size * (
                 pool_cfg.scheme == POOL_SCHEME_REPLICATED ? 1 : pool_cfg.pg_size-pool_cfg.parity_chunks
             );
             uint64_t meta_rev = 0;
@@ -941,7 +951,7 @@ bool cluster_client_t::try_send(cluster_op_t *op, int i)
                 .req = { .rw = {
                     .header = {
                         .magic = SECONDARY_OSD_OP_MAGIC,
-                        .id = next_op_id(),
+                        .id = op_id++,
                         .opcode = op->opcode == OSD_OP_READ_BITMAP ? OSD_OP_READ : op->opcode,
                     },
                     .inode = op->cur_inode,
@@ -973,7 +983,7 @@ int cluster_client_t::continue_sync(cluster_op_t *op)
 {
     if (op->state == 1)
         goto resume_1;
-    if (!dirty_osds.size())
+    if (immediate_commit || !dirty_osds.size())
     {
         // Sync is not required in the immediate_commit mode or if there are no dirty_osds
         op->retval = 0;
@@ -1069,7 +1079,7 @@ void cluster_client_t::send_sync(cluster_op_t *op, cluster_op_part_t *part)
         .req = {
             .hdr = {
                 .magic = SECONDARY_OSD_OP_MAGIC,
-                .id = next_op_id(),
+                .id = op_id++,
                 .opcode = OSD_OP_SYNC,
             },
         },
@@ -1130,8 +1140,7 @@ void cluster_client_t::handle_op_part(cluster_op_part_t *part)
     else
     {
         // OK
-        if (!(op->flags & OP_IMMEDIATE_COMMIT))
-            dirty_osds.insert(part->osd_num);
+        dirty_osds.insert(part->osd_num);
         part->flags |= PART_DONE;
         op->done_count++;
         if (op->opcode == OSD_OP_READ || op->opcode == OSD_OP_READ_BITMAP)
@@ -1153,12 +1162,12 @@ void cluster_client_t::copy_part_bitmap(cluster_op_t *op, cluster_op_part_t *par
 {
     // Copy (OR) bitmap
     auto & pool_cfg = st_cli.pool_config.at(INODE_POOL(op->cur_inode));
-    uint32_t pg_block_size = pool_cfg.data_block_size * (
+    uint32_t pg_block_size = bs_block_size * (
         pool_cfg.scheme == POOL_SCHEME_REPLICATED ? 1 : pool_cfg.pg_size-pool_cfg.parity_chunks
     );
-    uint32_t object_offset = (part->op.req.rw.offset - op->offset) / pool_cfg.bitmap_granularity;
-    uint32_t part_offset = (part->op.req.rw.offset % pg_block_size) / pool_cfg.bitmap_granularity;
-    uint32_t part_len = (op->opcode == OSD_OP_READ_BITMAP ? pg_block_size : part->op.req.rw.len) / pool_cfg.bitmap_granularity;
+    uint32_t object_offset = (part->op.req.rw.offset - op->offset) / bs_bitmap_granularity;
+    uint32_t part_offset = (part->op.req.rw.offset % pg_block_size) / bs_bitmap_granularity;
+    uint32_t part_len = (op->opcode == OSD_OP_READ_BITMAP ? pg_block_size : part->op.req.rw.len) / bs_bitmap_granularity;
     if (!(object_offset & 0x7) && !(part_offset & 0x7) && (part_len >= 8))
     {
         // Copy bytes
@@ -1181,5 +1190,5 @@ void cluster_client_t::copy_part_bitmap(cluster_op_t *op, cluster_op_part_t *par
 
 uint64_t cluster_client_t::next_op_id()
 {
-    return msgr.next_subop_id++;
+    return op_id++;
 }

src/cluster_client.h

@@ -6,6 +6,8 @@
 #include "messenger.h"
 #include "etcd_state_client.h"
 
+#define MIN_DATA_BLOCK_SIZE 4*1024
+#define MAX_DATA_BLOCK_SIZE 128*1024*1024
 #define DEFAULT_CLIENT_MAX_DIRTY_BYTES 32*1024*1024
 #define DEFAULT_CLIENT_MAX_DIRTY_OPS 1024
 #define INODE_LIST_DONE 1
@@ -77,7 +79,11 @@ class cluster_client_t
     timerfd_manager_t *tfd;
     ring_loop_t *ringloop;
 
+    uint64_t bs_block_size = 0;
+    uint32_t bs_bitmap_granularity = 0, bs_bitmap_size = 0;
     std::map<pool_id_t, uint64_t> pg_counts;
+    // WARNING: initially true so execute() doesn't create fake sync
+    bool immediate_commit = true;
     // FIXME: Implement inmemory_commit mode. Note that it requires to return overlapping reads from memory.
     uint64_t client_max_dirty_bytes = 0;
     uint64_t client_max_dirty_ops = 0;
@@ -85,6 +91,7 @@ class cluster_client_t
     int up_wait_retry_interval = 500; // ms
 
     int retry_timeout_id = 0;
+    uint64_t op_id = 1;
     std::vector<cluster_op_t*> offline_ops;
     cluster_op_t *op_queue_head = NULL, *op_queue_tail = NULL;
     std::map<object_id, cluster_buffer_t> dirty_buffers;
@@ -112,7 +119,7 @@ public:
     bool is_ready();
     void on_ready(std::function<void(void)> fn);
 
-    bool get_immediate_commit(uint64_t inode);
+    bool get_immediate_commit();
 
     static void copy_write(cluster_op_t *op, std::map<object_id, cluster_buffer_t> & dirty_buffers);
     void continue_ops(bool up_retry = false);
@@ -120,8 +127,8 @@ public:
         std::function<void(inode_list_t* lst, std::set<object_id>&& objects, pg_num_t pg_num, osd_num_t primary_osd, int status)> callback);
     int list_pg_count(inode_list_t *lst);
     void list_inode_next(inode_list_t *lst, int next_pgs);
-    //inline uint32_t get_bs_bitmap_granularity() { return st_cli.global_bitmap_granularity; }
-    //inline uint64_t get_bs_block_size() { return st_cli.global_block_size; }
+    inline uint32_t get_bs_bitmap_granularity() { return bs_bitmap_granularity; }
+    inline uint64_t get_bs_block_size() { return bs_block_size; }
     uint64_t next_op_id();
 
 protected: