WIP Use random_hier_combinations

Replace flatten_tree with extract_tree_levels
Implement multi-level tree extractor for hierarchical failure domains
2023-05-18 17:44:00 +03:00 · 2023-05-18 17:44:00 +03:00 · 2023-05-18 17:44:00 +03:00 · 2023-05-18 17:44:00 +03:00 · 2023-05-18 11:54:47 +03:00 · 2023-05-18 00:42:40 +03:00
129 changed files with 4006 additions and 1023 deletions
--- a/.gitea/workflows/buildenv.Dockerfile
+++ b/.gitea/workflows/buildenv.Dockerfile
@@ -0,0 +1,36 @@
 FROM node:16-bullseye
 WORKDIR /root
 ADD ./docker/vitastor.gpg /etc/apt/trusted.gpg.d
 RUN echo 'deb http://deb.debian.org/debian bullseye-backports main' >> /etc/apt/sources.list; \
    echo 'deb http://vitastor.io/debian bullseye main' >> /etc/apt/sources.list; \
    echo >> /etc/apt/preferences; \
    echo 'Package: *' >> /etc/apt/preferences; \
    echo 'Pin: release a=bullseye-backports' >> /etc/apt/preferences; \
    echo 'Pin-Priority: 500' >> /etc/apt/preferences; \
    echo >> /etc/apt/preferences; \
    echo 'Package: *' >> /etc/apt/preferences; \
    echo 'Pin: origin "vitastor.io"' >> /etc/apt/preferences; \
    echo 'Pin-Priority: 1000' >> /etc/apt/preferences; \
    grep '^deb ' /etc/apt/sources.list | perl -pe 's/^deb/deb-src/' >> /etc/apt/sources.list; \
    echo 'APT::Install-Recommends false;' >> /etc/apt/apt.conf; \
    echo 'APT::Install-Suggests false;' >> /etc/apt/apt.conf
 RUN apt-get update
 RUN apt-get -y install etcd qemu-system-x86 qemu-block-extra qemu-utils fio libasan5 \
    liburing1 liburing-dev libgoogle-perftools-dev devscripts libjerasure-dev cmake libibverbs-dev libisal-dev
 RUN apt-get -y build-dep fio qemu=`dpkg -s qemu-system-x86|grep ^Version:|awk '{print $2}'`
 RUN apt-get -y install jq lp-solve sudo
 RUN apt-get --download-only source fio qemu=`dpkg -s qemu-system-x86|grep ^Version:|awk '{print $2}'`
 RUN set -ex; \
    mkdir qemu-build; \
    cd qemu-build; \
    dpkg-source -x /root/qemu*.dsc; \
    cd qemu*/; \
    debian/rules configure-qemu || debian/rules b/configure-stamp; \
    cd b/qemu; \
    make -j8 config-poison.h || true; \
    make -j8 qapi/qapi-builtin-types.h
--- a/.gitea/workflows/test.Dockerfile
+++ b/.gitea/workflows/test.Dockerfile
@@ -0,0 +1,16 @@
 FROM git.yourcmc.ru/vitalif/vitastor/buildenv
 ADD . /root/vitastor
 RUN set -e -x; \
    mkdir -p /root/fio-build/; \
    cd /root/fio-build/; \
    dpkg-source -x /root/fio*.dsc; \
    cd /root/vitastor; \
    ln -s /root/fio-build/fio-*/ ./fio; \
    ln -s /root/qemu-build/qemu-*/ ./qemu; \
    ls /usr/include/linux/raw.h || cp ./debian/raw.h /usr/include/linux/raw.h; \
    mkdir build; \
    cd build; \
    cmake .. -DWITH_ASAN=yes -DWITH_QEMU=yes; \
    make -j16
--- a/.gitea/workflows/test.yml
+++ b/.gitea/workflows/test.yml
@@ -0,0 +1,552 @@
 name: Test
 on:
  push:
    branches:
    - '*'
    paths:
    - '.gitea/**'
    - 'src/**'
    - 'mon/**'
    - 'json11'
    - 'cpp-btree'
    - 'tests/**'
 env:
  BUILDENV_IMAGE: git.yourcmc.ru/vitalif/vitastor/buildenv
  TEST_IMAGE: git.yourcmc.ru/vitalif/vitastor/test
  OSD_ARGS: '--etcd_quick_timeout 2000'
 concurrency:
  group: ci-${{ github.ref }}
  cancel-in-progress: true
 jobs:
  buildenv:
    runs-on: ubuntu-latest
    container: git.yourcmc.ru/vitalif/gitea-ci-dind
    steps:
    - uses: actions/checkout@v3
    - name: Build and push
      run: |
        set -ex
        if ! docker manifest inspect $BUILDENV_IMAGE >/dev/null; then
          docker build -t $BUILDENV_IMAGE -f .gitea/workflows/buildenv.Dockerfile .
          docker login git.yourcmc.ru -u vitalif -p "${{secrets.TOKEN}}"
          docker push $BUILDENV_IMAGE
        fi
  build:
    runs-on: ubuntu-latest
    needs: buildenv
    container: git.yourcmc.ru/vitalif/gitea-ci-dind
    steps:
    - uses: actions/checkout@v3
      with:
        submodules: true
    - name: Build and push
      run: |
        set -ex
        if ! docker manifest inspect $TEST_IMAGE:$GITHUB_SHA >/dev/null; then
          docker build -t $TEST_IMAGE:$GITHUB_SHA -f .gitea/workflows/test.Dockerfile .
          docker login git.yourcmc.ru -u vitalif -p "${{secrets.TOKEN}}"
          docker push $TEST_IMAGE:$GITHUB_SHA
        fi
  make_test:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    # leak sanitizer sometimes crashes
    - run: cd /root/vitastor/build && ASAN_OPTIONS=detect_leaks=0 make -j16 test
  test_add_osd:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_add_osd.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_cas:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_cas.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_change_pg_count:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_change_pg_count.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_change_pg_count_ec:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: SCHEME=ec /root/vitastor/tests/test_change_pg_count.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_change_pg_size:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_change_pg_size.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_create_nomaxid:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_create_nomaxid.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_etcd_fail:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 10
      run: /root/vitastor/tests/test_etcd_fail.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_failure_domain:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_failure_domain.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_interrupted_rebalance:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 10
      run: /root/vitastor/tests/test_interrupted_rebalance.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_interrupted_rebalance_imm:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 10
      run: IMMEDIATE_COMMIT=1 /root/vitastor/tests/test_interrupted_rebalance.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_interrupted_rebalance_ec:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 10
      run: SCHEME=ec /root/vitastor/tests/test_interrupted_rebalance.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_interrupted_rebalance_ec_imm:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 10
      run: SCHEME=ec IMMEDIATE_COMMIT=1 /root/vitastor/tests/test_interrupted_rebalance.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_minsize_1:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_minsize_1.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_move_reappear:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_move_reappear.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_rebalance_verify:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_rebalance_verify.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_rebalance_verify_imm:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: IMMEDIATE_COMMIT=1 /root/vitastor/tests/test_rebalance_verify.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_rebalance_verify_ec:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: SCHEME=ec /root/vitastor/tests/test_rebalance_verify.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_rebalance_verify_ec_imm:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: SCHEME=ec IMMEDIATE_COMMIT=1 /root/vitastor/tests/test_rebalance_verify.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_rm:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_rm.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_snapshot:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_snapshot.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_snapshot_ec:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: SCHEME=ec /root/vitastor/tests/test_snapshot.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_splitbrain:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_splitbrain.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_write:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_write.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_write_xor:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: SCHEME=xor /root/vitastor/tests/test_write.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_write_no_same:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 3
      run: /root/vitastor/tests/test_write_no_same.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_heal_pg_size_2:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 10
      run: PG_SIZE=2 /root/vitastor/tests/test_heal.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
  test_heal_ec:
    runs-on: ubuntu-latest
    needs: build
    container: ${{env.TEST_IMAGE}}:${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: 10
      run: SCHEME=ec /root/vitastor/tests/test_heal.sh
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- $i --------"
          cat $i
          echo ""
        done
--- a/.gitea/workflows/tests-to-yaml.pl
+++ b/.gitea/workflows/tests-to-yaml.pl
@@ -0,0 +1,68 @@
 #!/usr/bin/perl
 use strict;
 for my $line (<>)
 {
    if ($line =~ /\.\/(test_[^\.]+)/s)
    {
        chomp $line;
        my $test_name = $1;
        my $timeout = 3;
        if ($test_name eq 'test_etcd_fail' || $test_name eq 'test_heal' || $test_name eq 'test_interrupted_rebalance')
        {
            $timeout = 10;
        }
        while ($line =~ /([^\s=]+)=(\S+)/gs)
        {
            if ($1 eq 'SCHEME' && $2 eq 'ec')
            {
                $test_name .= '_ec';
            }
            elsif ($1 eq 'SCHEME' && $2 eq 'xor')
            {
                $test_name .= '_xor';
            }
            elsif ($1 eq 'IMMEDIATE_COMMIT')
            {
                $test_name .= '_imm';
            }
            else
            {
                $test_name .= '_'.lc($1).'_'.$2;
            }
        }
        $line =~ s!\./test_!/root/vitastor/tests/test_!;
        # Gitea CI doesn't support artifacts yet, lol
        #- name: Upload results
        #  uses: actions/upload-artifact\@v3
        #  if: always()
        #  with:
        #    name: ${test_name}_result
        #    path: |
        #      /root/vitastor/testdata
        #      !/root/vitastor/testdata/*.bin
        #    retention-days: 5
        print <<"EOF"
  $test_name:
    runs-on: ubuntu-latest
    needs: build
    container: \${{env.TEST_IMAGE}}:\${{github.sha}}
    steps:
    - name: Run test
      id: test
      timeout-minutes: $timeout
      run: $line
    - name: Print logs
      if: always() && steps.test.outcome == 'failure'
      run: |
        for i in /root/vitastor/testdata/*.log /root/vitastor/testdata/*.txt; do
          echo "-------- \$i --------"
          cat \$i
          echo ""
        done
 EOF
 ;
    }
 }
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1,7 +1,7 @@
-cmake_minimum_required(VERSION 2.8)
+cmake_minimum_required(VERSION 2.8.12)
 project(vitastor)
-set(VERSION "0.8.4")
+set(VERSION "0.8.9")
 add_subdirectory(src)
--- a/VNPL-1.1-RU.txt
+++ b/VNPL-1.1-RU.txt
@@ -48,9 +48,9 @@ Vitastor, составлены для того, чтобы убедиться,
 интерфейс (прокси), опять же, без открытия в свободный публичный доступ как
 самой программы, так и прокси.
-  Сетевая Публичная Лицензия Vitastor разработана специально чтобы
+  Сетевая Публичная Лицензия Vitastor разработана специально, чтобы
 гарантировать, что в таких случаях и модифицированная версия программы, и
-прокси оставались доступными сообществу. Для этого лицензия требует от
+прокси останутся доступными сообществу. Для этого лицензия требует от
 операторов сетевых серверов предоставлять исходный код оригинальной программы,
 а также всех других программ, взаимодействующих с ней на их серверах,
 пользователям этих серверов, на условиях свободных лицензий. Таким образом,
--- a/csi/Makefile
+++ b/csi/Makefile
@@ -1,4 +1,4 @@
-VERSION ?= v0.8.4
+VERSION ?= v0.8.9
 all: build push
--- a/csi/deploy/004-csi-nodeplugin.yaml
+++ b/csi/deploy/004-csi-nodeplugin.yaml
@@ -49,7 +49,7 @@ spec:
            capabilities:
              add: ["SYS_ADMIN"]
            allowPrivilegeEscalation: true
-          image: vitalif/vitastor-csi:v0.8.4
+          image: vitalif/vitastor-csi:v0.8.9
          args:
            - "--node=$(NODE_ID)"
            - "--endpoint=$(CSI_ENDPOINT)"
--- a/csi/deploy/007-csi-provisioner.yaml
+++ b/csi/deploy/007-csi-provisioner.yaml
@@ -116,7 +116,7 @@ spec:
            privileged: true
            capabilities:
              add: ["SYS_ADMIN"]
-          image: vitalif/vitastor-csi:v0.8.4
+          image: vitalif/vitastor-csi:v0.8.9
          args:
            - "--node=$(NODE_ID)"
            - "--endpoint=$(CSI_ENDPOINT)"
--- a/csi/go.mod
+++ b/csi/go.mod
@@ -4,25 +4,10 @@ go 1.15
 require (
 	github.com/container-storage-interface/spec v1.4.0
 	github.com/coreos/bbolt v0.0.0-00010101000000-000000000000 // indirect
 	github.com/coreos/etcd v3.3.25+incompatible // indirect
 	github.com/coreos/go-semver v0.3.0 // indirect
 	github.com/coreos/go-systemd v0.0.0-20191104093116-d3cd4ed1dbcf // indirect
 	github.com/coreos/pkg v0.0.0-20180928190104-399ea9e2e55f // indirect
 	github.com/dustin/go-humanize v1.0.0 // indirect
 	github.com/golang/glog v0.0.0-20160126235308-23def4e6c14b
 	github.com/gorilla/websocket v1.4.2 // indirect
 	github.com/grpc-ecosystem/go-grpc-middleware v1.3.0 // indirect
 	github.com/grpc-ecosystem/go-grpc-prometheus v1.2.0 // indirect
 	github.com/grpc-ecosystem/grpc-gateway v1.16.0 // indirect
 	github.com/jonboulle/clockwork v0.2.2 // indirect
 	github.com/kubernetes-csi/csi-lib-utils v0.9.1
 	github.com/soheilhy/cmux v0.1.5 // indirect
 	github.com/tmc/grpc-websocket-proxy v0.0.0-20201229170055-e5319fda7802 // indirect
 	github.com/xiang90/probing v0.0.0-20190116061207-43a291ad63a2 // indirect
 	go.etcd.io/bbolt v0.0.0-00010101000000-000000000000 // indirect
 	go.etcd.io/etcd v3.3.25+incompatible
 	golang.org/x/net v0.0.0-20201202161906-c7110b5ffcbb
 	golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1 // indirect
 	google.golang.org/grpc v1.33.1
 	k8s.io/klog v1.0.0
 	k8s.io/utils v0.0.0-20210305010621-2afb4311ab10
--- a/csi/go.sum
+++ b/csi/go.sum
@@ -31,14 +31,11 @@ github.com/alecthomas/template v0.0.0-20160405071501-a0175ee3bccc/go.mod h1:LOuy
 github.com/alecthomas/template v0.0.0-20190718012654-fb15b899a751/go.mod h1:LOuyumcjzFXgccqObfd/Ljyb9UuFJ6TxHnclSeseNhc=
 github.com/alecthomas/units v0.0.0-20151022065526-2efee857e7cf/go.mod h1:ybxpYRFXyAe+OPACYpWeL0wqObRcbAqCMya13uyzqw0=
 github.com/alecthomas/units v0.0.0-20190717042225-c3de453c63f4/go.mod h1:ybxpYRFXyAe+OPACYpWeL0wqObRcbAqCMya13uyzqw0=
 github.com/antihax/optional v1.0.0/go.mod h1:uupD/76wgC+ih3iEmQUL+0Ugr19nfwCT1kdvxnR2qWY=
 github.com/beorn7/perks v0.0.0-20180321164747-3a771d992973/go.mod h1:Dwedo/Wpr24TaqPxmxbtue+5NUziq4I4S80YR8gNf3Q=
 github.com/beorn7/perks v1.0.0/go.mod h1:KWe93zE9D1o94FZ5RNwFwVgaQK1VOXiVxmqh+CedLV8=
 github.com/beorn7/perks v1.0.1 h1:VlbKKnNfV8bJzeqoa4cOKqO6bYr3WgKZxO8Z16+hsOM=
 github.com/beorn7/perks v1.0.1/go.mod h1:G2ZrVWU2WbWT9wwq4/hrbKbnv/1ERSJQ0ibhJ6rlkpw=
 github.com/blang/semver v3.5.0+incompatible/go.mod h1:kRBLl5iJ+tD4TcOOxsy/0fnwebNt5EWlYSAyrTnjyyk=
 github.com/census-instrumentation/opencensus-proto v0.2.1/go.mod h1:f6KPmirojxKA12rnyqOA5BBL4O983OfeGPqjHWSTneU=
 github.com/cespare/xxhash/v2 v2.1.1 h1:6MnRN8NT7+YBpUIWxHtefFZOKTAPgGjpQSxqLNn0+qY=
 github.com/cespare/xxhash/v2 v2.1.1/go.mod h1:VGX0DQ3Q6kWi7AoAeZDth3/j3BFtOZR5XLFGgcrjCOs=
 github.com/chzyer/logex v1.1.10/go.mod h1:+Ywpsq7O8HXn0nuIou7OrIPyXbp3wmkHB+jjWRnGsAI=
 github.com/chzyer/readline v0.0.0-20180603132655-2972be24d48e/go.mod h1:nSuG5e5PlCu98SY8svDHJxuZscDgtXS6KTTbou5AhLI=
@@ -46,25 +43,12 @@ github.com/chzyer/test v0.0.0-20180213035817-a1ea475d72b1/go.mod h1:Q3SI9o4m/ZMn
 github.com/container-storage-interface/spec v1.2.0/go.mod h1:6URME8mwIBbpVyZV93Ce5St17xBiQJQY67NDsuohiy4=
 github.com/container-storage-interface/spec v1.4.0 h1:ozAshSKxpJnYUfmkpZCTYyF/4MYeYlhdXbAvPvfGmkg=
 github.com/container-storage-interface/spec v1.4.0/go.mod h1:6URME8mwIBbpVyZV93Ce5St17xBiQJQY67NDsuohiy4=
 github.com/coreos/bbolt v1.3.5 h1:XFv7xaq7701j8ZSEzR28VohFYSlyakMyqNMU5FQH6Ac=
 github.com/coreos/bbolt v1.3.5/go.mod h1:G5EMThwa9y8QZGBClrRx5EY+Yw9kAhnjy3bSjsnlVTQ=
 github.com/coreos/etcd v3.3.25+incompatible h1:0GQEw6h3YnuOVdtwygkIfJ+Omx0tZ8/QkVyXI4LkbeY=
 github.com/coreos/etcd v3.3.25+incompatible/go.mod h1:uF7uidLiAD3TWHmW31ZFd/JWoc32PjwdhPthX9715RE=
 github.com/coreos/go-semver v0.3.0 h1:wkHLiw0WNATZnSG7epLsujiMCgPAc9xhjJ4tgnAxmfM=
 github.com/coreos/go-semver v0.3.0/go.mod h1:nnelYz7RCh+5ahJtPPxZlU+153eP4D4r3EedlOD2RNk=
 github.com/coreos/go-systemd v0.0.0-20191104093116-d3cd4ed1dbcf h1:iW4rZ826su+pqaw19uhpSCzhj44qo35pNgKFGqzDKkU=
 github.com/coreos/go-systemd v0.0.0-20191104093116-d3cd4ed1dbcf/go.mod h1:F5haX7vjVVG0kc13fIWeqUViNPyEJxv/OmvnBo0Yme4=
 github.com/coreos/pkg v0.0.0-20180928190104-399ea9e2e55f h1:lBNOc5arjvs8E5mO2tbpBpLoyyu8B6e44T7hJy6potg=
 github.com/coreos/pkg v0.0.0-20180928190104-399ea9e2e55f/go.mod h1:E3G3o1h8I7cfcXa63jLwjI0eiQQMgzzUDFVpN/nH/eA=
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/dgrijalva/jwt-go v3.2.0+incompatible h1:7qlOGliEKZXTDg6OTjfoBKDXWrumCAMpl/TFQ4/5kLM=
 github.com/dgrijalva/jwt-go v3.2.0+incompatible/go.mod h1:E3ru+11k8xSBh+hMPgOLZmtrrCbhqsmaPHjLKYnJCaQ=
 github.com/docker/spdystream v0.0.0-20160310174837-449fdfce4d96/go.mod h1:Qh8CwZgvJUkLughtfhJv5dyTYa91l1fOUCrgjqmcifM=
 github.com/docopt/docopt-go v0.0.0-20180111231733-ee0de3bc6815/go.mod h1:WwZ+bS3ebgob9U8Nd0kOddGdZWjyMGR8Wziv+TBNwSE=
 github.com/dustin/go-humanize v1.0.0 h1:VSnTsYCnlFHaM2/igO1h6X3HA71jcobQuxemgkq4zYo=
 github.com/dustin/go-humanize v1.0.0/go.mod h1:HtrtbFcZ19U5GC7JDqmcUSB87Iq5E25KnS6fMYU6eOk=
 github.com/elazarl/goproxy v0.0.0-20180725130230-947c36da3153/go.mod h1:/Zj4wYkgs4iZTTu3o/KG3Itv/qCCa8VVMlb3i9OVuzc=
 github.com/emicklei/go-restful v0.0.0-20170410110728-ff4f55a20633/go.mod h1:otzb+WCGbkyDHkqmQmT5YD2WR4BBwUdeQoFo8l/7tVs=
 github.com/envoyproxy/go-control-plane v0.9.0/go.mod h1:YTl/9mNaCwkRvm6d1a2C3ymFceY/DCBVvsKhRF0iEA4=
@@ -73,7 +57,6 @@ github.com/evanphx/json-patch v4.9.0+incompatible/go.mod h1:50XU6AFN0ol/bzJsmQLi
 github.com/fsnotify/fsnotify v1.4.7/go.mod h1:jwhsz4b93w/PPRr/qN1Yymfu8t87LnFCMoQvtojpjFo=
 github.com/fsnotify/fsnotify v1.4.9/go.mod h1:znqG4EE+3YCdAaPaxE2ZRY/06pZUdp0tY4IgpuI1SZQ=
 github.com/ghodss/yaml v0.0.0-20150909031657-73d445a93680/go.mod h1:4dBDuWmgqj2HViK6kFavaiC9ZROes6MMH2rRYeMEF04=
 github.com/ghodss/yaml v1.0.0/go.mod h1:4dBDuWmgqj2HViK6kFavaiC9ZROes6MMH2rRYeMEF04=
 github.com/go-gl/glfw/v3.3/glfw v0.0.0-20191125211704-12ad95a8df72/go.mod h1:tQ2UAYgL5IevRw8kRxooKSPJfGvJ9fJQFa0TUsXzTg8=
 github.com/go-kit/kit v0.8.0/go.mod h1:xBxKIO96dXMWWy0MnWVtmwkA9/13aqxPnvrjFYMA2as=
 github.com/go-kit/kit v0.9.0/go.mod h1:xBxKIO96dXMWWy0MnWVtmwkA9/13aqxPnvrjFYMA2as=
@@ -88,14 +71,10 @@ github.com/go-openapi/spec v0.0.0-20160808142527-6aced65f8501/go.mod h1:J8+jY1nA
 github.com/go-openapi/swag v0.0.0-20160704191624-1d0bd113de87/go.mod h1:DXUve3Dpr1UfpPtxFw+EFuQ41HhCWZfha5jSVRG7C7I=
 github.com/go-stack/stack v1.8.0/go.mod h1:v0f6uXyyMGvRgIKkXu+yp6POWl0qKG85gN/melR3HDY=
 github.com/gogo/protobuf v1.1.1/go.mod h1:r8qH/GZQm5c6nD/R0oafs1akxWv10x8SbQlK7atdtwQ=
 github.com/gogo/protobuf v1.3.1 h1:DqDEcV5aeaTmdFBePNpYsp3FlcVH/2ISVVM9Qf8PSls=
 github.com/gogo/protobuf v1.3.1/go.mod h1:SlYgWuQ5SjCEi6WLHjHCa1yvBfUnHcTbrrZtXPKa29o=
 github.com/gogo/protobuf v1.3.2 h1:Ov1cvc58UF3b5XjBnZv7+opcTcQFZebYjWzi34vdm4Q=
 github.com/gogo/protobuf v1.3.2/go.mod h1:P1XiOD3dCwIKUDQYPy72D8LYyHL2YPYrpS2s69NZV8Q=
 github.com/golang/glog v0.0.0-20160126235308-23def4e6c14b h1:VKtxabqXZkF25pY9ekfRL6a582T4P37/31XEstQ5p58=
 github.com/golang/glog v0.0.0-20160126235308-23def4e6c14b/go.mod h1:SBH7ygxi8pfUlaOkMMuAQtPIUF8ecWP5IEl/CR7VP2Q=
 github.com/golang/groupcache v0.0.0-20190702054246-869f871628b6/go.mod h1:cIg4eruTrX1D+g88fzRXU5OdNfaM+9IcxsU14FzY7Hc=
 github.com/golang/groupcache v0.0.0-20191227052852-215e87163ea7 h1:5ZkaAPbicIKTF2I64qf5Fh8Aa83Q/dnOafMYV0OMwjA=
 github.com/golang/groupcache v0.0.0-20191227052852-215e87163ea7/go.mod h1:cIg4eruTrX1D+g88fzRXU5OdNfaM+9IcxsU14FzY7Hc=
 github.com/golang/mock v1.1.1/go.mod h1:oTYuIxOrZwtPieC+H1uAHpcLFnEyAGVDL/k47Jfbm0A=
 github.com/golang/mock v1.2.0/go.mod h1:oTYuIxOrZwtPieC+H1uAHpcLFnEyAGVDL/k47Jfbm0A=
@@ -113,7 +92,6 @@ github.com/golang/protobuf v1.4.1/go.mod h1:U8fpvMrcmy5pZrNK1lt4xCsGvpyWQ/VVv6QD
 github.com/golang/protobuf v1.4.2 h1:+Z5KGCizgyZCbGh1KZqA0fcLLkwbsjIzS4aV2v7wJX0=
 github.com/golang/protobuf v1.4.2/go.mod h1:oDoupMAO8OvCJWAcko0GGGIgR6R6ocIYbsSw735rRwI=
 github.com/google/btree v0.0.0-20180813153112-4030bb1f1f0c/go.mod h1:lNA+9X1NB3Zf8V7Ke586lFgjr2dZNuvo3lPJSGZ5JPQ=
 github.com/google/btree v1.0.0 h1:0udJVsspx3VBr5FwtLhQQtuAsVc79tTq0ocGIPAU6qo=
 github.com/google/btree v1.0.0/go.mod h1:lNA+9X1NB3Zf8V7Ke586lFgjr2dZNuvo3lPJSGZ5JPQ=
 github.com/google/go-cmp v0.2.0/go.mod h1:oXzfMopK8JAjlY9xF4vHSVASa0yLyX7SntLO5aqRK0M=
 github.com/google/go-cmp v0.3.0/go.mod h1:8QqcDgzrUqlUb/G2PQTWiueGozuR1884gddMywk6iLU=
@@ -127,38 +105,24 @@ github.com/google/pprof v0.0.0-20181206194817-3ea8567a2e57/go.mod h1:zfwlbNMJ+OI
 github.com/google/pprof v0.0.0-20190515194954-54271f7e092f/go.mod h1:zfwlbNMJ+OItoe0UupaVj+oy1omPYYDuagoSzA8v9mc=
 github.com/google/pprof v0.0.0-20191218002539-d4f498aebedc/go.mod h1:ZgVRPoUq/hfqzAqh7sHMqb3I9Rq5C59dIz2SbBwJ4eM=
 github.com/google/renameio v0.1.0/go.mod h1:KWCgfxg9yswjAJkECMjeO8J8rahYeXnNhOm40UhjYkI=
 github.com/google/uuid v1.1.1 h1:Gkbcsh/GbpXz7lPftLA3P6TYMwjCLYm83jiFQZF/3gY=
 github.com/google/uuid v1.1.1/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
 github.com/googleapis/gax-go/v2 v2.0.4/go.mod h1:0Wqv26UfaUD9n4G6kQubkQ+KchISgw+vpHVxEJEs9eg=
 github.com/googleapis/gax-go/v2 v2.0.5/go.mod h1:DWXyrwAJ9X0FpwwEdw+IPEYBICEFu5mhpdKc/us6bOk=
 github.com/googleapis/gnostic v0.4.1/go.mod h1:LRhVm6pbyptWbWbuZ38d1eyptfvIytN3ir6b65WBswg=
 github.com/gorilla/websocket v1.4.2 h1:+/TMaTYc4QFitKJxsQ7Yye35DkWvkdLcvGKqM+x0Ufc=
 github.com/gorilla/websocket v1.4.2/go.mod h1:YR8l580nyteQvAITg2hZ9XVh4b55+EU/adAjf1fMHhE=
 github.com/gregjones/httpcache v0.0.0-20180305231024-9cad4c3443a7/go.mod h1:FecbI9+v66THATjSRHfNgh1IVFe/9kFxbXtjV0ctIMA=
 github.com/grpc-ecosystem/go-grpc-middleware v1.3.0 h1:+9834+KizmvFV7pXQGSXQTsaWhq2GjuNUt0aUU0YBYw=
 github.com/grpc-ecosystem/go-grpc-middleware v1.3.0/go.mod h1:z0ButlSOZa5vEBq9m2m2hlwIgKw+rp3sdCBRoJY+30Y=
 github.com/grpc-ecosystem/go-grpc-prometheus v1.2.0 h1:Ovs26xHkKqVztRpIrF/92BcuyuQ/YW4NSIpoGtfXNho=
 github.com/grpc-ecosystem/go-grpc-prometheus v1.2.0/go.mod h1:8NvIoxWQoOIhqOTXgfV/d3M/q6VIi02HzZEHgUlZvzk=
 github.com/grpc-ecosystem/grpc-gateway v1.16.0 h1:gmcG1KaJ57LophUzW0Hy8NmPhnMZb4M0+kPpLofRdBo=
 github.com/grpc-ecosystem/grpc-gateway v1.16.0/go.mod h1:BDjrQk3hbvj6Nolgz8mAMFbcEtjT1g+wF4CSlocrBnw=
 github.com/hashicorp/golang-lru v0.5.0/go.mod h1:/m3WP610KZHVQ1SGc6re/UDhFvYD7pJ4Ao+sR/qLZy8=
 github.com/hashicorp/golang-lru v0.5.1/go.mod h1:/m3WP610KZHVQ1SGc6re/UDhFvYD7pJ4Ao+sR/qLZy8=
 github.com/hpcloud/tail v1.0.0/go.mod h1:ab1qPbhIpdTxEkNHXyeSf5vhxWSCs/tWer42PpOxQnU=
 github.com/ianlancetaylor/demangle v0.0.0-20181102032728-5e5cf60278f6/go.mod h1:aSSvb/t6k1mPoxDqO4vJh6VOCGPwU4O0C2/Eqndh1Sc=
 github.com/imdario/mergo v0.3.5/go.mod h1:2EnlNZ0deacrJVfApfmtdGgDfMuh/nq6Ok1EcJh5FfA=
 github.com/jonboulle/clockwork v0.2.2 h1:UOGuzwb1PwsrDAObMuhUnj0p5ULPj8V/xJ7Kx9qUBdQ=
 github.com/jonboulle/clockwork v0.2.2/go.mod h1:Pkfl5aHPm1nk2H9h0bjmnJD/BcgbGXUBGnn1kMkgxc8=
 github.com/json-iterator/go v1.1.6/go.mod h1:+SdeFBvtyEkXs7REEP0seUULqWtbJapLOCVDaaPEHmU=
 github.com/json-iterator/go v1.1.10 h1:Kz6Cvnvv2wGdaG/V8yMvfkmNiXq9Ya2KUv4rouJJr68=
 github.com/json-iterator/go v1.1.10/go.mod h1:KdQUCv79m/52Kvf8AW2vK1V8akMuk1QjK/uOdHXbAo4=
 github.com/jstemmer/go-junit-report v0.0.0-20190106144839-af01ea7f8024/go.mod h1:6v2b51hI/fHJwM22ozAgKL4VKDeJcHhJFhtBdhmNjmU=
 github.com/jstemmer/go-junit-report v0.9.1/go.mod h1:Brl9GWCQeLvo8nXZwPNNblvFj/XSXhF0NWZEnDohbsk=
 github.com/julienschmidt/httprouter v1.2.0/go.mod h1:SYymIcj16QtmaHHD7aYtjjsJG7VTCxuUUipMqKk8s4w=
 github.com/kisielk/errcheck v1.2.0/go.mod h1:/BMXB+zMLi60iA8Vv6Ksmxu/1UDYcXs4uQLJ+jE2L00=
 github.com/kisielk/errcheck v1.5.0/go.mod h1:pFxgyoBC7bSaBwPgfKdkLd5X25qrDl4LWUI2bnpBCr8=
 github.com/kisielk/gotool v1.0.0/go.mod h1:XhKaO+MFFWcvkIS/tQcRk01m1F5IRFswLeQ+oQHNcck=
 github.com/konsorten/go-windows-terminal-sequences v1.0.1/go.mod h1:T0+1ngSBFLxvqU3pZ+m/2kptfBszLMUkC4ZK/EgS/cQ=
 github.com/konsorten/go-windows-terminal-sequences v1.0.3 h1:CE8S1cTafDpPvMhIxNJKvHsGVBgn1xWYf1NbHQhywc8=
 github.com/konsorten/go-windows-terminal-sequences v1.0.3/go.mod h1:T0+1ngSBFLxvqU3pZ+m/2kptfBszLMUkC4ZK/EgS/cQ=
 github.com/kr/logfmt v0.0.0-20140226030751-b84e30acd515/go.mod h1:+0opPa2QZZtGFBFZlji/RkVcI2GknAs/DXo4wKdlNEc=
 github.com/kr/pretty v0.1.0/go.mod h1:dAy3ld7l9f0ibDNOQOHHMYYIIbhfbHSm3C4ZsoJORNo=
@@ -171,14 +135,11 @@ github.com/kubernetes-csi/csi-lib-utils v0.9.1 h1:sGq6ifVujfMSkfTsMZip44Ttv8SDXv
 github.com/kubernetes-csi/csi-lib-utils v0.9.1/go.mod h1:8E2jVUX9j3QgspwHXa6LwyN7IHQDjW9jX3kwoWnSC+M=
 github.com/mailru/easyjson v0.0.0-20160728113105-d5b7844b561a/go.mod h1:C1wdFJiN94OJF2b5HbByQZoLdCWB1Yqtg26g4irojpc=
 github.com/matttproud/golang_protobuf_extensions v1.0.1/go.mod h1:D8He9yQNgCq6Z5Ld7szi9bcBfOoFv/3dc6xSMkL2PC0=
 github.com/matttproud/golang_protobuf_extensions v1.0.2-0.20181231171920-c182affec369 h1:I0XW9+e1XWDxdcEniV4rQAIOPUGDq67JSCiRCgGCZLI=
 github.com/matttproud/golang_protobuf_extensions v1.0.2-0.20181231171920-c182affec369/go.mod h1:BSXmuO+STAnVfrANrmjBb36TMTDstsz7MSK+HVaYKv4=
 github.com/moby/term v0.0.0-20200312100748-672ec06f55cd/go.mod h1:DdlQx2hp0Ss5/fLikoLlEeIYiATotOjgB//nb973jeo=
 github.com/modern-go/concurrent v0.0.0-20180228061459-e0a39a4cb421/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
 github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd h1:TRLaZ9cD/w8PVh93nsPXa1VrQ6jlwL5oN8l14QlcNfg=
 github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
 github.com/modern-go/reflect2 v0.0.0-20180701023420-4b7aa43c6742/go.mod h1:bx2lNnkwVCuqBIxFjflWJWanXIb3RllmbCylyMrvgv0=
 github.com/modern-go/reflect2 v1.0.1 h1:9f412s+6RmYXLWZSEzVVgPGK7C2PphHj5RJrvfx9AWI=
 github.com/modern-go/reflect2 v1.0.1/go.mod h1:bx2lNnkwVCuqBIxFjflWJWanXIb3RllmbCylyMrvgv0=
 github.com/munnerz/goautoneg v0.0.0-20120707110453-a547fc61f48d/go.mod h1:+n7T8mK8HuQTcFwEeznm/DIxMOiR9yIdICNftLE1DvQ=
 github.com/mwitkow/go-conntrack v0.0.0-20161129095857-cc309e4a2223/go.mod h1:qRWi+5nqEBWmkhHvq77mSJWrCKwh8bxhgT7d/eI7P4U=
@@ -188,38 +149,28 @@ github.com/onsi/ginkgo v1.6.0/go.mod h1:lLunBs/Ym6LB5Z9jYTR76FiuTmxDTDusOGeTQH+W
 github.com/onsi/ginkgo v1.11.0/go.mod h1:lLunBs/Ym6LB5Z9jYTR76FiuTmxDTDusOGeTQH+WWjE=
 github.com/onsi/gomega v0.0.0-20170829124025-dcabb60a477c/go.mod h1:C1qb7wdrVGGVU+Z6iS04AVkA3Q65CEZX59MT0QO5uiA=
 github.com/onsi/gomega v1.7.0/go.mod h1:ex+gbHU/CVuBBDIJjb2X0qEXbFg53c61hWP/1CpauHY=
 github.com/opentracing/opentracing-go v1.1.0/go.mod h1:UkNAQd3GIcIGf0SeVgPpRdFStlNbqXla1AfSYxPUl2o=
 github.com/peterbourgon/diskv v2.0.1+incompatible/go.mod h1:uqqh8zWWbv1HBMNONnaR/tNboyR3/BZd58JJSHlUSCU=
 github.com/pkg/errors v0.8.0/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
 github.com/pkg/errors v0.8.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
 github.com/pkg/errors v0.9.1 h1:FEBLx1zS214owpjy7qsBeixbURkuhQAwrK5UwLGTwt4=
 github.com/pkg/errors v0.9.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
 github.com/prometheus/client_golang v0.9.1/go.mod h1:7SWBe2y4D6OKWSNQJUaRYU/AaXPKyh/dDVn+NZz0KFw=
 github.com/prometheus/client_golang v1.0.0/go.mod h1:db9x61etRT2tGnBNRi70OPL5FsnadC4Ky3P0J6CfImo=
 github.com/prometheus/client_golang v1.7.1 h1:NTGy1Ja9pByO+xAeH/qiWnLrKtr3hJPNjaVUwnjpdpA=
 github.com/prometheus/client_golang v1.7.1/go.mod h1:PY5Wy2awLA44sXw4AOSfFBetzPP4j5+D6mVACh+pe2M=
 github.com/prometheus/client_model v0.0.0-20180712105110-5c3871d89910/go.mod h1:MbSGuTsp3dbXC40dX6PRTWyKYBIrTGTE9sqQNg2J8bo=
 github.com/prometheus/client_model v0.0.0-20190129233127-fd36f4220a90/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
 github.com/prometheus/client_model v0.0.0-20190812154241-14fe0d1b01d4/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
 github.com/prometheus/client_model v0.2.0 h1:uq5h0d+GuxiXLJLNABMgp2qUWDPiLvgCzz2dUR+/W/M=
 github.com/prometheus/client_model v0.2.0/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
 github.com/prometheus/common v0.4.1/go.mod h1:TNfzLD0ON7rHzMJeJkieUDPYmFC7Snx/y86RQel1bk4=
 github.com/prometheus/common v0.10.0 h1:RyRA7RzGXQZiW+tGMr7sxa85G1z0yOpM1qq5c8lNawc=
 github.com/prometheus/common v0.10.0/go.mod h1:Tlit/dnDKsSWFlCLTWaA1cyBgKHSMdTB80sz/V91rCo=
 github.com/prometheus/procfs v0.0.0-20181005140218-185b4288413d/go.mod h1:c3At6R/oaqEKCNdg8wHV1ftS6bRYblBhIjjI8uT2IGk=
 github.com/prometheus/procfs v0.0.2/go.mod h1:TjEm7ze935MbeOT/UhFTIMYKhuLP4wbCsTZCD3I8kEA=
 github.com/prometheus/procfs v0.1.3 h1:F0+tqvhOksq22sc6iCHF5WGlWjdwj92p0udFh1VFBS8=
 github.com/prometheus/procfs v0.1.3/go.mod h1:lV6e/gmhEcM9IjHGsFOCxxuZ+z1YqCvr4OA4YeYWdaU=
 github.com/rogpeppe/fastuuid v1.2.0/go.mod h1:jVj6XXZzXRy/MSR5jhDC/2q6DgLz+nrA6LYCDYWNEvQ=
 github.com/rogpeppe/go-internal v1.3.0/go.mod h1:M8bDsm7K2OlrFYOpmOWEs/qY81heoFRclV5y23lUDJ4=
 github.com/sirupsen/logrus v1.2.0/go.mod h1:LxeOpSwHxABJmUn/MG1IvRgCAasNZTLOkJPxbbu5VWo=
 github.com/sirupsen/logrus v1.4.2/go.mod h1:tLMulIdttU9McNUspp0xgXVQah82FyeX6MwdIuYE2rE=
 github.com/sirupsen/logrus v1.6.0 h1:UBcNElsrwanuuMsnGSlYmtmgbb23qDR5dG+6X6Oo89I=
 github.com/sirupsen/logrus v1.6.0/go.mod h1:7uNnSEd1DgxDLC74fIahvMZmmYsHGZGEOFrfsX/uA88=
 github.com/soheilhy/cmux v0.1.5 h1:jjzc5WVemNEDTLwv9tlmemhC73tI08BNOIGwBOo10Js=
 github.com/soheilhy/cmux v0.1.5/go.mod h1:T7TcVDs9LWfQgPlPsdngu6I6QIoyIFZDDC6sNE1GqG0=
 github.com/spf13/afero v1.2.2/go.mod h1:9ZxEEn6pIJ8Rxe320qSDBk6AsU0r9pR7Q4OcevTdifk=
 github.com/spf13/pflag v0.0.0-20170130214245-9ff6c6923cff/go.mod h1:DYY7MBk1bdzusC3SYhjObp+wFpr4gzcvqqNjLnInEg4=
 github.com/spf13/pflag v1.0.3/go.mod h1:DYY7MBk1bdzusC3SYhjObp+wFpr4gzcvqqNjLnInEg4=
@@ -231,24 +182,11 @@ github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UV
 github.com/stretchr/testify v1.4.0/go.mod h1:j7eGeouHqKxXV5pUuKE4zz7dFj8WfuZ+81PSLYec5m4=
 github.com/stretchr/testify v1.5.1 h1:nOGnQDM7FYENwehXlg/kFVnos3rEvtKTjRvOWSzb6H4=
 github.com/stretchr/testify v1.5.1/go.mod h1:5W2xD1RspED5o8YsWQXVCued0rvSQ+mT+I5cxcmMvtA=
 github.com/tmc/grpc-websocket-proxy v0.0.0-20201229170055-e5319fda7802 h1:uruHq4dN7GR16kFc5fp3d1RIYzJW5onx8Ybykw2YQFA=
 github.com/tmc/grpc-websocket-proxy v0.0.0-20201229170055-e5319fda7802/go.mod h1:ncp9v5uamzpCO7NfCPTXjqaC+bZgJeR0sMTm6dMHP7U=
 github.com/xiang90/probing v0.0.0-20190116061207-43a291ad63a2 h1:eY9dn8+vbi4tKz5Qo6v2eYzo7kUS51QINcR5jNpbZS8=
 github.com/xiang90/probing v0.0.0-20190116061207-43a291ad63a2/go.mod h1:UETIi67q53MR2AWcXfiuqkDkRtnGDLqkBTpCHuJHxtU=
 github.com/yuin/goldmark v1.1.27/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
 github.com/yuin/goldmark v1.2.1/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
 go.etcd.io/bbolt v1.3.5 h1:XAzx9gjCb0Rxj7EoqcClPD1d5ZBxZJk0jbuoPHenBt0=
 go.etcd.io/bbolt v1.3.5/go.mod h1:G5EMThwa9y8QZGBClrRx5EY+Yw9kAhnjy3bSjsnlVTQ=
 go.etcd.io/etcd v3.3.25+incompatible h1:V1RzkZJj9LqsJRy+TUBgpWSbZXITLB819lstuTFoZOY=
 go.etcd.io/etcd v3.3.25+incompatible/go.mod h1:yaeTdrJi5lOmYerz05bd8+V7KubZs8YSFZfzsF9A6aI=
 go.opencensus.io v0.21.0/go.mod h1:mSImk1erAIZhrmZN+AvHh14ztQfjbGwt4TtuofqLduU=
 go.opencensus.io v0.22.0/go.mod h1:+kGneAE2xo2IficOXnaByMWTGM9T73dGwxeWcUqIpI8=
 go.opencensus.io v0.22.2/go.mod h1:yxeiOL68Rb0Xd1ddK5vPZ/oVn4vY4Ynel7k9FzqtOIw=
 go.uber.org/atomic v1.4.0 h1:cxzIVoETapQEqDhQu3QfnvXAV4AlzcvUCxkVUFw3+EU=
 go.uber.org/atomic v1.4.0/go.mod h1:gD2HeocX3+yG+ygLZcrzQJaqmWj9AIm7n08wl/qW/PE=
 go.uber.org/multierr v1.1.0 h1:HoEmRHQPVSqub6w2z2d2EOVs2fjyFRGyofhKuyDq0QI=
 go.uber.org/multierr v1.1.0/go.mod h1:wR5kodmAFQ0UK8QlbwjlSNy0Z68gJhDJUG5sjR94q/0=
 go.uber.org/zap v1.10.0 h1:ORx85nbTijNz8ljznvCMR1ZBIPKFn3jQrag10X2AsuM=
 go.uber.org/zap v1.10.0/go.mod h1:vwi/ZaCAaUcBkycHslxD9B2zi4UTXhF60s6SWpuDF0Q=
 golang.org/x/crypto v0.0.0-20180904163835-0709b304e793/go.mod h1:6SG95UA2DQfeDnfUPMdvaQW0Q7yPrPDi9nlGo2tz2b4=
 golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
@@ -256,7 +194,6 @@ golang.org/x/crypto v0.0.0-20190510104115-cbcb75029529/go.mod h1:yigFU9vqHzYiE8U
 golang.org/x/crypto v0.0.0-20190605123033-f99c8df09eb5/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
 golang.org/x/crypto v0.0.0-20191011191535-87dc89f01550/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
 golang.org/x/crypto v0.0.0-20191206172530-e9b2fee46413/go.mod h1:LzIPMQfyMNhhGPhUkYOs5KpL4U8rLKemX1yGLhDgUto=
 golang.org/x/crypto v0.0.0-20200622213623-75b288015ac9 h1:psW17arqaxU48Z5kZ0CQnkZWQJsqcURM6tKiBApRjXI=
 golang.org/x/crypto v0.0.0-20200622213623-75b288015ac9/go.mod h1:LzIPMQfyMNhhGPhUkYOs5KpL4U8rLKemX1yGLhDgUto=
 golang.org/x/exp v0.0.0-20190121172915-509febef88a4/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
 golang.org/x/exp v0.0.0-20190306152737-a1d7652674e8/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
@@ -276,8 +213,6 @@ golang.org/x/mobile v0.0.0-20190719004257-d2bd2a29d028/go.mod h1:E/iHnbuqvinMTCc
 golang.org/x/mod v0.0.0-20190513183733-4bf6d317e70e/go.mod h1:mXi4GBBbnImb6dmsKGUJ2LatrhH/nqhxcFungHvyanc=
 golang.org/x/mod v0.1.0/go.mod h1:0QHyrYULN0/3qlju5TqG8bIK38QM8yzMo5ekMj3DlcY=
 golang.org/x/mod v0.1.1-0.20191105210325-c90efee705ee/go.mod h1:QqPTAvyqsEbceGzBzNggFXnrqF1CaUcvgkdR5Ot7KZg=
 golang.org/x/mod v0.2.0/go.mod h1:s0Qsj1ACt9ePp/hMypM3fl4fZqREWJwdYDEqhRiZZUA=
 golang.org/x/mod v0.3.0/go.mod h1:s0Qsj1ACt9ePp/hMypM3fl4fZqREWJwdYDEqhRiZZUA=
 golang.org/x/net v0.0.0-20180724234803-3673e40ba225/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
 golang.org/x/net v0.0.0-20180906233101-161cd47e91fd/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
 golang.org/x/net v0.0.0-20181114220301-adae6a3d119a/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
@@ -291,26 +226,20 @@ golang.org/x/net v0.0.0-20190603091049-60506f45cf65/go.mod h1:HSz+uSET+XFnRR8LxR
 golang.org/x/net v0.0.0-20190613194153-d28f0bde5980/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
 golang.org/x/net v0.0.0-20190620200207-3b0461eec859/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
 golang.org/x/net v0.0.0-20191209160850-c0dbc17a3553/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
 golang.org/x/net v0.0.0-20200226121028-0de0cce0169b/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
 golang.org/x/net v0.0.0-20200324143707-d3edc9973b7e/go.mod h1:qpuaurCH72eLCgpAm/N6yyVIVM9cpaDIP3A8BGJEC5A=
 golang.org/x/net v0.0.0-20200707034311-ab3426394381 h1:VXak5I6aEWmAXeQjA+QSZzlgNrpq9mjcfDemuexIKsU=
 golang.org/x/net v0.0.0-20200707034311-ab3426394381/go.mod h1:/O7V0waA8r7cgGh81Ro3o1hOxt32SMVPicZroKQ2sZA=
 golang.org/x/net v0.0.0-20200822124328-c89045814202/go.mod h1:/O7V0waA8r7cgGh81Ro3o1hOxt32SMVPicZroKQ2sZA=
 golang.org/x/net v0.0.0-20201021035429-f5854403a974/go.mod h1:sp8m0HH+o8qH0wwXwYZr8TS3Oi6o0r6Gce1SSxlDquU=
 golang.org/x/net v0.0.0-20201202161906-c7110b5ffcbb h1:eBmm0M9fYhWpKZLjQUUKka/LtIxf46G4fxeEz5KJr9U=
 golang.org/x/net v0.0.0-20201202161906-c7110b5ffcbb/go.mod h1:sp8m0HH+o8qH0wwXwYZr8TS3Oi6o0r6Gce1SSxlDquU=
 golang.org/x/oauth2 v0.0.0-20180821212333-d2e6202438be/go.mod h1:N/0e6XlmueqKjAGxoOufVs8QHGRruUQn6yWY3a++T0U=
 golang.org/x/oauth2 v0.0.0-20190226205417-e64efc72b421/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=
 golang.org/x/oauth2 v0.0.0-20190604053449-0f29369cfe45/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=
 golang.org/x/oauth2 v0.0.0-20191202225959-858c2ad4c8b6/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=
 golang.org/x/oauth2 v0.0.0-20200107190931-bf48bf16ab8d/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=
 golang.org/x/sync v0.0.0-20180314180146-1d60e4601c6f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20181108010431-42b317875d0f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20181221193216-37e7f081c4d4/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20190227155943-e225da77a7e6/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20190423024810-112230192c58/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20190911185100-cd5d95a43a6e/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20201020160332-67f06af15bc9/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sys v0.0.0-20180905080454-ebe1bf3edb33/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20180909124046-d0be0721c37e/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20181116152217-5ac8a444bdc5/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
@@ -326,11 +255,9 @@ golang.org/x/sys v0.0.0-20191005200804-aed5e4c7ecf9/go.mod h1:h1NjWce9XRLGQEsW7w
 golang.org/x/sys v0.0.0-20191204072324-ce4227a45e2e/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20191228213918-04cbcbbfeed8/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200106162015-b016eb3dc98e/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200202164722-d101bd2416d5/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200302150141-5c8b2ff67527/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200323222414-85ca7c5b95cd/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200615200032-f1bc736245b1/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200622214017-ed371f2e16b4 h1:5/PjkGUjvEU5Gl6BxmvKRPpqo2uNMv4rcHBMwzk/st8=
 golang.org/x/sys v0.0.0-20200622214017-ed371f2e16b4/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200930185726-fdedc70b468f h1:+Nyd8tzPX9R7BWHguqsrbFdRx3WQ/1ib8I44HXV5yTA=
 golang.org/x/sys v0.0.0-20200930185726-fdedc70b468f/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
@@ -341,7 +268,6 @@ golang.org/x/text v0.3.3 h1:cokOdA+Jmi5PJGXLlLllQSgYigAEfHXJAERHVMaCc2k=
 golang.org/x/text v0.3.3/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/time v0.0.0-20181108054448-85acf8d2951c/go.mod h1:tRJNPiyCQ0inRvYxbN9jk5I+vvW/OXSQhTDSoE431IQ=
 golang.org/x/time v0.0.0-20190308202827-9d24e82272b4/go.mod h1:tRJNPiyCQ0inRvYxbN9jk5I+vvW/OXSQhTDSoE431IQ=
 golang.org/x/time v0.0.0-20191024005414-555d28b269f0 h1:/5xXl8Y5W96D+TtHSlonuFqGHIWVuyCkGJLwGh9JJFs=
 golang.org/x/time v0.0.0-20191024005414-555d28b269f0/go.mod h1:tRJNPiyCQ0inRvYxbN9jk5I+vvW/OXSQhTDSoE431IQ=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
 golang.org/x/tools v0.0.0-20181011042414-1f849cf54d09/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
@@ -360,14 +286,10 @@ golang.org/x/tools v0.0.0-20190628153133-6cdbf07be9d0/go.mod h1:/rFqwRUd4F7ZHNgw
 golang.org/x/tools v0.0.0-20190816200558-6889da9d5479/go.mod h1:b+2E5dAYhXwXZwtnZ6UAqBI28+e2cm9otk0dWdXHAEo=
 golang.org/x/tools v0.0.0-20190911174233-4f2ddba30aff/go.mod h1:b+2E5dAYhXwXZwtnZ6UAqBI28+e2cm9otk0dWdXHAEo=
 golang.org/x/tools v0.0.0-20191012152004-8de300cfc20a/go.mod h1:b+2E5dAYhXwXZwtnZ6UAqBI28+e2cm9otk0dWdXHAEo=
 golang.org/x/tools v0.0.0-20191119224855-298f0cb1881e/go.mod h1:b+2E5dAYhXwXZwtnZ6UAqBI28+e2cm9otk0dWdXHAEo=
 golang.org/x/tools v0.0.0-20191125144606-a911d9008d1f/go.mod h1:b+2E5dAYhXwXZwtnZ6UAqBI28+e2cm9otk0dWdXHAEo=
 golang.org/x/tools v0.0.0-20191227053925-7b8e75db28f4/go.mod h1:TB2adYChydJhpapKDTa4BR/hXlZSLoq2Wpct/0txZ28=
 golang.org/x/tools v0.0.0-20200619180055-7c47624df98f/go.mod h1:EkVYQZoAsY45+roYkvgYkIh4xh/qjgUK9TdY2XT94GE=
 golang.org/x/tools v0.0.0-20210106214847-113979e3529a/go.mod h1:emZCQorbCU4vsT4fOWvOPXz4eW1wZW4PmDk9uLelYpA=
 golang.org/x/xerrors v0.0.0-20190717185122-a985d3407aa7/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 golang.org/x/xerrors v0.0.0-20191011141410-1b5146add898/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543 h1:E7g+9GITq07hpfrRu66IVDexMakfv52eLZ2CXBWiKr4=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1 h1:go1bK/D/BFZV2I8cIQd1NKEZ+0owSTG1fDTci4IqFcE=
 golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
@@ -388,8 +310,6 @@ google.golang.org/genproto v0.0.0-20190801165951-fa694d86fc64/go.mod h1:DMBHOl98
 google.golang.org/genproto v0.0.0-20190819201941-24fa4b261c55/go.mod h1:DMBHOl98Agz4BDEuKkezgsaosCRResVns1a3J2ZsMNc=
 google.golang.org/genproto v0.0.0-20190911173649-1774047e7e51/go.mod h1:IbNlFCBrqXvoKpeg0TB2l7cyZUmoaFKYIwrEpbDKLA8=
 google.golang.org/genproto v0.0.0-20191230161307-f3c370f40bfb/go.mod h1:n3cpQtvxv34hfy77yVDNjmbRyujviMdxYliBSkLhpCc=
 google.golang.org/genproto v0.0.0-20200423170343-7949de9c1215/go.mod h1:55QSHmfGQM9UVYDPBsyGGes0y52j32PQ3BqQfXhyH3c=
 google.golang.org/genproto v0.0.0-20200513103714-09dca8ec2884/go.mod h1:55QSHmfGQM9UVYDPBsyGGes0y52j32PQ3BqQfXhyH3c=
 google.golang.org/genproto v0.0.0-20200526211855-cb27e3aa2013 h1:+kGHl1aib/qcwaRi1CbqBZ1rk19r85MNUf8HaBghugY=
 google.golang.org/genproto v0.0.0-20200526211855-cb27e3aa2013/go.mod h1:NbSheEEYHJ7i3ixzK3sjbqSGDJWnxyFXZblF3eUsNvo=
 google.golang.org/grpc v1.25.1 h1:wdKvqQk7IttEw92GoRyKG2IDrUIpgpj6H6m81yfeMW0=
@@ -415,7 +335,6 @@ gopkg.in/inf.v0 v0.9.1/go.mod h1:cWUDdTG/fYaXco+Dcufb5Vnc6Gp2YChqWtbxRZE0mXw=
 gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7/go.mod h1:dt/ZhP58zS4L8KSrWDmTeBkI65Dw0HsyUHuEVlX15mw=
 gopkg.in/yaml.v2 v2.2.1/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI=
 gopkg.in/yaml.v2 v2.2.2/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI=
 gopkg.in/yaml.v2 v2.2.3/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI=
 gopkg.in/yaml.v2 v2.2.4/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI=
 gopkg.in/yaml.v2 v2.2.5/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI=
 gopkg.in/yaml.v2 v2.2.8 h1:obN1ZagJSUGI0Ek/LBmuj4SNLPfIny3KsKFopxRdj10=
@@ -444,5 +363,4 @@ k8s.io/utils v0.0.0-20210305010621-2afb4311ab10/go.mod h1:jPW/WVKK9YHAvNhRxK0md/
 rsc.io/binaryregexp v0.2.0/go.mod h1:qTv7/COck+e2FymRvadv62gMdZztPaShugOCi3I+8D8=
 sigs.k8s.io/structured-merge-diff/v4 v4.0.1/go.mod h1:bJZC9H9iH24zzfZ/41RGcq60oK1F7G282QMXDPYydCw=
 sigs.k8s.io/yaml v1.1.0/go.mod h1:UJmg0vDUVViEyp3mgSv9WPwZCDxu4rQW1olrI1uml+o=
 sigs.k8s.io/yaml v1.2.0 h1:kr/MCeFWJWTwyaHoR9c8EjH9OumOmoF9YGiZd7lFm/Q=
 sigs.k8s.io/yaml v1.2.0/go.mod h1:yfXDCHCao9+ENCvLSE62v9VSji2MKu5jeNfTrofGhJc=
--- a/csi/src/config.go
+++ b/csi/src/config.go
@@ -5,7 +5,7 @@ package vitastor
 const (
    vitastorCSIDriverName    = "csi.vitastor.io"
-    vitastorCSIDriverVersion = "0.8.4"
+    vitastorCSIDriverVersion = "0.8.9"
 )
 // Config struct fills the parameters of request or user input
--- a/csi/src/controllerserver.go
+++ b/csi/src/controllerserver.go
@@ -6,11 +6,11 @@ package vitastor
 import (
    "context"
    "encoding/json"
    "fmt"
    "strings"
    "bytes"
    "strconv"
    "time"
    "fmt"
    "os"
    "os/exec"
    "io/ioutil"
@@ -21,8 +21,6 @@ import (
    "google.golang.org/grpc/codes"
    "google.golang.org/grpc/status"
    "go.etcd.io/etcd/clientv3"
    "github.com/container-storage-interface/spec/lib/go/csi"
 )
@@ -114,6 +112,34 @@ func GetConnectionParams(params map[string]string) (map[string]string, []string,
    return ctxVars, etcdUrl, etcdPrefix
 }
 func invokeCLI(ctxVars map[string]string, args []string) ([]byte, error)
 {
    if (ctxVars["etcdUrl"] != "")
    {
        args = append(args, "--etcd_address", ctxVars["etcdUrl"])
    }
    if (ctxVars["etcdPrefix"] != "")
    {
        args = append(args, "--etcd_prefix", ctxVars["etcdPrefix"])
    }
    if (ctxVars["configPath"] != "")
    {
        args = append(args, "--config_path", ctxVars["configPath"])
    }
    c := exec.Command("/usr/bin/vitastor-cli", args...)
    var stdout, stderr bytes.Buffer
    c.Stdout = &stdout
    c.Stderr = &stderr
    err := c.Run()
    stderrStr := string(stderr.Bytes())
    if (err != nil)
    {
        klog.Errorf("vitastor-cli %s failed: %s, status %s\n", strings.Join(args, " "), stderrStr, err)
        return nil, status.Error(codes.Internal, stderrStr+" (status "+err.Error()+")")
    }
    return stdout.Bytes(), nil
 }
 // Create the volume
 func (cs *ControllerServer) CreateVolume(ctx context.Context, req *csi.CreateVolumeRequest) (*csi.CreateVolumeResponse, error)
 {
@@ -146,128 +172,41 @@ func (cs *ControllerServer) CreateVolume(ctx context.Context, req *csi.CreateVol
        volSize = ((capRange.GetRequiredBytes() + MB - 1) / MB) * MB
    }
-    // FIXME: The following should PROBABLY be implemented externally in a management tool
+    ctxVars, etcdUrl, _ := GetConnectionParams(req.Parameters)
    ctxVars, etcdUrl, etcdPrefix := GetConnectionParams(req.Parameters)
    if (len(etcdUrl) == 0)
    {
        return nil, status.Error(codes.InvalidArgument, "no etcdUrl in storage class configuration and no etcd_address in vitastor.conf")
    }
-    // Connect to etcd
+    // Create image using vitastor-cli
-    cli, err := clientv3.New(clientv3.Config{
+    _, err := invokeCLI(ctxVars, []string{ "create", volName, "-s", fmt.Sprintf("%v", volSize), "--pool", fmt.Sprintf("%v", poolId) })
        DialTimeout: ETCD_TIMEOUT,
        Endpoints: etcdUrl,
    })
    if (err != nil)
    {
-        return nil, status.Error(codes.Internal, "failed to connect to etcd at "+strings.Join(etcdUrl, ",")+": "+err.Error())
+        if (strings.Index(err.Error(), "already exists") > 0)
    }
    defer cli.Close()
    var imageId uint64 = 0
    for
    {
        // Check if the image exists
        ctx, cancel := context.WithTimeout(context.Background(), ETCD_TIMEOUT)
        resp, err := cli.Get(ctx, etcdPrefix+"/index/image/"+volName)
        cancel()
        if (err != nil)
        {
-            return nil, status.Error(codes.Internal, "failed to read key from etcd: "+err.Error())
+            stat, err := invokeCLI(ctxVars, []string{ "ls", "--json", volName })
        }
        if (len(resp.Kvs) > 0)
        {
            kv := resp.Kvs[0]
            var v InodeIndex
            err := json.Unmarshal(kv.Value, &v)
            if (err != nil)
            {
-                return nil, status.Error(codes.Internal, "invalid /index/image/"+volName+" key in etcd: "+err.Error())
+                return nil, err
            }
-            poolId = v.PoolId
+            var inodeCfg []InodeConfig
-            imageId = v.Id
+            err = json.Unmarshal(stat, &inodeCfg)
            inodeCfgKey := fmt.Sprintf("/config/inode/%d/%d", poolId, imageId)
            ctx, cancel := context.WithTimeout(context.Background(), ETCD_TIMEOUT)
            resp, err := cli.Get(ctx, etcdPrefix+inodeCfgKey)
            cancel()
            if (err != nil)
            {
-                return nil, status.Error(codes.Internal, "failed to read key from etcd: "+err.Error())
+                return nil, status.Error(codes.Internal, "Invalid JSON in vitastor-cli ls: "+err.Error())
            }
-            if (len(resp.Kvs) == 0)
+            if (len(inodeCfg) == 0)
            {
-                return nil, status.Error(codes.Internal, "missing "+inodeCfgKey+" key in etcd")
+                return nil, status.Error(codes.Internal, "vitastor-cli create said that image already exists, but ls can't find it")
            }
-            var inodeCfg InodeConfig
+            if (inodeCfg[0].Size < uint64(volSize))
            err = json.Unmarshal(resp.Kvs[0].Value, &inodeCfg)
            if (err != nil)
            {
                return nil, status.Error(codes.Internal, "invalid "+inodeCfgKey+" key in etcd: "+err.Error())
            }
            if (inodeCfg.Size < uint64(volSize))
            {
                return nil, status.Error(codes.Internal, "image "+volName+" is already created, but size is less than expected")
            }
        }
        else
        {
-            // Find a free ID
+            return nil, err
            // Create image metadata in a transaction verifying that the image doesn't exist yet AND ID is still free
            maxIdKey := fmt.Sprintf("%s/index/maxid/%d", etcdPrefix, poolId)
            ctx, cancel := context.WithTimeout(context.Background(), ETCD_TIMEOUT)
            resp, err := cli.Get(ctx, maxIdKey)
            cancel()
            if (err != nil)
            {
                return nil, status.Error(codes.Internal, "failed to read key from etcd: "+err.Error())
            }
            var modRev int64
            var nextId uint64
            if (len(resp.Kvs) > 0)
            {
                var err error
                nextId, err = strconv.ParseUint(string(resp.Kvs[0].Value), 10, 64)
                if (err != nil)
                {
                    return nil, status.Error(codes.Internal, maxIdKey+" contains invalid ID")
                }
                modRev = resp.Kvs[0].ModRevision
                nextId++
            }
            else
            {
                nextId = 1
            }
            inodeIdxJson, _ := json.Marshal(InodeIndex{
                Id: nextId,
                PoolId: poolId,
            })
            inodeCfgJson, _ := json.Marshal(InodeConfig{
                Name: volName,
                Size: uint64(volSize),
            })
            ctx, cancel = context.WithTimeout(context.Background(), ETCD_TIMEOUT)
            txnResp, err := cli.Txn(ctx).If(
                clientv3.Compare(clientv3.ModRevision(fmt.Sprintf("%s/index/maxid/%d", etcdPrefix, poolId)), "=", modRev),
                clientv3.Compare(clientv3.CreateRevision(fmt.Sprintf("%s/index/image/%s", etcdPrefix, volName)), "=", 0),
                clientv3.Compare(clientv3.CreateRevision(fmt.Sprintf("%s/config/inode/%d/%d", etcdPrefix, poolId, nextId)), "=", 0),
            ).Then(
                clientv3.OpPut(fmt.Sprintf("%s/index/maxid/%d", etcdPrefix, poolId), fmt.Sprintf("%d", nextId)),
                clientv3.OpPut(fmt.Sprintf("%s/index/image/%s", etcdPrefix, volName), string(inodeIdxJson)),
                clientv3.OpPut(fmt.Sprintf("%s/config/inode/%d/%d", etcdPrefix, poolId, nextId), string(inodeCfgJson)),
            ).Commit()
            cancel()
            if (err != nil)
            {
                return nil, status.Error(codes.Internal, "failed to commit transaction in etcd: "+err.Error())
            }
            if (txnResp.Succeeded)
            {
                imageId = nextId
                break
            }
            // Start over if the transaction fails
        }
    }
@@ -299,97 +238,12 @@ func (cs *ControllerServer) DeleteVolume(ctx context.Context, req *csi.DeleteVol
    }
    volName := ctxVars["name"]
-    _, etcdUrl, etcdPrefix := GetConnectionParams(ctxVars)
+    ctxVars, _, _ = GetConnectionParams(ctxVars)
    if (len(etcdUrl) == 0)
    {
        return nil, status.Error(codes.InvalidArgument, "no etcdUrl in storage class configuration and no etcd_address in vitastor.conf")
    }
-    cli, err := clientv3.New(clientv3.Config{
+    _, err = invokeCLI(ctxVars, []string{ "rm", volName })
        DialTimeout: ETCD_TIMEOUT,
        Endpoints: etcdUrl,
    })
    if (err != nil)
    {
-        return nil, status.Error(codes.Internal, "failed to connect to etcd at "+strings.Join(etcdUrl, ",")+": "+err.Error())
+        return nil, err
    }
    defer cli.Close()
    // Find inode by name
    ctx, cancel := context.WithTimeout(context.Background(), ETCD_TIMEOUT)
    resp, err := cli.Get(ctx, etcdPrefix+"/index/image/"+volName)
    cancel()
    if (err != nil)
    {
        return nil, status.Error(codes.Internal, "failed to read key from etcd: "+err.Error())
    }
    if (len(resp.Kvs) == 0)
    {
        return nil, status.Error(codes.NotFound, "volume "+volName+" does not exist")
    }
    var idx InodeIndex
    err = json.Unmarshal(resp.Kvs[0].Value, &idx)
    if (err != nil)
    {
        return nil, status.Error(codes.Internal, "invalid /index/image/"+volName+" key in etcd: "+err.Error())
    }
    // Get inode config
    inodeCfgKey := fmt.Sprintf("%s/config/inode/%d/%d", etcdPrefix, idx.PoolId, idx.Id)
    ctx, cancel = context.WithTimeout(context.Background(), ETCD_TIMEOUT)
    resp, err = cli.Get(ctx, inodeCfgKey)
    cancel()
    if (err != nil)
    {
        return nil, status.Error(codes.Internal, "failed to read key from etcd: "+err.Error())
    }
    if (len(resp.Kvs) == 0)
    {
        return nil, status.Error(codes.NotFound, "volume "+volName+" does not exist")
    }
    var inodeCfg InodeConfig
    err = json.Unmarshal(resp.Kvs[0].Value, &inodeCfg)
    if (err != nil)
    {
        return nil, status.Error(codes.Internal, "invalid "+inodeCfgKey+" key in etcd: "+err.Error())
    }
    // Delete inode data by invoking vitastor-cli
    args := []string{
        "rm-data", "--etcd_address", strings.Join(etcdUrl, ","),
        "--pool", fmt.Sprintf("%d", idx.PoolId),
        "--inode", fmt.Sprintf("%d", idx.Id),
    }
    if (ctxVars["configPath"] != "")
    {
        args = append(args, "--config_path", ctxVars["configPath"])
    }
    c := exec.Command("/usr/bin/vitastor-cli", args...)
    var stderr bytes.Buffer
    c.Stdout = nil
    c.Stderr = &stderr
    err = c.Run()
    stderrStr := string(stderr.Bytes())
    if (err != nil)
    {
        klog.Errorf("vitastor-cli rm-data failed: %s, status %s\n", stderrStr, err)
        return nil, status.Error(codes.Internal, stderrStr+" (status "+err.Error()+")")
    }
    // Delete inode config in etcd
    ctx, cancel = context.WithTimeout(context.Background(), ETCD_TIMEOUT)
    txnResp, err := cli.Txn(ctx).Then(
        clientv3.OpDelete(fmt.Sprintf("%s/index/image/%s", etcdPrefix, volName)),
        clientv3.OpDelete(fmt.Sprintf("%s/config/inode/%d/%d", etcdPrefix, idx.PoolId, idx.Id)),
    ).Commit()
    cancel()
    if (err != nil)
    {
        return nil, status.Error(codes.Internal, "failed to delete keys in etcd: "+err.Error())
    }
    if (!txnResp.Succeeded)
    {
        return nil, status.Error(codes.Internal, "failed to delete keys in etcd: transaction failed")
    }
    return &csi.DeleteVolumeResponse{}, nil
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,10 +1,10 @@
-vitastor (0.8.4-1) unstable; urgency=medium
+vitastor (0.8.9-1) unstable; urgency=medium
  * Bugfixes
 -- Vitaliy Filippov <vitalif@yourcmc.ru>  Fri, 03 Jun 2022 02:09:44 +0300
-vitastor (0.8.4-1) unstable; urgency=medium
+vitastor (0.8.9-1) unstable; urgency=medium
  * Implement NFS proxy
  * Add documentation
--- a/debian/vitastor.Dockerfile
+++ b/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-0.8.4; \
+    cp -r /root/vitastor vitastor-0.8.9; \
-    cd vitastor-0.8.4; \
+    cd vitastor-0.8.9; \
    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-$REL; \
-    tar --sort=name --mtime='2020-01-01' --owner=0 --group=0 --exclude=debian -cJf vitastor_0.8.4.orig.tar.xz vitastor-0.8.4; \
+    tar --sort=name --mtime='2020-01-01' --owner=0 --group=0 --exclude=debian -cJf vitastor_0.8.9.orig.tar.xz vitastor-0.8.9; \
-    cd vitastor-0.8.4; \
+    cd vitastor-0.8.9; \
    V=$(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; \
--- a/docker/vitastor.gpg
+++ b/docker/vitastor.gpg
--- a/docs/config.en.md
+++ b/docs/config.en.md
@@ -17,14 +17,16 @@ Configuration parameters can be set in 3 places:
 - Configuration file (`/etc/vitastor/vitastor.conf` or other path)
 - etcd key `/vitastor/config/global`. Most variables can be set there, but etcd
  connection parameters should obviously be set in the configuration file.
- Command line of Vitastor components: OSD, mon, fio and QEMU options,
+- Command line of Vitastor components: OSD (when you run it without vitastor-disk),
-  OpenStack/Proxmox/etc configuration. The latter doesn't allow to set all
+  mon, fio and QEMU options, OpenStack/Proxmox/etc configuration. The latter
-  variables directly, but it allows to override the configuration file and
+  doesn't allow to set all variables directly, but it allows to override the
-  set everything you need inside it.
+  configuration file and set everything you need inside it.
 - OSD superblocks created by [vitastor-disk](../usage/disk.en.md) contain
  primarily disk layout parameters of specific OSDs. In fact, these parameters
  are automatically passed into the command line of vitastor-osd process, so
  they have the same "status" as command-line parameters.
 In the future, additional configuration methods may be added:
 - OSD superblock which will, by design, contain parameters related to the disk
  layout and to one specific OSD.
 - OSD-specific keys in etcd like `/vitastor/config/osd/<number>`.
 ## Parameter Reference
--- a/docs/config.ru.md
+++ b/docs/config.ru.md
@@ -19,14 +19,17 @@
 - Ключе в etcd `/vitastor/config/global`. Большая часть параметров может
  задаваться там, кроме, естественно, самих параметров соединения с etcd,
  которые должны задаваться в файле конфигурации
- В командной строке компонентов Vitastor: OSD, монитора, опциях fio и QEMU,
+- В командной строке компонентов Vitastor: OSD (при ручном запуске без vitastor-disk),
-  настроек OpenStack, Proxmox и т.п. Последние, как правило, не включают полный
+  монитора, опциях fio и QEMU, настроек OpenStack, Proxmox и т.п. Последние,
-  набор параметров напрямую, но разрешают определить путь к файлу конфигурации
+  как правило, не включают полный набор параметров напрямую, но позволяют
-  и задать любые параметры в нём.
+  определить путь к файлу конфигурации и задать любые параметры в нём.
 - В суперблоке OSD, записываемом [vitastor-disk](../usage/disk.ru.md) - параметры,
  связанные с дисковым форматом и с этим конкретным OSD. На самом деле,
  при запуске OSD эти параметры автоматически передаются в командную строку
  процесса vitastor-osd, то есть по "статусу" они эквивалентны параметрам
  командной строки OSD.
 В будущем также могут быть добавлены другие способы конфигурации:
 - Суперблок OSD, в котором будут храниться параметры OSD, связанные с дисковым
  форматом и с этим конкретным OSD.
 - OSD-специфичные ключи в etcd типа `/vitastor/config/osd/<номер>`.
 ## Список параметров
--- a/docs/config/network.en.md
+++ b/docs/config/network.en.md
@@ -19,6 +19,7 @@ between clients, OSDs and etcd.
 - [rdma_max_sge](#rdma_max_sge)
 - [rdma_max_msg](#rdma_max_msg)
 - [rdma_max_recv](#rdma_max_recv)
 - [rdma_max_send](#rdma_max_send)
 - [peer_connect_interval](#peer_connect_interval)
 - [peer_connect_timeout](#peer_connect_timeout)
 - [osd_idle_timeout](#osd_idle_timeout)
@@ -74,6 +75,12 @@ to work. For example, Mellanox ConnectX-3 and older adapters don't have
 Implicit ODP, so they're unsupported by Vitastor. Run `ibv_devinfo -v` as
 root to list available RDMA devices and their features.
 Remember that you also have to configure your network switches if you use
 RoCE/RoCEv2, otherwise you may experience unstable performance. Refer to
 the manual of your network vendor for details about setting up the switch
 for RoCEv2 correctly. Usually it means setting up Lossless Ethernet with
 PFC (Priority Flow Control) and ECN (Explicit Congestion Notification).
 ## rdma_port_num
 - Type: integer
@@ -116,20 +123,30 @@ required to change this parameter.
 ## rdma_max_msg
 - Type: integer
- Default: 1048576
+- Default: 132096
 Maximum size of a single RDMA send or receive operation in bytes.
 ## rdma_max_recv
 - Type: integer
 - Default: 16
 Maximum number of RDMA receive buffers per connection (RDMA requires
 preallocated buffers to receive data). Each buffer is `rdma_max_msg` bytes
 in size. So this setting directly affects memory usage: a single Vitastor
 RDMA client uses `rdma_max_recv * rdma_max_msg * OSD_COUNT` bytes of memory.
 Default is roughly 2 MB * number of OSDs.
 ## rdma_max_send
 - Type: integer
 - Default: 8
-Maximum number of parallel RDMA receive operations. Note that this number
+Maximum number of outstanding RDMA send operations per connection. Should be
-of receive buffers `rdma_max_msg` in size are allocated for each client,
+less than `rdma_max_recv` so the receiving side doesn't run out of buffers.
-so this setting actually affects memory usage. This is because RDMA receive
+Doesn't affect memory usage - additional memory isn't allocated for send
-operations are (sadly) still not zero-copy in Vitastor. It may be fixed in
+operations.
 later versions.
 ## peer_connect_interval
--- a/docs/config/network.ru.md
+++ b/docs/config/network.ru.md
@@ -19,6 +19,7 @@
 - [rdma_max_sge](#rdma_max_sge)
 - [rdma_max_msg](#rdma_max_msg)
 - [rdma_max_recv](#rdma_max_recv)
 - [rdma_max_send](#rdma_max_send)
 - [peer_connect_interval](#peer_connect_interval)
 - [peer_connect_timeout](#peer_connect_timeout)
 - [osd_idle_timeout](#osd_idle_timeout)
@@ -78,6 +79,13 @@ Implicit On-Demand Paging (Implicit ODP) и Scatter/Gather (SG). Наприме
 суперпользователя, чтобы посмотреть список доступных RDMA-устройств, их
 параметры и возможности.
 Обратите внимание, что если вы используете RoCE/RoCEv2, вам также необходимо
 правильно настроить для него коммутаторы, иначе вы можете столкнуться с
 нестабильной производительностью. Подробную информацию о настройке
 коммутатора для RoCEv2 ищите в документации производителя. Обычно это
 подразумевает настройку сети без потерь на основе PFC (Priority Flow
 Control) и ECN (Explicit Congestion Notification).
 ## rdma_port_num
 - Тип: целое число
@@ -121,22 +129,32 @@ OSD в любом случае согласовывают реальное зн
 ## rdma_max_msg
 - Тип: целое число
- Значение по умолчанию: 1048576
+- Значение по умолчанию: 132096
 Максимальный размер одной RDMA-операции отправки или приёма.
 ## rdma_max_recv
 - Тип: целое число
 - Значение по умолчанию: 16
 Максимальное число буферов для RDMA-приёма данных на одно соединение
 (RDMA требует заранее выделенных буферов для приёма данных). Каждый буфер
 имеет размер `rdma_max_msg` байт. Таким образом, настройка прямо влияет на
 потребление памяти - один Vitastor-клиент с RDMA использует
 `rdma_max_recv * rdma_max_msg * ЧИСЛО_OSD` байт памяти, по умолчанию -
 примерно 2 МБ * число OSD.
 ## rdma_max_send
 - Тип: целое число
 - Значение по умолчанию: 8
-Максимальное число параллельных RDMA-операций получения данных. Следует
+Максимальное число RDMA-операций отправки, отправляемых в очередь одного
-иметь в виду, что данное число буферов размером `rdma_max_msg` выделяется
+соединения. Желательно, чтобы оно было меньше `rdma_max_recv`, чтобы
-для каждого подключённого клиентского соединения, так что данная настройка
+у принимающей стороны в процессе работы не заканчивались буферы на приём.
-влияет на потребление памяти. Это так потому, что RDMA-приём данных в
+Не влияет на потребление памяти - дополнительная память на операции отправки
-Vitastor, увы, всё равно не является zero-copy, т.е. всё равно 1 раз
+не выделяется.
 копирует данные в памяти. Данная особенность, возможно, будет исправлена в
 более новых версиях Vitastor.
 ## peer_connect_interval
--- a/docs/config/src/network.yml
+++ b/docs/config/src/network.yml
@@ -53,6 +53,12 @@
    to work. For example, Mellanox ConnectX-3 and older adapters don't have
    Implicit ODP, so they're unsupported by Vitastor. Run `ibv_devinfo -v` as
    root to list available RDMA devices and their features.
    Remember that you also have to configure your network switches if you use
    RoCE/RoCEv2, otherwise you may experience unstable performance. Refer to
    the manual of your network vendor for details about setting up the switch
    for RoCEv2 correctly. Usually it means setting up Lossless Ethernet with
    PFC (Priority Flow Control) and ECN (Explicit Congestion Notification).
  info_ru: |
    Название RDMA-устройства для связи с Vitastor OSD (например, "rocep5s0f0").
    Имейте в виду, что поддержка RDMA в Vitastor требует функций устройства
@@ -61,6 +67,13 @@
    потому не поддерживаются в Vitastor. Запустите `ibv_devinfo -v` от имени
    суперпользователя, чтобы посмотреть список доступных RDMA-устройств, их
    параметры и возможности.
    Обратите внимание, что если вы используете RoCE/RoCEv2, вам также необходимо
    правильно настроить для него коммутаторы, иначе вы можете столкнуться с
    нестабильной производительностью. Подробную информацию о настройке
    коммутатора для RoCEv2 ищите в документации производителя. Обычно это
    подразумевает настройку сети без потерь на основе PFC (Priority Flow
    Control) и ECN (Explicit Congestion Notification).
 - name: rdma_port_num
  type: int
  default: 1
@@ -114,26 +127,39 @@
    так что менять этот параметр обычно не нужно.
 - name: rdma_max_msg
  type: int
-  default: 1048576
+  default: 132096
  info: Maximum size of a single RDMA send or receive operation in bytes.
  info_ru: Максимальный размер одной RDMA-операции отправки или приёма.
 - name: rdma_max_recv
  type: int
  default: 16
  info: |
    Maximum number of RDMA receive buffers per connection (RDMA requires
    preallocated buffers to receive data). Each buffer is `rdma_max_msg` bytes
    in size. So this setting directly affects memory usage: a single Vitastor
    RDMA client uses `rdma_max_recv * rdma_max_msg * OSD_COUNT` bytes of memory.
    Default is roughly 2 MB * number of OSDs.
  info_ru: |
    Максимальное число буферов для RDMA-приёма данных на одно соединение
    (RDMA требует заранее выделенных буферов для приёма данных). Каждый буфер
    имеет размер `rdma_max_msg` байт. Таким образом, настройка прямо влияет на
    потребление памяти - один Vitastor-клиент с RDMA использует
    `rdma_max_recv * rdma_max_msg * ЧИСЛО_OSD` байт памяти, по умолчанию -
    примерно 2 МБ * число OSD.
 - name: rdma_max_send
  type: int
  default: 8
  info: |
-    Maximum number of parallel RDMA receive operations. Note that this number
+    Maximum number of outstanding RDMA send operations per connection. Should be
-    of receive buffers `rdma_max_msg` in size are allocated for each client,
+    less than `rdma_max_recv` so the receiving side doesn't run out of buffers.
-    so this setting actually affects memory usage. This is because RDMA receive
+    Doesn't affect memory usage - additional memory isn't allocated for send
-    operations are (sadly) still not zero-copy in Vitastor. It may be fixed in
+    operations.
    later versions.
  info_ru: |
-    Максимальное число параллельных RDMA-операций получения данных. Следует
+    Максимальное число RDMA-операций отправки, отправляемых в очередь одного
-    иметь в виду, что данное число буферов размером `rdma_max_msg` выделяется
+    соединения. Желательно, чтобы оно было меньше `rdma_max_recv`, чтобы
-    для каждого подключённого клиентского соединения, так что данная настройка
+    у принимающей стороны в процессе работы не заканчивались буферы на приём.
-    влияет на потребление памяти. Это так потому, что RDMA-приём данных в
+    Не влияет на потребление памяти - дополнительная память на операции отправки
-    Vitastor, увы, всё равно не является zero-copy, т.е. всё равно 1 раз
+    не выделяется.
    копирует данные в памяти. Данная особенность, возможно, будет исправлена в
    более новых версиях Vitastor.
 - name: peer_connect_interval
  type: sec
  min: 1
--- a/docs/installation/packages.en.md
+++ b/docs/installation/packages.en.md
@@ -20,15 +20,18 @@
 ## CentOS
 - Add Vitastor package repository:
-  - CentOS 7: `yum install https://vitastor.io/rpms/centos/7/vitastor-release-1.0-1.el7.noarch.rpm`
+  - CentOS 7: `yum install https://vitastor.io/rpms/centos/7/vitastor-release.rpm`
-  - CentOS 8: `dnf install https://vitastor.io/rpms/centos/8/vitastor-release-1.0-1.el8.noarch.rpm`
+  - CentOS 8: `dnf install https://vitastor.io/rpms/centos/8/vitastor-release.rpm`
  - AlmaLinux 9 and other RHEL 9 clones (Rocky, Oracle...): `dnf install https://vitastor.io/rpms/centos/9/vitastor-release.rpm`
 - Enable EPEL: `yum/dnf install epel-release`
 - Enable additional CentOS repositories:
  - CentOS 7: `yum install centos-release-scl`
  - CentOS 8: `dnf install centos-release-advanced-virtualization`
  - RHEL 9 clones: not required
 - Enable elrepo-kernel:
  - CentOS 7: `yum install https://www.elrepo.org/elrepo-release-7.el7.elrepo.noarch.rpm`
  - CentOS 8: `dnf install https://www.elrepo.org/elrepo-release-8.el8.elrepo.noarch.rpm`
  - RHEL 9 clones: `dnf install https://www.elrepo.org/elrepo-release-9.el9.elrepo.noarch.rpm`
 - Install packages: `yum/dnf install vitastor lpsolve etcd kernel-ml qemu-kvm`
 ## Installation requirements
--- a/docs/installation/packages.ru.md
+++ b/docs/installation/packages.ru.md
@@ -20,15 +20,18 @@
 ## CentOS
 - Добавьте в систему репозиторий Vitastor:
-  - CentOS 7: `yum install https://vitastor.io/rpms/centos/7/vitastor-release-1.0-1.el7.noarch.rpm`
+  - CentOS 7: `yum install https://vitastor.io/rpms/centos/7/vitastor-release.rpm`
-  - CentOS 8: `dnf install https://vitastor.io/rpms/centos/8/vitastor-release-1.0-1.el8.noarch.rpm`
+  - CentOS 8: `dnf install https://vitastor.io/rpms/centos/8/vitastor-release.rpm`
  - AlmaLinux 9 и другие клоны RHEL 9 (Rocky, Oracle...): `dnf install https://vitastor.io/rpms/centos/9/vitastor-release.rpm`
 - Включите EPEL: `yum/dnf install epel-release`
 - Включите дополнительные репозитории CentOS:
  - CentOS 7: `yum install centos-release-scl`
  - CentOS 8: `dnf install centos-release-advanced-virtualization`
  - Клоны RHEL 9: не нужно
 - Включите elrepo-kernel:
  - CentOS 7: `yum install https://www.elrepo.org/elrepo-release-7.el7.elrepo.noarch.rpm`
  - CentOS 8: `dnf install https://www.elrepo.org/elrepo-release-8.el8.elrepo.noarch.rpm`
  - Клоны RHEL 9: `dnf install https://www.elrepo.org/elrepo-release-9.el9.elrepo.noarch.rpm`
 - Установите пакеты: `yum/dnf install vitastor lpsolve etcd kernel-ml qemu-kvm`
 ## Установочные требования
--- a/docs/installation/proxmox.en.md
+++ b/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 (6.4-7.4 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
+  buster for 6.4, bullseye for 7.4, pve7.1 for 7.1, pve7.2 for 7.2, pve7.3 for 7.3
 - 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),
--- a/docs/installation/proxmox.ru.md
+++ b/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 6.4-7.4):
 - 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
+  buster для 6.4, bullseye для 7.4, pve7.1 для 7.1, pve7.2 для 7.2, pve7.3 для 7.3
 - 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
--- a/docs/intro/quickstart.en.md
+++ b/docs/intro/quickstart.en.md
@@ -45,7 +45,9 @@ On the monitor hosts:
  }
  ```
 - Initialize OSDs:
-  - SSD-only: `vitastor-disk prepare /dev/sdXXX [/dev/sdYYY ...]`
+  - SSD-only: `vitastor-disk prepare /dev/sdXXX [/dev/sdYYY ...]`. You can add
    `--disable_data_fsync off` to leave disk cache enabled if you use desktop
    SSDs without capacitors.
  - Hybrid, SSD+HDD: `vitastor-disk prepare --hybrid /dev/sdXXX [/dev/sdYYY ...]`.
    Pass all your devices (HDD and SSD) to this script &mdash; 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
@@ -53,7 +55,9 @@ On the monitor hosts:
    but some free unpartitioned space must be available because the script creates new partitions for journals.
 - You can change OSD configuration in units or in `vitastor.conf`.
  Check [Configuration Reference](../config.en.md) for parameter descriptions.
- If all your drives have capacitors, create global configuration in etcd: \
+- If all your drives have capacitors, and even if not, but if you ran `vitastor-disk`
  without `--disable_data_fsync off` at the first step, then put the following
  setting into etcd: \
  `etcdctl --endpoints=... put /vitastor/config/global '{"immediate_commit":"all"}'`
 - Start all OSDs: `systemctl start vitastor.target`
@@ -70,11 +74,15 @@ 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.
 If you use HDDs you should also add `"block_size": 1048576` to pool configuration.
 The other option is to add it into /vitastor/config/global, in this case it will
 apply to all pools by default.
 ## Check cluster status
 `vitastor-cli status`
--- a/docs/intro/quickstart.ru.md
+++ b/docs/intro/quickstart.ru.md
@@ -45,7 +45,9 @@
  }
  ```
 - Инициализуйте OSD:
-  - SSD: `vitastor-disk prepare /dev/sdXXX [/dev/sdYYY ...]`
+  - SSD: `vitastor-disk prepare /dev/sdXXX [/dev/sdYYY ...]`. Если вы используете
    десктопные SSD без конденсаторов, можете оставить кэш включённым, добавив
    опцию `--disable_data_fsync off`.
  - 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
@@ -54,8 +56,11 @@
    для журналов, на SSD должно быть доступно свободное нераспределённое место.
 - Вы можете менять параметры OSD в юнитах systemd или в `vitastor.conf`. Описания параметров
  смотрите в [справке по конфигурации](../config.ru.md).
- Если все ваши диски - серверные с конденсаторами, пропишите это в глобальную конфигурацию в etcd: \
+- Если все ваши диски - серверные с конденсаторами, и даже если нет, но при этом
-  `etcdctl --endpoints=... put /vitastor/config/global '{"immediate_commit":"all"}'`
+  вы не добавляли опцию `--disable_data_fsync off` на первом шаге, а `vitastor-disk`
  не ругался на невозможность отключения кэша дисков, пропишите следующую настройку
  в глобальную конфигурацию в etcd: \
  `etcdctl --endpoints=... put /vitastor/config/global '{"immediate_commit":"all"}'`.
 - Запустите все OSD: `systemctl start vitastor.target`
 ## Создайте пул
@@ -71,11 +76,15 @@ 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.
 Если вы используете HDD-диски, то добавьте в конфигурацию пулов опцию `"block_size": 1048576`.
 Также эту опцию можно добавить в /vitastor/config/global, в этом случае она будет
 применяться ко всем пулам по умолчанию.
 ## Проверьте состояние кластера
 `vitastor-cli status`
--- a/docs/performance/theoretical.en.md
+++ b/docs/performance/theoretical.en.md
@@ -35,15 +35,24 @@ Write amplification for 4 KB blocks is usually 3-5 in Vitastor:
 If you manage to get an SSD which handles 512 byte blocks well (Optane?) you may
 lower 1, 3 and 4 to 512 bytes (1/8 of data size) and get WA as low as 2.375.
 Implemented NVDIMM support can basically eliminate WA at all - all extra writes will
 go to DRAM memory. But this requires a test cluster with NVDIMM - please contact me
 if you want to provide me with such cluster for tests.
 Lazy fsync also reduces WA for parallel workloads because journal blocks are only
 written when they fill up or fsync is requested.
 ## In Practice
-In practice, using tests from [Understanding Performance](understanding.en.md)
+In practice, using tests from [Understanding Performance](understanding.en.md), decent TCP network,
-and good server-grade SSD/NVMe drives, you should head for:
+good server-grade SSD/NVMe drives and disabled CPU power saving, you should head for:
 - At least 5000 T1Q1 replicated read and write iops (maximum 0.2ms latency)
 - At least 5000 T1Q1 EC read IOPS and at least 2200 EC write IOPS (maximum 0.45ms latency)
 - At least ~80k parallel read iops or ~30k write iops per 1 core (1 OSD)
 - Disk-speed or wire-speed linear reads and writes, whichever is the bottleneck in your case
 Lower results may mean that you have bad drives, bad network or some kind of misconfiguration.
 Current latency records:
 - 9668 T1Q1 replicated write iops (0.103 ms latency) with TCP and NVMe
 - 9143 T1Q1 replicated read iops (0.109 ms latency) with TCP and NVMe
--- a/docs/performance/theoretical.ru.md
+++ b/docs/performance/theoretical.ru.md
@@ -36,6 +36,25 @@ WA (мультипликатор записи) для 4 КБ блоков в Vit
 Если вы найдёте SSD, хорошо работающий с 512-байтными блоками данных (Optane?),
 то 1, 3 и 4 можно снизить до 512 байт (1/8 от размера данных) и получить WA всего 2.375.
 Если реализовать поддержку NVDIMM, то WA можно, условно говоря, ликвидировать вообще - все
 дополнительные операции записи смогут обслуживаться DRAM памятью. Но для этого необходим
 тестовый кластер с NVDIMM - пишите, если готовы предоставить такой для тестов.
 Кроме того, WA снижается при использовании отложенного/ленивого сброса при параллельной
 нагрузке, т.к. блоки журнала записываются на диск только когда они заполняются или явным
 образом запрашивается fsync.
 ## На практике
 На практике, используя тесты fio со страницы [Понимание сути производительности систем хранения](understanding.ru.md),
 нормальную TCP-сеть, хорошие серверные SSD/NVMe, при отключённом энергосбережении процессоров вы можете рассчитывать на:
 - От 5000 IOPS в 1 поток (T1Q1) и на чтение, и на запись при использовании репликации (задержка до 0.2мс)
 - От 5000 IOPS в 1 поток (T1Q1) на чтение и 2200 IOPS в 1 поток на запись при использовании EC (задержка до 0.45мс)
 - От 80000 IOPS на чтение в параллельном режиме на 1 ядро, от 30000 IOPS на запись на 1 ядро (на 1 OSD)
 - Скорость параллельного линейного чтения и записи, равная меньшему значению из скорости дисков или сети
 Худшие результаты означают, что у вас либо медленные диски, либо медленная сеть, либо что-то неправильно настроено.
 Зафиксированный на данный момент рекорд задержки:
 - 9668 IOPS (0.103 мс задержка) в 1 поток (T1Q1) на запись с TCP и NVMe при использовании репликации
 - 9143 IOPS (0.109 мс задержка) в 1 поток (T1Q1) на чтение с TCP и NVMe при использовании репликации
--- a/docs/usage/disk.en.md
+++ b/docs/usage/disk.en.md
@@ -1,4 +1,4 @@
-[Documentation](../../README.md#documentation) → Usage → Disk Tool
+[Documentation](../../README.md#documentation) → Usage → Disk management tool
 -----
--- a/docs/usage/disk.ru.md
+++ b/docs/usage/disk.ru.md
@@ -1,4 +1,4 @@
-[Документация](../../README-ru.md#документация) → Использование → Управление дисками
+[Документация](../../README-ru.md#документация) → Использование → Инструмент управления дисками
 -----
--- a/2
+++ b/2
--- a/mon/PGUtil.js
+++ b/mon/PGUtil.js
@@ -43,16 +43,16 @@ function finish_pg_history(merged_history)
    merged_history.all_peers = Object.values(merged_history.all_peers);
 }
-function scale_pg_count(prev_pgs, prev_pg_history, new_pg_history, new_pg_count)
+function scale_pg_count(prev_pgs, real_prev_pgs, prev_pg_history, new_pg_history, new_pg_count)
 {
-    const old_pg_count = prev_pgs.length;
+    const old_pg_count = real_prev_pgs.length;
    // Add all possibly intersecting PGs to the history of new PGs
    if (!(new_pg_count % old_pg_count))
    {
        // New PG count is a multiple of old PG count
        for (let i = 0; i < new_pg_count; i++)
        {
-            add_pg_history(new_pg_history, i, prev_pgs, prev_pg_history, i % old_pg_count);
+            add_pg_history(new_pg_history, i, real_prev_pgs, prev_pg_history, i % old_pg_count);
            finish_pg_history(new_pg_history[i]);
        }
    }
@@ -64,7 +64,7 @@ function scale_pg_count(prev_pgs, prev_pg_history, new_pg_history, new_pg_count)
        {
            for (let j = 0; j < mul; j++)
            {
-                add_pg_history(new_pg_history, i, prev_pgs, prev_pg_history, i+j*new_pg_count);
+                add_pg_history(new_pg_history, i, real_prev_pgs, prev_pg_history, i+j*new_pg_count);
            }
            finish_pg_history(new_pg_history[i]);
        }
@@ -76,7 +76,7 @@ function scale_pg_count(prev_pgs, prev_pg_history, new_pg_history, new_pg_count)
        let merged_history = {};
        for (let i = 0; i < old_pg_count; i++)
        {
-            add_pg_history(merged_history, 1, prev_pgs, prev_pg_history, i);
+            add_pg_history(merged_history, 1, real_prev_pgs, prev_pg_history, i);
        }
        finish_pg_history(merged_history[1]);
        for (let i = 0; i < new_pg_count; i++)
@@ -90,15 +90,15 @@ function scale_pg_count(prev_pgs, prev_pg_history, new_pg_history, new_pg_count)
        new_pg_history[i] = null;
    }
    // Just for the lp_solve optimizer - pick a "previous" PG for each "new" one
-    if (old_pg_count < new_pg_count)
+    if (prev_pgs.length < new_pg_count)
    {
-        for (let i = old_pg_count; i < new_pg_count; i++)
+        for (let i = prev_pgs.length; i < new_pg_count; i++)
        {
-            prev_pgs[i] = prev_pgs[i % old_pg_count];
+            prev_pgs[i] = prev_pgs[i % prev_pgs.length];
        }
    }
-    else if (old_pg_count > new_pg_count)
+    else if (prev_pgs.length > new_pg_count)
    {
-        prev_pgs.splice(new_pg_count, old_pg_count-new_pg_count);
+        prev_pgs.splice(new_pg_count, prev_pgs.length-new_pg_count);
    }
 }
--- a/mon/lp-optimizer.js
+++ b/mon/lp-optimizer.js
@@ -50,7 +50,8 @@ async function lp_solve(text)
    return { score, vars };
 }
-async function optimize_initial({ osd_tree, pg_count, pg_size = 3, pg_minsize = 2, max_combinations = 10000, parity_space = 1, ordered = false })
+async function optimize_initial({ osd_tree, pg_count, pg_size = 3, pg_minsize = 2, hier_sizes = null,
    max_combinations = 10000, parity_space = 1, ordered = false, seq_layout = false })
 {
    if (!pg_count || !osd_tree)
    {
@@ -58,7 +59,7 @@ async function optimize_initial({ osd_tree, pg_count, pg_size = 3, pg_minsize =
    }
    const all_weights = Object.assign({}, ...Object.values(osd_tree));
    const total_weight = Object.values(all_weights).reduce((a, c) => Number(a) + Number(c), 0);
-    const all_pgs = Object.values(random_combinations(osd_tree, pg_size, max_combinations, parity_space > 1));
+    const all_pgs = Object.values(random_hier_combinations(osd_tree, hier_sizes || [ pg_size, 1 ], max_combinations, parity_space > 1, seq_layout));
    const pg_per_osd = {};
    for (const pg of all_pgs)
    {
@@ -216,39 +217,45 @@ function calc_intersect_weights(old_pg_size, pg_size, pg_count, prev_weights, al
    return move_weights;
 }
-function add_valid_previous(osd_tree, prev_weights, all_pgs)
+function build_parent_per_leaf(osd_tree, res = {}, parents = [])
 {
    for (const item in osd_tree)
    {
        if (osd_tree[item] instanceof Object)
            build_parent_per_leaf(osd_tree[item], res, [ ...parents, item ]);
        else
            res[item] = parents;
    }
    return res;
 }
 function add_valid_previous(osd_tree, prev_weights, all_pgs, hier_sizes)
 {
    // Add previous combinations that are still valid
-    const hosts = Object.keys(osd_tree).sort();
+    const parent_per_osd = build_parent_per_leaf(osd_tree);
    const host_per_osd = {};
    for (const host in osd_tree)
    {
        for (const osd in osd_tree[host])
        {
            host_per_osd[osd] = host;
        }
    }
    skip_pg: for (const pg_name in prev_weights)
    {
-        const seen_hosts = {};
+        const seen = [];
        const pg = pg_name.substr(3).split(/_/);
        for (const osd of pg)
        {
-            if (!host_per_osd[osd] || seen_hosts[host_per_osd[osd]])
+            if (!parent_per_osd[osd])
            {
                continue skip_pg;
            for (let i = 0; i < parent_per_osd[osd].length; i++)
            {
                seen[parent_per_osd[osd][i]]++;
                if (seen[parent_per_osd[osd][i]] > hier_sizes[i])
                    continue skip_pg;
            }
            seen_hosts[host_per_osd[osd]] = true;
        }
        if (!all_pgs[pg_name])
        {
            all_pgs[pg_name] = pg;
        }
    }
 }
 // Try to minimize data movement
-async function optimize_change({ prev_pgs: prev_int_pgs, osd_tree, pg_size = 3, pg_minsize = 2, max_combinations = 10000, parity_space = 1, ordered = false })
+async function optimize_change({ prev_pgs: prev_int_pgs, osd_tree, pg_size = 3, pg_minsize = 2,
    hier_sizes = null, max_combinations = 10000, parity_space = 1, ordered = false, seq_layout = false })
 {
    if (!osd_tree)
    {
@@ -273,10 +280,10 @@ async function optimize_change({ prev_pgs: prev_int_pgs, osd_tree, pg_size = 3,
    }
    const old_pg_size = prev_int_pgs[0].length;
    // Get all combinations
-    let all_pgs = random_combinations(osd_tree, pg_size, max_combinations, parity_space > 1);
+    let all_pgs = random_hier_combinations(osd_tree, hier_sizes || [ pg_size, 1 ], max_combinations, parity_space > 1, seq_layout);
    if (old_pg_size == pg_size)
    {
-        add_valid_previous(osd_tree, prev_weights, all_pgs);
+        add_valid_previous(osd_tree, prev_weights, all_pgs, hier_sizes || [ pg_size, 1 ]);
    }
    all_pgs = Object.values(all_pgs);
    const pg_per_osd = {};
@@ -502,41 +509,147 @@ function put_aligned_pgs(aligned_pgs, int_pgs, prev_int_pgs, keygen)
    }
 }
-// Convert multi-level osd_tree = { level: number|string, id?: string, size?: number, children?: osd_tree }[]
+// Convert multi-level tree_node = { level: number|string, id?: string, size?: number, children?: tree_node[] }
 // levels = { string: number }
-// to a two-level osd_tree suitable for all_combinations()
+// to a multi-level OSD tree suitable for random_hier_combinations()
-function flatten_tree(osd_tree, levels, failure_domain_level, osd_level, domains = {}, i = { i: 1 })
+// (or in case of just 2 levels - for all_combinations() / random_combinations())
 //
 // Example:
 // tree_node = { level: 'dc', children: [ { level: 'rack', children: [ { level: 'host', children: [ { level: 'osd', size: 10 } ] } ] } ] }
 // extract_levels = [ 'rack', 'osd' ]
 // level_defs = { dc: 1, rack: 2, host: 3, osd: 4 }
 //
 // Result:
 // { rack0: { osd1: 10 } }
 function extract_tree_levels(tree_node, extract_levels, level_defs, new_tree = { idx: 1, items: {} })
 {
-    osd_level = levels[osd_level] || osd_level;
+    const next_level = Number(level_defs[extract_levels[0]] || extract_levels[0]) || 0;
-    failure_domain_level = levels[failure_domain_level] || failure_domain_level;
+    const level_name = level_defs[extract_levels[0]] ? extract_levels[0] : 'l'+extract_levels[0]+'_';
-    for (const node of osd_tree)
+    const is_leaf = extract_levels.length == 1;
    if ((level_defs[tree_node.level] || tree_node.level) >= next_level)
    {
-        if ((levels[node.level] || node.level) < failure_domain_level)
+        if (!is_leaf)
        {
-            flatten_tree(node.children||[], levels, failure_domain_level, osd_level, domains, i);
+            // Insert a (possibly fake) level
            const nt = { idx: 1, items: {} };
            new_tree.items[level_name+(new_tree.idx++)] = nt.items;
            extract_tree_levels(tree_node, extract_levels.slice(1), level_defs, nt);
        }
        else
        {
-            domains['dom'+(i.i++)] = extract_osds([ node ], levels, osd_level);
+            // Insert a leaf node
            const leaf_id = tree_node.id || (level_name+(new_tree.idx++));
            new_tree.items[leaf_id] = tree_node.size;
        }
    }
-    return domains;
+    else
    {
        for (const child_node of tree_node.children||[])
        {
            extract_tree_levels(child_node, extract_levels, level_defs, new_tree);
        }
    }
    return new_tree.items;
 }
-function extract_osds(osd_tree, levels, osd_level, osds = {})
+// generate random PGs with hierarchical failure domains, i.e. for example 3 DC each with 2 HOSTS
 // osd_tree = { level3_id: { level2_id: { level1_id: scalar_value } }, ... }
 //   osd_tree may contain arbitrary number of levels, but level count must be the same across the whole tree
 // size_per_level = number of items to select on each level, for example [3, 2, 1].
 //   must have the same number of items as the osd_tree level count.
 // count = PG count to generate
 // ordered = don't treat (x,y) and (y,x) as equal
 // seq_layout = true for the [DC1,DC1,DC2,DC2,DC3,DC3] layout, false for [DC1,DC2,DC3,DC1,DC2,DC3] layout
 function random_hier_combinations(osd_tree, size_per_level, count, ordered, seq_layout)
 {
-    for (const node of osd_tree)
+    let seed = 0x5f020e43;
    const rng = () =>
    {
-        if ((levels[node.level] || node.level) >= osd_level)
+        seed ^= seed << 13;
        seed ^= seed >> 17;
        seed ^= seed << 5;
        return seed + 2147483648;
    };
    const get_max_level = (o) =>
    {
        let lvl = 0;
        while (o instanceof Object)
        {
-            osds[node.id] = node.size;
+            for (const k in o)
            {
                lvl++;
                o = o[k];
                break;
            }
        }
-        else
+        return lvl;
    };
    const max_level = get_max_level(osd_tree);
    const gen_pg = (select) =>
    {
        let pg = [ osd_tree ];
        for (let level = 0; level < max_level; level++)
        {
-            extract_osds(node.children||[], levels, osd_level, osds);
+            let npg = [];
            for (let i = 0; i < pg.length; i++)
            {
                const keys = pg[i] instanceof Object ? Object.keys(pg[i]) : [];
                const max_keys = keys.length < size_per_level[level] ? keys.length : size_per_level[level];
                for (let j = 0; j < max_keys; j++)
                {
                    const r = select(level, i, j, (ordered ? keys.length : (keys.length - (max_keys - j - 1))));
                    const el = pg[i][keys[r]] instanceof Object ? pg[i][keys[r]] : keys[r];
                    npg[seq_layout ? i*size_per_level[level]+j : j*pg.length+i] = el;
                    keys.splice(ordered ? r : 0, ordered ? 1 : (r+1));
                }
                for (let j = max_keys; j < size_per_level[level]; j++)
                    npg[seq_layout ? i*size_per_level[level]+j : j*pg.length+i] = NO_OSD;
            }
            pg = npg;
        }
        return pg;
    };
    const r = {};
    // Generate random combinations including each OSD at least once
    let has_next = true;
    let ctr = [];
    while (has_next)
    {
        let pg = gen_pg((level, i, j, n) =>
        {
            if (i == 0 && j == 0)
            {
                // Select a pre-determined OSD in the first position on each level
                const r = ctr[level] == null || ctr[level][1] != n ? 0 : ctr[level][0];
                ctr[level] = [ r, n ];
                return r;
            }
            return rng() % n;
        });
        for (let i = ctr.length-1; i >= 0; i--)
        {
            ctr[i][0]++;
            if (ctr[i][0] < ctr[i][1])
                break;
            else
                ctr[i] = null;
        }
        has_next = ctr[0] != null;
        const cyclic_pgs = [ pg ];
        if (ordered)
            for (let i = 1; i < pg.size; i++)
                cyclic_pgs.push([ ...pg.slice(i), ...pg.slice(0, i) ]);
        for (const pg of cyclic_pgs)
            r['pg_'+pg.join('_')] = pg;
    }
-    return osds;
+    // Generate purely random combinations
    while (count > 0)
    {
        let pg = gen_pg((l, i, j, n) => rng() % n);
        r['pg_'+pg.join('_')] = pg;
        count--;
    }
    return r;
 }
 // ordered = don't treat (x,y) and (y,x) as equal
@@ -752,11 +865,12 @@ module.exports = {
    pg_weights_space_efficiency,
    pg_list_space_efficiency,
    pg_per_osd_space_efficiency,
-    flatten_tree,
+    extract_tree_levels,
    lp_solve,
    make_int_pgs,
    align_pgs,
    random_combinations,
    random_hier_combinations,
    all_combinations,
 };
--- a/mon/make-etcd
+++ b/mon/make-etcd
@@ -79,7 +79,7 @@ StartLimitInterval=0
 RestartSec=10
 [Install]
-WantedBy=local.target
+WantedBy=multi-user.target
 `);
    await system(`useradd etcd`);
    await system(`systemctl daemon-reload`);
--- a/mon/mon.js
+++ b/mon/mon.js
@@ -51,8 +51,9 @@ const etcd_tree = {
            // THIS IS JUST A POOR MAN'S CONFIG DOCUMENTATION
            // etcd connection
            config_path: "/etc/vitastor/vitastor.conf",
            etcd_address: "10.0.115.10:2379/v3",
            etcd_prefix: "/vitastor",
            // etcd connection - configurable online
            etcd_address: "10.0.115.10:2379/v3",
            // mon
            etcd_mon_ttl: 30, // min: 10
            etcd_mon_timeout: 1000, // ms. min: 0
@@ -70,14 +71,15 @@ const etcd_tree = {
            rdma_gid_index: 0,
            rdma_mtu: 4096,
            rdma_max_sge: 128,
-            rdma_max_send: 32,
+            rdma_max_send: 8,
-            rdma_max_recv: 8,
+            rdma_max_recv: 16,
-            rdma_max_msg: 1048576,
+            rdma_max_msg: 132096,
            log_level: 0,
            block_size: 131072,
            disk_alignment: 4096,
            bitmap_granularity: 4096,
            immediate_commit: false, // 'all' or 'small'
            // client and osd - configurable online
            log_level: 0,
            client_dirty_limit: 33554432,
            peer_connect_interval: 5, // seconds. min: 1
            peer_connect_timeout: 5, // seconds. min: 1
@@ -95,18 +97,19 @@ const etcd_tree = {
            osd_network: null, // "192.168.7.0/24" or an array of masks
            bind_address: "0.0.0.0",
            bind_port: 0,
            readonly: false,
            osd_memlock: false,
            // osd - configurable online
            autosync_interval: 5,
            autosync_writes: 128,
            client_queue_depth: 128, // unused
            recovery_queue_depth: 4,
            recovery_sync_batch: 16,
            readonly: false,
            no_recovery: false,
            no_rebalance: false,
            print_stats_interval: 3,
            slow_log_interval: 10,
            inode_vanish_time: 60,
            osd_memlock: false,
            // blockstore - fixed in superblock
            block_size,
            disk_alignment,
@@ -125,14 +128,15 @@ const etcd_tree = {
            meta_offset,
            disable_meta_fsync,
            disable_device_lock,
-            // blockstore - configurable
+            // blockstore - configurable offline
            max_write_iodepth,
            min_flusher_count: 1,
            max_flusher_count: 256,
            inmemory_metadata,
            inmemory_journal,
            journal_sector_buffer_count,
            journal_no_same_sector_overwrites,
            // blockstore - configurable online
            max_write_iodepth,
            min_flusher_count: 1,
            max_flusher_count: 256,
            throttle_small_writes: false,
            throttle_target_iops: 100,
            throttle_target_mbs: 100,
@@ -155,6 +159,10 @@ const etcd_tree = {
                // number of parity chunks, required for EC
                parity_chunks?: 1,
                pg_count: 100,
                // failure_domain = string | { string: int }
                // the second case specifies multiple failure domains. example:
                // { datacenter: 3, host: 2 } - means 3 datacenters with 2 hosts each, for EC 4+2
                // guarantees availability on outage of either 1 datacenter or 2 hosts
                failure_domain: 'host',
                max_osd_combinations: 10000,
                // block_size, bitmap_granularity, immediate_commit must match all OSDs used in that pool
@@ -261,7 +269,7 @@ const etcd_tree = {
            /* <pool_id>: {
                <pg_id>: {
                    primary: osd_num_t,
-                    state: ("starting"|"peering"|"peered"|"incomplete"|"active"|"repeering"|"stopping"|"offline"|
+                    state: ("starting"|"peering"|"incomplete"|"active"|"repeering"|"stopping"|"offline"|
                        "degraded"|"has_incomplete"|"has_degraded"|"has_misplaced"|"has_unclean"|
                        "has_invalid"|"left_on_dead")[],
                }
@@ -844,7 +852,7 @@ class Mon
        }
        for (const node_id in tree)
        {
-            if (node_id === '')
+            if (node_id === '' || tree[node_id].level === 'osd' && (!tree[node_id].size || tree[node_id].size <= 0))
            {
                continue;
            }
@@ -952,9 +960,9 @@ class Mon
        return alive_set[this.rng() % alive_set.length];
    }
-    save_new_pgs_txn(request, pool_id, up_osds, osd_tree, prev_pgs, new_pgs, pg_history)
+    save_new_pgs_txn(save_to, request, pool_id, up_osds, osd_tree, prev_pgs, new_pgs, pg_history)
    {
-        const aff_osds = this.get_affinity_osds(this.state.config.pools[pool_id], up_osds, osd_tree);
+        const aff_osds = this.get_affinity_osds(this.state.config.pools[pool_id] || {}, up_osds, osd_tree);
        const pg_items = {};
        this.reset_rng();
        new_pgs.map((osd_set, i) =>
@@ -1005,14 +1013,14 @@ class Mon
                });
            }
        }
-        this.state.config.pgs.items = this.state.config.pgs.items || {};
+        save_to.items = save_to.items || {};
        if (!new_pgs.length)
        {
-            delete this.state.config.pgs.items[pool_id];
+            delete save_to.items[pool_id];
        }
        else
        {
-            this.state.config.pgs.items[pool_id] = pg_items;
+            save_to.items[pool_id] = pg_items;
        }
    }
@@ -1023,6 +1031,32 @@ class Mon
        pool_cfg.parity_chunks = Math.floor(pool_cfg.parity_chunks) || undefined;
        pool_cfg.pg_count = Math.floor(pool_cfg.pg_count);
        pool_cfg.failure_domain = pool_cfg.failure_domain || 'host';
        if (pool_cfg.failure_domain instanceof Object)
        {
            for (const key in pool_cfg.failure_domain)
            {
                const cnt = parseInt(pool_cfg.failure_domain[key]);
                if (!cnt || cnt <= 0)
                {
                    if (warn)
                        console.log('Pool '+pool_id+' specifies invalid item count for failure domain \"'+key+'\"');
                    return false;
                }
                if (key !== 'host' && key != 'osd' && !(key in this.config.placement_levels||{}))
                {
                    if (warn)
                        console.log('Pool '+pool_id+' uses invalid failure domain \"'+key+'\"');
                    return false;
                }
            }
        }
        else if (pool_cfg.failure_domain !== 'host' && pool_cfg.failure_domain != 'osd' &&
            !(pool_cfg.failure_domain in this.config.placement_levels||{}))
        {
            if (warn)
                console.log('Pool '+pool_id+' uses invalid failure domain \"'+pool_cfg.failure_domain+'\"');
            return false;
        }
        pool_cfg.max_osd_combinations = Math.floor(pool_cfg.max_osd_combinations) || 10000;
        if (!/^[1-9]\d*$/.exec(''+pool_id))
        {
@@ -1108,27 +1142,23 @@ class Mon
    filter_osds_by_tags(orig_tree, flat_tree, tags)
    {
        if (!tags)
-        {
+            return 1;
            return;
        }
        for (const tag of (tags instanceof Array ? tags : [ tags ]))
        {
-            for (const host in flat_tree)
+            for (const item in flat_tree)
            {
-                let found = 0;
+                if (flat_tree[item] instanceof Object)
                for (const osd in flat_tree[host])
                {
-                    if (!orig_tree[osd].tags || !orig_tree[osd].tags[tag])
+                    if (!filter_osds_by_tags(orig_tree, flat_tree[item], tags))
-                        delete flat_tree[host][osd];
+                        delete flat_tree[item];
                    else
                        found++;
                }
                if (!found)
                {
                    delete flat_tree[host];
                }
                else if (!orig_tree[item].tags || !orig_tree[item].tags[tag])
                    delete flat_tree[item];
            }
        }
        for (const item in flat_tree)
            return 1;
        return 0;
    }
    get_affinity_osds(pool_cfg, up_osds, osd_tree)
@@ -1156,6 +1186,7 @@ class Mon
        if (this.state.config.pgs.hash != tree_hash)
        {
            // Something has changed
            const new_config_pgs = JSON.parse(JSON.stringify(this.state.config.pgs));
            const etcd_request = { compare: [], success: [] };
            for (const pool_id in (this.state.config.pgs||{}).items||{})
            {
@@ -1176,7 +1207,7 @@ class Mon
                    etcd_request.success.push({ requestDeleteRange: {
                        key: b64(this.etcd_prefix+'/pool/stats/'+pool_id),
                    } });
-                    this.save_new_pgs_txn(etcd_request, pool_id, up_osds, osd_tree, prev_pgs, [], []);
+                    this.save_new_pgs_txn(new_config_pgs, etcd_request, pool_id, up_osds, osd_tree, prev_pgs, [], []);
                }
            }
            for (const pool_id in this.state.config.pools)
@@ -1186,9 +1217,11 @@ class Mon
                {
                    continue;
                }
-                let pool_tree = osd_tree[pool_cfg.root_node || ''];
+                let pool_tree = osd_tree[pool_cfg.root_node || ''] || {};
-                pool_tree = pool_tree ? pool_tree.children : [];
+                const failure_domains = pool_cfg.failure_domain instanceof Object
-                pool_tree = LPOptimizer.flatten_tree(pool_tree, levels, pool_cfg.failure_domain, 'osd');
+                    ? [ ...Object.keys(pool_cfg.failure_domain), 'osd' ]
                    : [ pool_cfg.failure_domain, 'osd' ];
                pool_tree = LPOptimizer.extract_tree_levels(pool_tree, failure_domains, levels);
                this.filter_osds_by_tags(osd_tree, pool_tree, pool_cfg.osd_tags);
                // These are for the purpose of building history.osd_sets
                const real_prev_pgs = [];
@@ -1215,6 +1248,9 @@ class Mon
                    pg_count: pool_cfg.pg_count,
                    pg_size: pool_cfg.pg_size,
                    pg_minsize: pool_cfg.pg_minsize,
                    hier_sizes: pool_cfg.failure_domain instanceof Object
                        ? [ ...Object.values(pool_cfg.failure_domain), 1 ]
                        : null,
                    max_combinations: pool_cfg.max_osd_combinations,
                    ordered: pool_cfg.scheme != 'replicated',
                };
@@ -1230,7 +1266,7 @@ class Mon
                            return;
                        }
                        const new_pg_history = [];
-                        PGUtil.scale_pg_count(prev_pgs, pg_history, new_pg_history, pool_cfg.pg_count);
+                        PGUtil.scale_pg_count(prev_pgs, real_prev_pgs, pg_history, new_pg_history, pool_cfg.pg_count);
                        pg_history = new_pg_history;
                    }
                    for (const pg of prev_pgs)
@@ -1270,7 +1306,7 @@ class Mon
                    } });
                }
                LPOptimizer.print_change_stats(optimize_result);
-                const pg_effsize = Math.min(pool_cfg.pg_size, Object.keys(pool_tree).length);
+                const pg_effsize = Math.min(pool_cfg.pg_size, Object.keys(pool_tree).length); // FIXME requires hier support too
                this.state.pool.stats[pool_id] = {
                    used_raw_tb: (this.state.pool.stats[pool_id]||{}).used_raw_tb || 0,
                    total_raw_tb: optimize_result.space,
@@ -1283,14 +1319,15 @@ class Mon
                    key: b64(this.etcd_prefix+'/pool/stats/'+pool_id),
                    value: b64(JSON.stringify(this.state.pool.stats[pool_id])),
                } });
-                this.save_new_pgs_txn(etcd_request, pool_id, up_osds, osd_tree, real_prev_pgs, optimize_result.int_pgs, pg_history);
+                this.save_new_pgs_txn(new_config_pgs, etcd_request, pool_id, up_osds, osd_tree, real_prev_pgs, optimize_result.int_pgs, pg_history);
            }
-            this.state.config.pgs.hash = tree_hash;
+            new_config_pgs.hash = tree_hash;
-            await this.save_pg_config(etcd_request);
+            await this.save_pg_config(new_config_pgs, etcd_request);
        }
        else
        {
            // Nothing changed, but we still want to recheck the distribution of primaries
            let new_config_pgs;
            let changed = false;
            for (const pool_id in this.state.config.pools)
            {
@@ -1310,31 +1347,35 @@ class Mon
                        const new_primary = this.pick_primary(pool_id, pg_cfg.osd_set, up_osds, aff_osds);
                        if (pg_cfg.primary != new_primary)
                        {
                            if (!new_config_pgs)
                            {
                                new_config_pgs = JSON.parse(JSON.stringify(this.state.config.pgs));
                            }
                            console.log(
                                `Moving pool ${pool_id} (${pool_cfg.name || 'unnamed'}) PG ${pg_num}`+
                                ` primary OSD from ${pg_cfg.primary} to ${new_primary}`
                            );
                            changed = true;
-                            pg_cfg.primary = new_primary;
+                            new_config_pgs.items[pool_id][pg_num].primary = new_primary;
                        }
                    }
                }
            }
            if (changed)
            {
-                await this.save_pg_config();
+                await this.save_pg_config(new_config_pgs);
            }
        }
    }
-    async save_pg_config(etcd_request = { compare: [], success: [] })
+    async save_pg_config(new_config_pgs, etcd_request = { compare: [], success: [] })
    {
        etcd_request.compare.push(
            { key: b64(this.etcd_prefix+'/mon/master'), target: 'LEASE', lease: ''+this.etcd_lease_id },
            { key: b64(this.etcd_prefix+'/config/pgs'), target: 'MOD', mod_revision: ''+this.etcd_watch_revision, result: 'LESS' },
        );
        etcd_request.success.push(
-            { requestPut: { key: b64(this.etcd_prefix+'/config/pgs'), value: b64(JSON.stringify(this.state.config.pgs)) } },
+            { requestPut: { key: b64(this.etcd_prefix+'/config/pgs'), value: b64(JSON.stringify(new_config_pgs)) } },
        );
        const res = await this.etcd_call('/kv/txn', etcd_request, this.config.etcd_mon_timeout, 0);
        if (!res.succeeded)
@@ -1813,6 +1854,7 @@ function POST(url, body, timeout)
            clearTimeout(timer_id);
            let res_body = '';
            res.setEncoding('utf8');
            res.on('error', (error) => ok({ error }));
            res.on('data', chunk => { res_body += chunk; });
            res.on('end', () =>
            {
@@ -1832,6 +1874,8 @@ function POST(url, body, timeout)
                }
            });
        });
        req.on('error', (error) => ok({ error }));
        req.on('close', () => ok({ error: new Error('Connection closed prematurely') }));
        req.write(body_text);
        req.end();
    });
--- a/mon/test-optimize-undersized.js
+++ b/mon/test-optimize-undersized.js
@@ -36,7 +36,7 @@ const crush_tree = [
    ] },
 ];
-const osd_tree = LPOptimizer.flatten_tree(crush_tree, {}, 1, 3);
+const osd_tree = LPOptimizer.extract_tree_levels({ level: -Infinity, children: crush_tree }, [ 1, 3 ], {});
 console.log(osd_tree);
 async function run()
@@ -47,32 +47,32 @@ async function run()
    LPOptimizer.print_change_stats(res, false);
    assert(res.space == 0);
    console.log('\nAdding 1st failure domain:');
-    cur_tree['dom1'] = osd_tree['dom1'];
+    cur_tree['l1_1'] = osd_tree['l1_1'];
    res = await LPOptimizer.optimize_change({ prev_pgs: res.int_pgs, osd_tree: cur_tree, pg_size: 3 });
    LPOptimizer.print_change_stats(res, false);
    assert(res.space == 12 && res.total_space == 12);
    console.log('\nAdding 2nd failure domain:');
-    cur_tree['dom2'] = osd_tree['dom2'];
+    cur_tree['l1_2'] = osd_tree['l1_2'];
    res = await LPOptimizer.optimize_change({ prev_pgs: res.int_pgs, osd_tree: cur_tree, pg_size: 3 });
    LPOptimizer.print_change_stats(res, false);
    assert(res.space == 24 && res.total_space == 24);
    console.log('\nAdding 3rd failure domain:');
-    cur_tree['dom3'] = osd_tree['dom3'];
+    cur_tree['l1_3'] = osd_tree['l1_3'];
    res = await LPOptimizer.optimize_change({ prev_pgs: res.int_pgs, osd_tree: cur_tree, pg_size: 3 });
    LPOptimizer.print_change_stats(res, false);
    assert(res.space == 36 && res.total_space == 36);
    console.log('\nRemoving 3rd failure domain:');
-    delete cur_tree['dom3'];
+    delete cur_tree['l1_3'];
    res = await LPOptimizer.optimize_change({ prev_pgs: res.int_pgs, osd_tree: cur_tree, pg_size: 3 });
    LPOptimizer.print_change_stats(res, false);
    assert(res.space == 24 && res.total_space == 24);
    console.log('\nRemoving 2nd failure domain:');
-    delete cur_tree['dom2'];
+    delete cur_tree['l1_2'];
    res = await LPOptimizer.optimize_change({ prev_pgs: res.int_pgs, osd_tree: cur_tree, pg_size: 3 });
    LPOptimizer.print_change_stats(res, false);
    assert(res.space == 12 && res.total_space == 12);
    console.log('\nRemoving 1st failure domain:');
-    delete cur_tree['dom1'];
+    delete cur_tree['l1_1'];
    res = await LPOptimizer.optimize_change({ prev_pgs: res.int_pgs, osd_tree: cur_tree, pg_size: 3 });
    LPOptimizer.print_change_stats(res, false);
    assert(res.space == 0);
--- a/mon/test-optimize.js
+++ b/mon/test-optimize.js
@@ -108,7 +108,11 @@ async function run()
    LPOptimizer.print_change_stats(res, false);
    console.log('\n256 PGs, size=3, failure domain=rack');
-    res = await LPOptimizer.optimize_initial({ osd_tree: LPOptimizer.flatten_tree(crush_tree, {}, 1, 3), pg_size: 3, pg_count: 256 });
+    res = await LPOptimizer.optimize_initial({
        osd_tree: LPOptimizer.extract_tree_levels({ level: -Infinity, children: crush_tree }, [ 1, 3 ], {}),
        pg_size: 3,
        pg_count: 256,
    });
    LPOptimizer.print_change_stats(res, false);
 }
--- a/mon/test-random-hier.js
+++ b/mon/test-random-hier.js
@@ -0,0 +1,56 @@
 // Copyright (c) Vitaliy Filippov, 2019+
 // License: VNPL-1.1 (see README.md for details)
 const LPOptimizer = require('./lp-optimizer.js');
 const osd_tree = {
    100: { 110: { 111: 1, 112: 1 }, 120: { 121: 1, 122: 1 } },
    200: { 210: { 211: 1, 212: 1 }, 220: { 221: 1, 222: 1 } },
    300: { 310: { 311: 1, 312: 1 }, 320: { 321: 1, 322: 1 } },
    400: { 410: { 411: 1, 412: 1 }, 420: { 421: 1, 422: 1 } },
    500: { 510: { 511: 1, 512: 1 }, 520: { 521: 1, 522: 1 } },
 };
 const osd_tree2 = {
    100: { 111: 1, 112: 1, 121: 1, 122: 1 },
    200: { 211: 1, 212: 1, 221: 1, 222: 1 },
    300: { 311: 1, 312: 1, 321: 1, 322: 1 },
    400: { 411: 1, 412: 1, 421: 1, 422: 1 },
    500: { 511: 1, 512: 1, 521: 1, 522: 1 },
 };
 const osd_tree3 = {
    100: { 111: 1, 112: 1, 121: 1, 122: 1 },
    200: { 211: 1, 212: 1, 221: 1, 222: 1 },
    300: { 311: 1, 312: 1, 321: 1, 322: 1 },
    400: { 411: 1, 412: 1, 421: 1, 422: 1 },
    500: { 511: 1 },
 };
 async function run()
 {
    let r;
    console.log(r = LPOptimizer.random_hier_combinations(osd_tree, [ 3, 2, 1 ], 10000, false, true));
    console.log(r = LPOptimizer.random_hier_combinations(osd_tree2, [ 3, 2 ], 0, false, true));
    // Will contain 'Z':
    console.log(r = LPOptimizer.random_combinations(osd_tree2, 6, 0, true));
    console.log(r = LPOptimizer.extract_tree_levels(
        { level: 'dc', children: [
            { level: 'rack', children: [
                { level: 'host', children: [
                    { level: 'osd', id: 'OSD5', size: 10 },
                ] },
            ] },
            { level: 'osd', id: 'OSD10', size: 10 },
        ] },
        [ 'rack', 'osd' ],
        { dc: 1, rack: 2, host: 3, osd: 4 }
    ));
    if (JSON.stringify(r) != '{"rack1":{"OSD5":10},"rack2":{"OSD10":10}}')
        throw new Error('extract_tree_levels failed');
    // should not contain Z:
    console.log(r = LPOptimizer.random_hier_combinations(osd_tree3, [ 3, 2 ], 0, false, true));
    console.log('OK');
 }
 run().catch(console.error);
--- a/mon/vitastor-mon.service
+++ b/mon/vitastor-mon.service
@@ -15,4 +15,4 @@ StartLimitInterval=0
 RestartSec=10
 [Install]
-WantedBy=vitastor.target
+WantedBy=multi-user.target
--- a/patches/VitastorPlugin.pm
+++ b/patches/VitastorPlugin.pm
@@ -139,6 +139,7 @@ sub options
 {
    return {
        shared => { optional => 1 },
        content => { optional => 1 },
        nodes => { optional => 1 },
        disable => { optional => 1 },
        vitastor_etcd_address => { optional => 1 },
--- a/patches/cinder-vitastor.py
+++ b/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 = '0.8.4'
+VERSION = '0.8.9'
 LOG = logging.getLogger(__name__)
--- a/patches/pve-qemu-7.2-vitastor.patch
+++ b/patches/pve-qemu-7.2-vitastor.patch
@@ -0,0 +1,169 @@
 Index: pve-qemu-kvm-7.2.0/block/meson.build
 ===================================================================
 --- pve-qemu-kvm-7.2.0.orig/block/meson.build
 +++ pve-qemu-kvm-7.2.0/block/meson.build
@@ -113,6 +113,7 @@ 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: pve-qemu-kvm-7.2.0/meson.build
 ===================================================================
 --- pve-qemu-kvm-7.2.0.orig/meson.build
 +++ pve-qemu-kvm-7.2.0/meson.build
@@ -1026,6 +1026,26 @@ 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
@@ -1865,6 +1885,7 @@ 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())
@@ -3957,6 +3978,7 @@ 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: pve-qemu-kvm-7.2.0/meson_options.txt
 ===================================================================
 --- pve-qemu-kvm-7.2.0.orig/meson_options.txt
 +++ pve-qemu-kvm-7.2.0/meson_options.txt
@@ -169,6 +169,8 @@ 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: pve-qemu-kvm-7.2.0/qapi/block-core.json
 ===================================================================
 --- pve-qemu-kvm-7.2.0.orig/qapi/block-core.json
 +++ pve-qemu-kvm-7.2.0/qapi/block-core.json
@@ -3213,7 +3213,7 @@
             'raw', 'rbd',
             { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
             'pbs',
 -            'ssh', 'throttle', 'vdi', 'vhdx',
 +            'ssh', 'throttle', 'vdi', 'vhdx', 'vitastor',
             { 'name': 'virtio-blk-vfio-pci', 'if': 'CONFIG_BLKIO' },
             { 'name': 'virtio-blk-vhost-user', 'if': 'CONFIG_BLKIO' },
             { 'name': 'virtio-blk-vhost-vdpa', 'if': 'CONFIG_BLKIO' },
@@ -4223,6 +4223,28 @@
             '*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.
@@ -4671,6 +4693,7 @@
       'throttle':   'BlockdevOptionsThrottle',
       'vdi':        'BlockdevOptionsGenericFormat',
       'vhdx':       'BlockdevOptionsGenericFormat',
 +      'vitastor':   'BlockdevOptionsVitastor',
       'virtio-blk-vfio-pci':
                     { 'type': 'BlockdevOptionsVirtioBlkVfioPci',
                       'if': 'CONFIG_BLKIO' },
@@ -5072,6 +5095,17 @@
             '*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
@@ -5269,6 +5303,7 @@
       'ssh':            'BlockdevCreateOptionsSsh',
       'vdi':            'BlockdevCreateOptionsVdi',
       'vhdx':           'BlockdevCreateOptionsVhdx',
 +      'vitastor':       'BlockdevCreateOptionsVitastor',
       'vmdk':           'BlockdevCreateOptionsVmdk',
       'vpc':            'BlockdevCreateOptionsVpc'
   } }
 Index: pve-qemu-kvm-7.2.0/scripts/ci/org.centos/stream/8/x86_64/configure
 ===================================================================
 --- pve-qemu-kvm-7.2.0.orig/scripts/ci/org.centos/stream/8/x86_64/configure
 +++ pve-qemu-kvm-7.2.0/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 \
--- a/patches/qemu-6.2-vitastor.patch
+++ b/patches/qemu-6.2-vitastor.patch
@@ -0,0 +1,169 @@
 diff --git a/block/meson.build b/block/meson.build
 index deb73ca389..e269f599a1 100644
 --- a/block/meson.build
 +++ b/block/meson.build
@@ -78,6 +78,7 @@ 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()
 diff --git a/meson.build b/meson.build
 index 96de1a6ef9..2e3994777d 100644
 --- a/meson.build
 +++ b/meson.build
@@ -838,6 +838,26 @@ if not get_option('rbd').auto() or have_block
   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
@@ -1455,6 +1475,7 @@ config_host_data.set('CONFIG_LINUX_AIO', libaio.found())
 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())
@@ -3412,6 +3433,7 @@ 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}
 diff --git a/meson_options.txt b/meson_options.txt
 index e392323732..5b56007475 100644
 --- a/meson_options.txt
 +++ b/meson_options.txt
@@ -121,6 +121,8 @@ option('lzo', type : 'feature', value : 'auto',
        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',
 diff --git a/qapi/block-core.json b/qapi/block-core.json
 index 1d3dd9cb48..88453405e5 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
@@ -2930,7 +2930,7 @@
             'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels',
             'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
             { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
 -            'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
 +            'ssh', 'throttle', 'vdi', 'vhdx', 'vitastor', 'vmdk', 'vpc', 'vvfat' ] }
 ##
 # @BlockdevOptionsFile:
@@ -3864,6 +3864,28 @@
             '*key-secret': 'str',
             '*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:
 #
@@ -4259,6 +4281,7 @@
       'throttle':   'BlockdevOptionsThrottle',
       'vdi':        'BlockdevOptionsGenericFormat',
       'vhdx':       'BlockdevOptionsGenericFormat',
 +      'vitastor':   'BlockdevOptionsVitastor',
       'vmdk':       'BlockdevOptionsGenericCOWFormat',
       'vpc':        'BlockdevOptionsGenericFormat',
       'vvfat':      'BlockdevOptionsVVFAT'
@@ -4647,6 +4670,17 @@
             '*cluster-size' :   'size',
             '*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:
 #
@@ -4846,6 +4880,7 @@
       'ssh':            'BlockdevCreateOptionsSsh',
       'vdi':            'BlockdevCreateOptionsVdi',
       'vhdx':           'BlockdevCreateOptionsVhdx',
 +      'vitastor':       'BlockdevCreateOptionsVitastor',
       'vmdk':           'BlockdevCreateOptionsVmdk',
       'vpc':            'BlockdevCreateOptionsVpc'
   } }
 diff --git a/scripts/meson-buildoptions.sh b/scripts/meson-buildoptions.sh
 index 7a17ff4218..cdddbf32aa 100644
 --- a/scripts/meson-buildoptions.sh
 +++ b/scripts/meson-buildoptions.sh
@@ -69,6 +69,7 @@ meson_options_help() {
   printf "%s\n" '  oss             OSS sound support'
   printf "%s\n" '  pa              PulseAudio sound support'
   printf "%s\n" '  rbd             Ceph block device driver'
 +  printf "%s\n" '  vitastor        Vitastor block device driver'
   printf "%s\n" '  sdl             SDL user interface'
   printf "%s\n" '  sdl-image       SDL Image support for icons'
   printf "%s\n" '  seccomp         seccomp support'
@@ -210,6 +211,8 @@ _meson_option_parse() {
     --disable-pa) printf "%s" -Dpa=disabled ;;
     --enable-rbd) printf "%s" -Drbd=enabled ;;
     --disable-rbd) printf "%s" -Drbd=disabled ;;
 +    --enable-vitastor) printf "%s" -Dvitastor=enabled ;;
 +    --disable-vitastor) printf "%s" -Dvitastor=disabled ;;
     --enable-sdl) printf "%s" -Dsdl=enabled ;;
     --disable-sdl) printf "%s" -Dsdl=disabled ;;
     --enable-sdl-image) printf "%s" -Dsdl_image=enabled ;;
--- a/patches/qemu-7.0-vitastor.patch
+++ b/patches/qemu-7.0-vitastor.patch
@@ -0,0 +1,190 @@
 diff --git a/block/meson.build b/block/meson.build
 index 0b2a60c99b..d923713804 100644
 --- a/block/meson.build
 +++ b/block/meson.build
@@ -98,6 +98,7 @@ 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()
 diff --git a/meson.build b/meson.build
 index 861de93c4f..272f72af11 100644
 --- a/meson.build
 +++ b/meson.build
@@ -884,6 +884,26 @@ if not get_option('rbd').auto() or have_block
   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
@@ -1546,6 +1566,7 @@ config_host_data.set('CONFIG_LIBPMEM', libpmem.found())
 config_host_data.set('CONFIG_NUMA', numa.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_SDL', sdl.found())
 config_host_data.set('CONFIG_SDL_IMAGE', sdl_image.found())
 config_host_data.set('CONFIG_SECCOMP', seccomp.found())
@@ -3709,6 +3730,7 @@ 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}
 diff --git a/meson_options.txt b/meson_options.txt
 index 52b11cead4..d8d0868174 100644
 --- a/meson_options.txt
 +++ b/meson_options.txt
@@ -149,6 +149,8 @@ option('lzo', type : 'feature', value : 'auto',
        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',
 diff --git a/qapi/block-core.json b/qapi/block-core.json
 index beeb91952a..1c98dc0e12 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
@@ -2929,7 +2929,7 @@
             'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels',
             'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
             { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
 -            'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
 +            'ssh', 'throttle', 'vdi', 'vhdx', 'vitastor', 'vmdk', 'vpc', 'vvfat' ] }
 ##
 # @BlockdevOptionsFile:
@@ -3863,6 +3863,28 @@
             '*key-secret': 'str',
             '*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:
 #
@@ -4277,6 +4299,7 @@
       'throttle':   'BlockdevOptionsThrottle',
       'vdi':        'BlockdevOptionsGenericFormat',
       'vhdx':       'BlockdevOptionsGenericFormat',
 +      'vitastor':   'BlockdevOptionsVitastor',
       'vmdk':       'BlockdevOptionsGenericCOWFormat',
       'vpc':        'BlockdevOptionsGenericFormat',
       'vvfat':      'BlockdevOptionsVVFAT'
@@ -4665,6 +4688,17 @@
             '*cluster-size' :   'size',
             '*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:
 #
@@ -4864,6 +4898,7 @@
       'ssh':            'BlockdevCreateOptionsSsh',
       'vdi':            'BlockdevCreateOptionsVdi',
       'vhdx':           'BlockdevCreateOptionsVhdx',
 +      'vitastor':       'BlockdevCreateOptionsVitastor',
       'vmdk':           'BlockdevCreateOptionsVmdk',
       'vpc':            'BlockdevCreateOptionsVpc'
   } }
 diff --git a/scripts/ci/org.centos/stream/8/x86_64/configure b/scripts/ci/org.centos/stream/8/x86_64/configure
 index 9850dd4444..72b1287520 100755
 --- a/scripts/ci/org.centos/stream/8/x86_64/configure
 +++ b/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 \
@@ -181,6 +181,7 @@
 --enable-opengl \
 --enable-pie \
 --enable-rbd \
 +--enable-vitastor \
 --enable-rdma \
 --enable-seccomp \
 --enable-snappy \
 diff --git a/scripts/meson-buildoptions.sh b/scripts/meson-buildoptions.sh
 index 1e26f4571e..370898d48c 100644
 --- a/scripts/meson-buildoptions.sh
 +++ b/scripts/meson-buildoptions.sh
@@ -98,6 +98,7 @@ meson_options_help() {
   printf "%s\n" '  qed             qed image format support'
   printf "%s\n" '  qga-vss         build QGA VSS support (broken with MinGW)'
   printf "%s\n" '  rbd             Ceph block device driver'
 +  printf "%s\n" '  vitastor        Vitastor block device driver'
   printf "%s\n" '  replication     replication support'
   printf "%s\n" '  sdl             SDL user interface'
   printf "%s\n" '  sdl-image       SDL Image support for icons'
@@ -289,6 +290,8 @@ _meson_option_parse() {
     --disable-qom-cast-debug) printf "%s" -Dqom_cast_debug=false ;;
     --enable-rbd) printf "%s" -Drbd=enabled ;;
     --disable-rbd) printf "%s" -Drbd=disabled ;;
 +    --enable-vitastor) printf "%s" -Dvitastor=enabled ;;
 +    --disable-vitastor) printf "%s" -Dvitastor=disabled ;;
     --enable-replication) printf "%s" -Dreplication=enabled ;;
     --disable-replication) printf "%s" -Dreplication=disabled ;;
     --enable-rng-none) printf "%s" -Drng_none=true ;;
--- a/patches/qemu-7.1-vitastor.patch
+++ b/patches/qemu-7.1-vitastor.patch
@@ -0,0 +1,190 @@
 diff --git a/block/meson.build b/block/meson.build
 index 60bc305597..89a042216f 100644
 --- a/block/meson.build
 +++ b/block/meson.build
@@ -98,6 +98,7 @@ 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()
 diff --git a/meson.build b/meson.build
 index 20fddbd707..600db4e2fb 100644
 --- a/meson.build
 +++ b/meson.build
@@ -967,6 +967,26 @@ if not get_option('rbd').auto() or have_block
   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
@@ -1799,6 +1819,7 @@ config_host_data.set('CONFIG_NUMA', numa.found())
 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())
@@ -3954,6 +3975,7 @@ 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}
 diff --git a/meson_options.txt b/meson_options.txt
 index e58e158396..9747b38fd0 100644
 --- a/meson_options.txt
 +++ b/meson_options.txt
@@ -167,6 +167,8 @@ option('lzo', type : 'feature', value : 'auto',
        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',
 diff --git a/qapi/block-core.json b/qapi/block-core.json
 index 2173e7734a..5a4900b322 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
@@ -2955,7 +2955,7 @@
             'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels',
             'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
             { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
 -            'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
 +            'ssh', 'throttle', 'vdi', 'vhdx', 'vitastor', 'vmdk', 'vpc', 'vvfat' ] }
 ##
 # @BlockdevOptionsFile:
@@ -3883,6 +3883,28 @@
             '*key-secret': 'str',
             '*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:
 #
@@ -4327,6 +4349,7 @@
       'throttle':   'BlockdevOptionsThrottle',
       'vdi':        'BlockdevOptionsGenericFormat',
       'vhdx':       'BlockdevOptionsGenericFormat',
 +      'vitastor':   'BlockdevOptionsVitastor',
       'vmdk':       'BlockdevOptionsGenericCOWFormat',
       'vpc':        'BlockdevOptionsGenericFormat',
       'vvfat':      'BlockdevOptionsVVFAT'
@@ -4717,6 +4740,17 @@
             '*cluster-size' :   'size',
             '*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:
 #
@@ -4915,6 +4949,7 @@
       'ssh':            'BlockdevCreateOptionsSsh',
       'vdi':            'BlockdevCreateOptionsVdi',
       'vhdx':           'BlockdevCreateOptionsVhdx',
 +      'vitastor':       'BlockdevCreateOptionsVitastor',
       'vmdk':           'BlockdevCreateOptionsVmdk',
       'vpc':            'BlockdevCreateOptionsVpc'
   } }
 diff --git a/scripts/ci/org.centos/stream/8/x86_64/configure b/scripts/ci/org.centos/stream/8/x86_64/configure
 index a7f92aff90..53dc55be2e 100755
 --- a/scripts/ci/org.centos/stream/8/x86_64/configure
 +++ b/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 \
 diff --git a/scripts/meson-buildoptions.sh b/scripts/meson-buildoptions.sh
 index 359b04e0e6..f5b85ba78c 100644
 --- a/scripts/meson-buildoptions.sh
 +++ b/scripts/meson-buildoptions.sh
@@ -135,6 +135,7 @@ meson_options_help() {
   printf "%s\n" '  qed             qed image format support'
   printf "%s\n" '  qga-vss         build QGA VSS support (broken with MinGW)'
   printf "%s\n" '  rbd             Ceph block device driver'
 +  printf "%s\n" '  vitastor        Vitastor block device driver'
   printf "%s\n" '  rdma            Enable RDMA-based migration'
   printf "%s\n" '  replication     replication support'
   printf "%s\n" '  sdl             SDL user interface'
@@ -370,6 +371,8 @@ _meson_option_parse() {
     --disable-qom-cast-debug) printf "%s" -Dqom_cast_debug=false ;;
     --enable-rbd) printf "%s" -Drbd=enabled ;;
     --disable-rbd) printf "%s" -Drbd=disabled ;;
 +    --enable-vitastor) printf "%s" -Dvitastor=enabled ;;
 +    --disable-vitastor) printf "%s" -Dvitastor=disabled ;;
     --enable-rdma) printf "%s" -Drdma=enabled ;;
     --disable-rdma) printf "%s" -Drdma=disabled ;;
     --enable-replication) printf "%s" -Dreplication=enabled ;;
--- a/patches/qemu-7.2-vitastor.patch
+++ b/patches/qemu-7.2-vitastor.patch
@@ -0,0 +1,190 @@
 diff --git a/block/meson.build b/block/meson.build
 index b7c68b83a3..95d8a6f15d 100644
 --- a/block/meson.build
 +++ b/block/meson.build
@@ -100,6 +100,7 @@ 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()
 diff --git a/meson.build b/meson.build
 index 5c6b5a1c75..f31f73612e 100644
 --- a/meson.build
 +++ b/meson.build
@@ -1026,6 +1026,26 @@ if not get_option('rbd').auto() or have_block
   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
@@ -1861,6 +1881,7 @@ config_host_data.set('CONFIG_NUMA', numa.found())
 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())
@@ -3945,6 +3966,7 @@ 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}
 diff --git a/meson_options.txt b/meson_options.txt
 index 4b749ca549..6b37bd6b77 100644
 --- a/meson_options.txt
 +++ b/meson_options.txt
@@ -169,6 +169,8 @@ option('lzo', type : 'feature', value : 'auto',
        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',
 diff --git a/qapi/block-core.json b/qapi/block-core.json
 index 95ac4fa634..7a240827e4 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
@@ -2959,7 +2959,7 @@
             'parallels', 'preallocate', 'qcow', 'qcow2', 'qed', 'quorum',
             'raw', 'rbd',
             { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
 -            'ssh', 'throttle', 'vdi', 'vhdx',
 +            'ssh', 'throttle', 'vdi', 'vhdx', 'vitastor',
             { 'name': 'virtio-blk-vfio-pci', 'if': 'CONFIG_BLKIO' },
             { 'name': 'virtio-blk-vhost-user', 'if': 'CONFIG_BLKIO' },
             { 'name': 'virtio-blk-vhost-vdpa', 'if': 'CONFIG_BLKIO' },
@@ -3957,6 +3957,28 @@
             '*key-secret': 'str',
             '*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:
 #
@@ -4405,6 +4427,7 @@
       'throttle':   'BlockdevOptionsThrottle',
       'vdi':        'BlockdevOptionsGenericFormat',
       'vhdx':       'BlockdevOptionsGenericFormat',
 +      'vitastor':   'BlockdevOptionsVitastor',
       'virtio-blk-vfio-pci':
                     { 'type': 'BlockdevOptionsVirtioBlkVfioPci',
                       'if': 'CONFIG_BLKIO' },
@@ -4804,6 +4827,17 @@
             '*cluster-size' :   'size',
             '*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:
 #
@@ -5002,6 +5036,7 @@
       'ssh':            'BlockdevCreateOptionsSsh',
       'vdi':            'BlockdevCreateOptionsVdi',
       'vhdx':           'BlockdevCreateOptionsVhdx',
 +      'vitastor':       'BlockdevCreateOptionsVitastor',
       'vmdk':           'BlockdevCreateOptionsVmdk',
       'vpc':            'BlockdevCreateOptionsVpc'
   } }
 diff --git a/scripts/ci/org.centos/stream/8/x86_64/configure b/scripts/ci/org.centos/stream/8/x86_64/configure
 index a7f92aff90..53dc55be2e 100755
 --- a/scripts/ci/org.centos/stream/8/x86_64/configure
 +++ b/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 \
 diff --git a/scripts/meson-buildoptions.sh b/scripts/meson-buildoptions.sh
 index aa6e30ea91..c45d21c40f 100644
 --- a/scripts/meson-buildoptions.sh
 +++ b/scripts/meson-buildoptions.sh
@@ -135,6 +135,7 @@ meson_options_help() {
   printf "%s\n" '  qed             qed image format support'
   printf "%s\n" '  qga-vss         build QGA VSS support (broken with MinGW)'
   printf "%s\n" '  rbd             Ceph block device driver'
 +  printf "%s\n" '  vitastor        Vitastor block device driver'
   printf "%s\n" '  rdma            Enable RDMA-based migration'
   printf "%s\n" '  replication     replication support'
   printf "%s\n" '  sdl             SDL user interface'
@@ -376,6 +377,8 @@ _meson_option_parse() {
     --disable-qom-cast-debug) printf "%s" -Dqom_cast_debug=false ;;
     --enable-rbd) printf "%s" -Drbd=enabled ;;
     --disable-rbd) printf "%s" -Drbd=disabled ;;
 +    --enable-vitastor) printf "%s" -Dvitastor=enabled ;;
 +    --disable-vitastor) printf "%s" -Dvitastor=disabled ;;
     --enable-rdma) printf "%s" -Drdma=enabled ;;
     --disable-rdma) printf "%s" -Drdma=disabled ;;
     --enable-replication) printf "%s" -Dreplication=enabled ;;
--- a/patches/qemu-8.0-vitastor.patch
+++ b/patches/qemu-8.0-vitastor.patch
@@ -0,0 +1,190 @@
 diff --git a/block/meson.build b/block/meson.build
 index 382bec0e7d..af6207dbce 100644
 --- a/block/meson.build
 +++ b/block/meson.build
@@ -101,6 +101,7 @@ 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()
 diff --git a/meson.build b/meson.build
 index c44d05a13f..ebedb42843 100644
 --- a/meson.build
 +++ b/meson.build
@@ -1028,6 +1028,26 @@ if not get_option('rbd').auto() or have_block
   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
@@ -1878,6 +1898,7 @@ endif
 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())
@@ -4002,6 +4023,7 @@ 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}
 diff --git a/meson_options.txt b/meson_options.txt
 index fc9447d267..c4ac55c283 100644
 --- a/meson_options.txt
 +++ b/meson_options.txt
@@ -173,6 +173,8 @@ option('lzo', type : 'feature', value : 'auto',
        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',
 diff --git a/qapi/block-core.json b/qapi/block-core.json
 index c05ad0c07e..f5eb701604 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
@@ -3054,7 +3054,7 @@
             'parallels', 'preallocate', 'qcow', 'qcow2', 'qed', 'quorum',
             'raw', 'rbd',
             { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
 -            'ssh', 'throttle', 'vdi', 'vhdx',
 +            'ssh', 'throttle', 'vdi', 'vhdx', 'vitastor',
             { 'name': 'virtio-blk-vfio-pci', 'if': 'CONFIG_BLKIO' },
             { 'name': 'virtio-blk-vhost-user', 'if': 'CONFIG_BLKIO' },
             { 'name': 'virtio-blk-vhost-vdpa', 'if': 'CONFIG_BLKIO' },
@@ -4073,6 +4073,28 @@
             '*key-secret': 'str',
             '*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:
 #
@@ -4521,6 +4543,7 @@
       'throttle':   'BlockdevOptionsThrottle',
       'vdi':        'BlockdevOptionsGenericFormat',
       'vhdx':       'BlockdevOptionsGenericFormat',
 +      'vitastor':   'BlockdevOptionsVitastor',
       'virtio-blk-vfio-pci':
                     { 'type': 'BlockdevOptionsVirtioBlkVfioPci',
                       'if': 'CONFIG_BLKIO' },
@@ -4920,6 +4943,17 @@
             '*cluster-size' :   'size',
             '*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:
 #
@@ -5118,6 +5152,7 @@
       'ssh':            'BlockdevCreateOptionsSsh',
       'vdi':            'BlockdevCreateOptionsVdi',
       'vhdx':           'BlockdevCreateOptionsVhdx',
 +      'vitastor':       'BlockdevCreateOptionsVitastor',
       'vmdk':           'BlockdevCreateOptionsVmdk',
       'vpc':            'BlockdevCreateOptionsVpc'
   } }
 diff --git a/scripts/ci/org.centos/stream/8/x86_64/configure b/scripts/ci/org.centos/stream/8/x86_64/configure
 index 6e8983f39c..1b0b9fcf3e 100755
 --- a/scripts/ci/org.centos/stream/8/x86_64/configure
 +++ b/scripts/ci/org.centos/stream/8/x86_64/configure
@@ -32,7 +32,7 @@
 --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 \
 diff --git a/scripts/meson-buildoptions.sh b/scripts/meson-buildoptions.sh
 index 009fab1515..95914e6ebc 100644
 --- a/scripts/meson-buildoptions.sh
 +++ b/scripts/meson-buildoptions.sh
@@ -142,6 +142,7 @@ meson_options_help() {
   printf "%s\n" '  qed             qed image format support'
   printf "%s\n" '  qga-vss         build QGA VSS support (broken with MinGW)'
   printf "%s\n" '  rbd             Ceph block device driver'
 +  printf "%s\n" '  vitastor        Vitastor block device driver'
   printf "%s\n" '  rdma            Enable RDMA-based migration'
   printf "%s\n" '  replication     replication support'
   printf "%s\n" '  sdl             SDL user interface'
@@ -388,6 +389,8 @@ _meson_option_parse() {
     --disable-qom-cast-debug) printf "%s" -Dqom_cast_debug=false ;;
     --enable-rbd) printf "%s" -Drbd=enabled ;;
     --disable-rbd) printf "%s" -Drbd=disabled ;;
 +    --enable-vitastor) printf "%s" -Dvitastor=enabled ;;
 +    --disable-vitastor) printf "%s" -Dvitastor=disabled ;;
     --enable-rdma) printf "%s" -Drdma=enabled ;;
     --disable-rdma) printf "%s" -Drdma=disabled ;;
     --enable-replication) printf "%s" -Dreplication=enabled ;;
--- a/rpm/build-tarball.sh
+++ b/rpm/build-tarball.sh
@@ -7,13 +7,12 @@ set -e
 VITASTOR=$(dirname $0)
 VITASTOR=$(realpath "$VITASTOR/..")
-if [ -d /opt/rh/gcc-toolset-9 ]; then
+EL=$(rpm --eval '%dist')
 if [ "$EL" = ".el8" ]; then
    # CentOS 8
    EL=8
    . /opt/rh/gcc-toolset-9/enable
-else
+elif [ "$EL" = ".el7" ]; then
    # CentOS 7
    EL=7
    . /opt/rh/devtoolset-9/enable
 fi
 cd ~/rpmbuild/SPECS
@@ -25,4 +24,4 @@ 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-0.8.4/#' --exclude 'rpm/*.rpm' -czf $VITASTOR/../vitastor-0.8.4$(rpm --eval '%dist').tar.gz *
+tar --transform 's#^#vitastor-0.8.9/#' --exclude 'rpm/*.rpm' -czf $VITASTOR/../vitastor-0.8.9$(rpm --eval '%dist').tar.gz *
--- a/rpm/qemu-kvm-7.0-el9.spec.patch
+++ b/rpm/qemu-kvm-7.0-el9.spec.patch
@@ -0,0 +1,93 @@
 --- qemu-kvm.spec.orig	2023-02-28 08:04:06.000000000 +0000
 +++ qemu-kvm.spec	2023-04-27 22:29:18.094878829 +0000
@@ -100,8 +100,6 @@
 %endif
 %global target_list %{kvm_target}-softmmu
 -%global block_drivers_rw_list qcow2,raw,file,host_device,nbd,iscsi,rbd,blkdebug,luks,null-co,nvme,copy-on-read,throttle,compress
 -%global block_drivers_ro_list vdi,vmdk,vhdx,vpc,https
 %define qemudocdir %{_docdir}/%{name}
 %global firmwaredirs "%{_datadir}/qemu-firmware:%{_datadir}/ipxe/qemu:%{_datadir}/seavgabios:%{_datadir}/seabios"
@@ -129,6 +127,7 @@ Requires: %{name}-device-usb-host = %{ep
 Requires: %{name}-device-usb-redirect = %{epoch}:%{version}-%{release}   \
 %endif                                                           \
 Requires: %{name}-block-rbd = %{epoch}:%{version}-%{release}     \
 +Requires: %{name}-block-vitastor = %{epoch}:%{version}-%{release}\
 Requires: %{name}-audio-pa = %{epoch}:%{version}-%{release}
 # Since SPICE is removed from RHEL-9, the following Obsoletes:
@@ -151,7 +150,7 @@ Obsoletes: %{name}-block-ssh <= %{epoch}
 Summary: QEMU is a machine emulator and virtualizer
 Name: qemu-kvm
 Version: 7.0.0
 -Release: 13%{?rcrel}%{?dist}%{?cc_suffix}.2
 +Release: 13.vitastor%{?rcrel}%{?dist}%{?cc_suffix}
 # Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
 # Epoch 15 used for RHEL 8
 # Epoch 17 used for RHEL 9 (due to release versioning offset in RHEL 8.5)
@@ -174,6 +173,7 @@ Source28: 95-kvm-memlock.conf
 Source30: kvm-s390x.conf
 Source31: kvm-x86.conf
 Source36: README.tests
 +Source37: qemu-vitastor.c
 Patch0004: 0004-Initial-redhat-build.patch
@@ -498,6 +498,7 @@ Patch171: kvm-i386-do-kvm_put_msr_featur
 Patch172: kvm-target-i386-kvm-fix-kvmclock_current_nsec-Assertion-.patch
 # For bz#2168221 - while live-migrating many instances concurrently, libvirt sometimes return internal error: migration was active, but no RAM info was set [rhel-9.1.0.z]
 Patch173: kvm-migration-Read-state-once.patch
 +Patch174: qemu-7.0-vitastor.patch
 # Source-git patches
@@ -531,6 +532,7 @@ BuildRequires: libcurl-devel
 %if %{have_block_rbd}
 BuildRequires: librbd-devel
 %endif
 +BuildRequires: vitastor-client-devel
 # We need both because the 'stap' binary is probed for by configure
 BuildRequires: systemtap
 BuildRequires: systemtap-sdt-devel
@@ -718,6 +720,14 @@ using the rbd protocol.
 %endif
 +%package  block-vitastor
 +Summary: QEMU Vitastor block driver
 +Requires: %{name}-common%{?_isa} = %{epoch}:%{version}-%{release}
 +
 +%description block-vitastor
 +This package provides the additional Vitastor block driver for QEMU.
 +
 +
 %package  audio-pa
 Summary: QEMU PulseAudio audio driver
 Requires: %{name}-common%{?_isa} = %{epoch}:%{version}-%{release}
@@ -811,6 +821,7 @@ This package provides usbredir support.
 %prep
 %setup -q -n qemu-%{version}%{?rcstr}
 %autopatch -p1
 +cp %{SOURCE37} ./block/vitastor.c
 %global qemu_kvm_build qemu_kvm_build
 mkdir -p %{qemu_kvm_build}
@@ -1032,6 +1043,7 @@ run_configure \
 %if %{have_block_rbd}
   --enable-rbd \
 %endif
 +  --enable-vitastor \
 %if %{have_librdma}
   --enable-rdma \
 %endif
@@ -1511,6 +1523,9 @@ useradd -r -u 107 -g qemu -G kvm -d / -s
 %files block-rbd
 %{_libdir}/%{name}/block-rbd.so
 %endif
 +%files block-vitastor
 +%{_libdir}/%{name}/block-vitastor.so
 +
 %files audio-pa
 %{_libdir}/%{name}/audio-pa.so
--- a/rpm/vitastor-el7.Dockerfile
+++ b/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-0.8.4.el7.tar.gz ~/rpmbuild/SOURCES; \
+    cp /root/vitastor-0.8.9.el7.tar.gz ~/rpmbuild/SOURCES; \
    cp vitastor-el7.spec ~/rpmbuild/SPECS/vitastor.spec; \
    cd ~/rpmbuild/SPECS/; \
    rpmbuild -ba vitastor.spec; \
--- a/rpm/vitastor-el7.spec
+++ b/rpm/vitastor-el7.spec
@@ -1,11 +1,11 @@
 Name:           vitastor
-Version:        0.8.4
+Version:        0.8.9
 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-0.8.4.el7.tar.gz
+Source0:        vitastor-0.8.9.el7.tar.gz
 BuildRequires:  liburing-devel >= 0.6
 BuildRequires:  gperftools-devel
@@ -35,6 +35,7 @@ Summary:        Vitastor - OSD
 Requires:       libJerasure2
 Requires:       libisa-l
 Requires:       liburing >= 0.6
 Requires:       liburing < 2
 Requires:       vitastor-client = %{version}-%{release}
 Requires:       util-linux
 Requires:       parted
@@ -59,6 +60,7 @@ scheduling cluster-level operations.
 %package -n vitastor-client
 Summary:        Vitastor - client
 Requires:       liburing >= 0.6
 Requires:       liburing < 2
 %description -n vitastor-client
--- a/rpm/vitastor-el8.Dockerfile
+++ b/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-0.8.4.el8.tar.gz ~/rpmbuild/SOURCES; \
+    cp /root/vitastor-0.8.9.el8.tar.gz ~/rpmbuild/SOURCES; \
    cp vitastor-el8.spec ~/rpmbuild/SPECS/vitastor.spec; \
    cd ~/rpmbuild/SPECS/; \
    rpmbuild -ba vitastor.spec; \
--- a/rpm/vitastor-el8.spec
+++ b/rpm/vitastor-el8.spec
@@ -1,11 +1,11 @@
 Name:           vitastor
-Version:        0.8.4
+Version:        0.8.9
 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-0.8.4.el8.tar.gz
+Source0:        vitastor-0.8.9.el8.tar.gz
 BuildRequires:  liburing-devel >= 0.6
 BuildRequires:  gperftools-devel
@@ -34,6 +34,7 @@ Summary:        Vitastor - OSD
 Requires:       libJerasure2
 Requires:       libisa-l
 Requires:       liburing >= 0.6
 Requires:       liburing < 2
 Requires:       vitastor-client = %{version}-%{release}
 Requires:       util-linux
 Requires:       parted
@@ -57,6 +58,7 @@ scheduling cluster-level operations.
 %package -n vitastor-client
 Summary:        Vitastor - client
 Requires:       liburing >= 0.6
 Requires:       liburing < 2
 %description -n vitastor-client
--- a/rpm/vitastor-el9.Dockerfile
+++ b/rpm/vitastor-el9.Dockerfile
@@ -0,0 +1,28 @@
 # Build packages for AlmaLinux 9 inside a container
 # cd ..; podman build -t vitastor-el9 -v `pwd`/packages:/root/packages -f rpm/vitastor-el9.Dockerfile .
 FROM almalinux:9
 WORKDIR /root
 RUN sed -i 's/enabled=0/enabled=1/' /etc/yum.repos.d/*.repo
 RUN dnf -y install epel-release dnf-plugins-core
 RUN dnf -y install https://vitastor.io/rpms/centos/9/vitastor-release-1.0-1.el9.noarch.rpm
 RUN dnf -y install gcc-c++ gperftools-devel fio nodejs rpm-build jerasure-devel libisa-l-devel gf-complete-devel rdma-core-devel libarchive liburing-devel cmake
 RUN dnf download --source fio
 RUN rpm --nomd5 -i fio*.src.rpm
 RUN cd ~/rpmbuild/SPECS && dnf builddep -y --spec fio.spec
 ADD . /root/vitastor
 RUN set -e; \
    cd /root/vitastor/rpm; \
    sh build-tarball.sh; \
    cp /root/vitastor-0.8.9.el9.tar.gz ~/rpmbuild/SOURCES; \
    cp vitastor-el9.spec ~/rpmbuild/SPECS/vitastor.spec; \
    cd ~/rpmbuild/SPECS/; \
    rpmbuild -ba vitastor.spec; \
    mkdir -p /root/packages/vitastor-el9; \
    rm -rf /root/packages/vitastor-el9/*; \
    cp ~/rpmbuild/RPMS/*/vitastor* /root/packages/vitastor-el9/; \
    cp ~/rpmbuild/SRPMS/vitastor* /root/packages/vitastor-el9/
--- a/rpm/vitastor-el9.spec
+++ b/rpm/vitastor-el9.spec
@@ -0,0 +1,158 @@
 Name:           vitastor
 Version:        0.8.9
 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-0.8.9.el9.tar.gz
 BuildRequires:  liburing-devel >= 0.6
 BuildRequires:  gperftools-devel
 BuildRequires:  gcc-c++
 BuildRequires:  nodejs >= 10
 BuildRequires:  jerasure-devel
 BuildRequires:  libisa-l-devel
 BuildRequires:  gf-complete-devel
 BuildRequires:  rdma-core-devel
 BuildRequires:  cmake
 Requires:       vitastor-osd = %{version}-%{release}
 Requires:       vitastor-mon = %{version}-%{release}
 Requires:       vitastor-client = %{version}-%{release}
 Requires:       vitastor-client-devel = %{version}-%{release}
 Requires:       vitastor-fio = %{version}-%{release}
 %description
 Vitastor is a small, simple and fast clustered block storage (storage for VM drives),
 architecturally similar to Ceph which means strong consistency, primary-replication,
 symmetric clustering and automatic data distribution over any number of drives of any
 size with configurable redundancy (replication or erasure codes/XOR).
 %package -n vitastor-osd
 Summary:        Vitastor - OSD
 Requires:       vitastor-client = %{version}-%{release}
 Requires:       util-linux
 Requires:       parted
 %description -n vitastor-osd
 Vitastor object storage daemon, i.e. server program that stores data.
 %package -n vitastor-mon
 Summary:        Vitastor - monitor
 Requires:       nodejs >= 10
 Requires:       lpsolve
 %description -n vitastor-mon
 Vitastor monitor, i.e. server program responsible for watching cluster state and
 scheduling cluster-level operations.
 %package -n vitastor-client
 Summary:        Vitastor - client
 %description -n vitastor-client
 Vitastor client library and command-line interface.
 %package -n vitastor-client-devel
 Summary:        Vitastor - development files
 Group:          Development/Libraries
 Requires:       vitastor-client = %{version}-%{release}
 %description -n vitastor-client-devel
 Vitastor library headers for development.
 %package -n vitastor-fio
 Summary:        Vitastor - fio drivers
 Group:          Development/Libraries
 Requires:       vitastor-client = %{version}-%{release}
 Requires:       fio = 3.27-7.el9
 %description -n vitastor-fio
 Vitastor fio drivers for benchmarking.
 %prep
 %setup -q
 %build
 %cmake
 %cmake_build
 %install
 rm -rf $RPM_BUILD_ROOT
 %cmake_install
 cd mon
 npm install
 cd ..
 mkdir -p %buildroot/usr/lib/vitastor
 cp -r mon %buildroot/usr/lib/vitastor
 mkdir -p %buildroot/lib/systemd/system
 cp mon/vitastor.target mon/vitastor-mon.service mon/vitastor-osd@.service %buildroot/lib/systemd/system
 mkdir -p %buildroot/lib/udev/rules.d
 cp mon/90-vitastor.rules %buildroot/lib/udev/rules.d
 %files
 %doc GPL-2.0.txt VNPL-1.1.txt README.md README-ru.md
 %files -n vitastor-osd
 %_bindir/vitastor-osd
 %_bindir/vitastor-disk
 %_bindir/vitastor-dump-journal
 /lib/systemd/system/vitastor-osd@.service
 /lib/systemd/system/vitastor.target
 /lib/udev/rules.d/90-vitastor.rules
 %pre -n vitastor-osd
 groupadd -r -f vitastor 2>/dev/null ||:
 useradd -r -g vitastor -s /sbin/nologin -c "Vitastor daemons" -M -d /nonexistent vitastor 2>/dev/null ||:
 install -o vitastor -g vitastor -d /var/log/vitastor
 mkdir -p /etc/vitastor
 %files -n vitastor-mon
 /usr/lib/vitastor/mon
 /lib/systemd/system/vitastor-mon.service
 %pre -n vitastor-mon
 groupadd -r -f vitastor 2>/dev/null ||:
 useradd -r -g vitastor -s /sbin/nologin -c "Vitastor daemons" -M -d /nonexistent vitastor 2>/dev/null ||:
 mkdir -p /etc/vitastor
 %files -n vitastor-client
 %_bindir/vitastor-nbd
 %_bindir/vitastor-nfs
 %_bindir/vitastor-cli
 %_bindir/vitastor-rm
 %_bindir/vita
 %_libdir/libvitastor_blk.so*
 %_libdir/libvitastor_client.so*
 %files -n vitastor-client-devel
 %_includedir/vitastor_c.h
 %_libdir/pkgconfig
 %files -n vitastor-fio
 %_libdir/libfio_vitastor.so
 %_libdir/libfio_vitastor_blk.so
 %_libdir/libfio_vitastor_sec.so
 %changelog
--- a/src/CMakeLists.txt
+++ b/src/CMakeLists.txt
@@ -1,8 +1,9 @@
-cmake_minimum_required(VERSION 2.8)
+cmake_minimum_required(VERSION 2.8.12)
 project(vitastor)
 include(GNUInstallDirs)
 include(CTest)
 set(WITH_QEMU false CACHE BOOL "Build QEMU driver inside Vitastor source tree")
 set(WITH_FIO true CACHE BOOL "Build FIO driver")
@@ -15,7 +16,7 @@ if("${CMAKE_INSTALL_PREFIX}" MATCHES "^/usr/local/?$")
 	set(CMAKE_INSTALL_RPATH "${CMAKE_INSTALL_PREFIX}/${CMAKE_INSTALL_LIBDIR}")
 endif()
-add_definitions(-DVERSION="0.8.4")
+add_definitions(-DVERSION="0.8.9")
 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)
@@ -55,6 +56,14 @@ if (ISAL_LIBRARIES)
 	add_definitions(-DWITH_ISAL)
 endif (ISAL_LIBRARIES)
 add_custom_target(build_tests)
 add_custom_target(test
 	COMMAND
 	echo leak:tcmalloc > ${CMAKE_CURRENT_BINARY_DIR}/lsan-suppress.txt &&
 	env LSAN_OPTIONS=suppressions=${CMAKE_CURRENT_BINARY_DIR}/lsan-suppress.txt ${CMAKE_CTEST_COMMAND}
 )
 add_dependencies(test build_tests)
 include_directories(
 	../
 	/usr/include/jerasure
@@ -145,7 +154,6 @@ add_library(vitastor_client SHARED
 set_target_properties(vitastor_client PROPERTIES PUBLIC_HEADER "vitastor_c.h")
 target_link_libraries(vitastor_client
 	vitastor_common
 	tcmalloc_minimal
 	${LIBURING_LIBRARIES}
 	${IBVERBS_LIBRARIES}
 )
@@ -235,14 +243,17 @@ 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 EXCLUDE_FROM_ALL osd_rmw_test.cpp allocator.cpp)
 add_executable(osd_rmw_test osd_rmw_test.cpp allocator.cpp)
 target_link_libraries(osd_rmw_test Jerasure ${ISAL_LIBRARIES} tcmalloc_minimal)
 add_dependencies(build_tests osd_rmw_test)
 add_test(NAME osd_rmw_test COMMAND osd_rmw_test)
 if (ISAL_LIBRARIES)
-	add_executable(osd_rmw_test_je osd_rmw_test.cpp allocator.cpp)
+	add_executable(osd_rmw_test_je EXCLUDE_FROM_ALL 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)
 	add_dependencies(build_tests osd_rmw_test_je)
 	add_test(NAME osd_rmw_test_jerasure COMMAND osd_rmw_test_je)
 endif (ISAL_LIBRARIES)
 # stub_uring_osd
@@ -257,11 +268,15 @@ target_link_libraries(stub_uring_osd
 )
 # osd_peering_pg_test
-add_executable(osd_peering_pg_test osd_peering_pg_test.cpp osd_peering_pg.cpp)
+add_executable(osd_peering_pg_test EXCLUDE_FROM_ALL osd_peering_pg_test.cpp osd_peering_pg.cpp)
 target_link_libraries(osd_peering_pg_test tcmalloc_minimal)
 add_dependencies(build_tests osd_peering_pg_test)
 add_test(NAME osd_peering_pg_test COMMAND osd_peering_pg_test)
 # test_allocator
-add_executable(test_allocator test_allocator.cpp allocator.cpp)
+add_executable(test_allocator EXCLUDE_FROM_ALL test_allocator.cpp allocator.cpp)
 add_dependencies(build_tests test_allocator)
 add_test(NAME test_allocator COMMAND test_allocator)
 # test_cas
 add_executable(test_cas
@@ -281,12 +296,15 @@ target_link_libraries(test_crc32
 # test_cluster_client
 add_executable(test_cluster_client
 	EXCLUDE_FROM_ALL
 	test_cluster_client.cpp
 	pg_states.cpp osd_ops.cpp cluster_client.cpp cluster_client_list.cpp msgr_op.cpp mock/messenger.cpp msgr_stop.cpp
 	etcd_state_client.cpp timerfd_manager.cpp ../json11/json11.cpp
 )
 target_compile_definitions(test_cluster_client PUBLIC -D__MOCK__)
 target_include_directories(test_cluster_client PUBLIC ${CMAKE_SOURCE_DIR}/src/mock)
 add_dependencies(build_tests test_cluster_client)
 add_test(NAME test_cluster_client COMMAND test_cluster_client)
 ## test_blockstore, test_shit
 #add_executable(test_blockstore test_blockstore.cpp)
--- a/src/blockstore.cpp
+++ b/src/blockstore.cpp
@@ -13,6 +13,11 @@ blockstore_t::~blockstore_t()
    delete impl;
 }
 void blockstore_t::parse_config(blockstore_config_t & config)
 {
    impl->parse_config(config, false);
 }
 void blockstore_t::loop()
 {
    impl->loop();
--- a/src/blockstore.h
+++ b/src/blockstore.h
@@ -107,7 +107,7 @@ Input:
 - buf = pre-allocated obj_ver_id array <len> units long
 Output:
- retval = 0 or negative error number (-EINVAL, -ENOENT if no such version or -EBUSY if not synced)
+- retval = 0 or negative error number (-ENOENT if no such version for stabilize)
 ## BS_OP_SYNC_STAB_ALL
@@ -165,6 +165,9 @@ public:
    blockstore_t(blockstore_config_t & config, ring_loop_t *ringloop, timerfd_manager_t *tfd);
    ~blockstore_t();
    // Update configuration
    void parse_config(blockstore_config_t & config);
    // Event loop
    void loop();
--- a/src/blockstore_flush.cpp
+++ b/src/blockstore_flush.cpp
@@ -932,7 +932,7 @@ bool journal_flusher_co::fsync_batch(bool fsync_meta, int wait_base)
    resume_1:
        if (!cur_sync->state)
        {
-            if (flusher->syncing_flushers >= flusher->cur_flusher_count || !flusher->flush_queue.size())
+            if (flusher->syncing_flushers >= flusher->active_flushers || !flusher->flush_queue.size())
            {
                // Sync batch is ready. Do it.
                await_sqe(0);
--- a/src/blockstore_impl.cpp
+++ b/src/blockstore_impl.cpp
@@ -11,7 +11,7 @@ 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;
-    parse_config(config);
+    parse_config(config, true);
    zero_object = (uint8_t*)memalign_or_die(MEM_ALIGNMENT, dsk.data_block_size);
    try
    {
@@ -171,7 +171,7 @@ void blockstore_impl_t::loop()
                    // Can't submit SYNC before previous writes
                    continue;
                }
-                wr_st = continue_sync(op, false);
+                wr_st = continue_sync(op);
                if (wr_st != 2)
                {
                    has_writes = wr_st > 0 ? 1 : 2;
@@ -307,6 +307,18 @@ void blockstore_impl_t::check_wait(blockstore_op_t *op)
        }
        PRIV(op)->wait_for = 0;
    }
    else if (PRIV(op)->wait_for == WAIT_FREE)
    {
        if (!data_alloc->get_free_count() && big_to_flush > 0)
        {
 #ifdef BLOCKSTORE_DEBUG
            printf("Still waiting for free space on the data device\n");
 #endif
            return;
        }
        flusher->release_trim();
        PRIV(op)->wait_for = 0;
    }
    else
    {
        throw std::runtime_error("BUG: op->wait_for value is unexpected");
@@ -325,7 +337,7 @@ void blockstore_impl_t::enqueue_op(blockstore_op_t *op)
    {
        // Basic verification not passed
        op->retval = -EINVAL;
-        std::function<void (blockstore_op_t*)>(op->callback)(op);
+        ringloop->set_immediate([op]() { std::function<void (blockstore_op_t*)>(op->callback)(op); });
        return;
    }
    if (op->opcode == BS_OP_SYNC_STAB_ALL)
@@ -368,16 +380,21 @@ void blockstore_impl_t::enqueue_op(blockstore_op_t *op)
    }
    if ((op->opcode == BS_OP_WRITE || op->opcode == BS_OP_WRITE_STABLE || op->opcode == BS_OP_DELETE) && !enqueue_write(op))
    {
-        std::function<void (blockstore_op_t*)>(op->callback)(op);
+        ringloop->set_immediate([op]() { std::function<void (blockstore_op_t*)>(op->callback)(op); });
        return;
    }
    init_op(op);
    submit_queue.push_back(op);
    ringloop->wakeup();
 }
 void blockstore_impl_t::init_op(blockstore_op_t *op)
 {
    // Call constructor without allocating memory. We'll call destructor before returning op back
    new ((void*)op->private_data) blockstore_op_private_t;
    PRIV(op)->wait_for = 0;
    PRIV(op)->op_state = 0;
    PRIV(op)->pending_ops = 0;
    submit_queue.push_back(op);
    ringloop->wakeup();
 }
 static bool replace_stable(object_id oid, uint64_t version, int search_start, int search_end, obj_ver_id* list)
--- a/src/blockstore_impl.h
+++ b/src/blockstore_impl.h
@@ -160,6 +160,8 @@ 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
 {
@@ -216,6 +218,11 @@ struct pool_shard_settings_t
    uint32_t pg_stripe_size;
 };
 #define STAB_SPLIT_DONE 1
 #define STAB_SPLIT_WAIT 2
 #define STAB_SPLIT_SYNC 3
 #define STAB_SPLIT_TODO 4
 class blockstore_impl_t
 {
    blockstore_disk_t dsk;
@@ -258,6 +265,7 @@ class blockstore_impl_t
    struct journal_t journal;
    journal_flusher_t *flusher;
    int big_to_flush = 0;
    int write_iodepth = 0;
    bool live = false, queue_stall = false;
@@ -277,7 +285,6 @@ class blockstore_impl_t
    friend class journal_flusher_t;
    friend class journal_flusher_co;
    void parse_config(blockstore_config_t & config);
    void calc_lengths();
    void open_data();
    void open_meta();
@@ -299,6 +306,7 @@ class blockstore_impl_t
    blockstore_init_journal* journal_init_reader;
    void check_wait(blockstore_op_t *op);
    void init_op(blockstore_op_t *op);
    // Read
    int dequeue_read(blockstore_op_t *read_op);
@@ -318,7 +326,7 @@ class blockstore_impl_t
    void handle_write_event(ring_data_t *data, blockstore_op_t *op);
    // Sync
-    int continue_sync(blockstore_op_t *op, bool queue_has_in_progress_sync);
+    int continue_sync(blockstore_op_t *op);
    void ack_sync(blockstore_op_t *op);
    // Stabilize
@@ -326,6 +334,8 @@ class blockstore_impl_t
    int continue_stable(blockstore_op_t *op);
    void mark_stable(const obj_ver_id & ov, bool forget_dirty = false);
    void stabilize_object(object_id oid, uint64_t max_ver);
    blockstore_op_t* selective_sync(blockstore_op_t *op);
    int split_stab_op(blockstore_op_t *op, std::function<int(obj_ver_id v)> decider);
    // Rollback
    int dequeue_rollback(blockstore_op_t *op);
@@ -341,6 +351,8 @@ public:
    blockstore_impl_t(blockstore_config_t & config, ring_loop_t *ringloop, timerfd_manager_t *tfd);
    ~blockstore_impl_t();
    void parse_config(blockstore_config_t & config, bool init);
    // Event loop
    void loop();
--- a/src/blockstore_open.cpp
+++ b/src/blockstore_open.cpp
@@ -4,8 +4,54 @@
 #include <sys/file.h>
 #include "blockstore_impl.h"
-void blockstore_impl_t::parse_config(blockstore_config_t & config)
+void blockstore_impl_t::parse_config(blockstore_config_t & config, bool init)
 {
    // Online-configurable options:
    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);
    }
    min_flusher_count = strtoull(config["min_flusher_count"].c_str(), NULL, 10);
    max_write_iodepth = strtoull(config["max_write_iodepth"].c_str(), NULL, 10);
    throttle_small_writes = config["throttle_small_writes"] == "true" || config["throttle_small_writes"] == "1" || config["throttle_small_writes"] == "yes";
    throttle_target_iops = strtoull(config["throttle_target_iops"].c_str(), NULL, 10);
    throttle_target_mbs = strtoull(config["throttle_target_mbs"].c_str(), NULL, 10);
    throttle_target_parallelism = strtoull(config["throttle_target_parallelism"].c_str(), NULL, 10);
    throttle_threshold_us = strtoull(config["throttle_threshold_us"].c_str(), NULL, 10);
    if (!max_flusher_count)
    {
        max_flusher_count = 256;
    }
    if (!min_flusher_count || journal.flush_journal)
    {
        min_flusher_count = 1;
    }
    if (!max_write_iodepth)
    {
        max_write_iodepth = 128;
    }
    if (!throttle_target_iops)
    {
        throttle_target_iops = 100;
    }
    if (!throttle_target_mbs)
    {
        throttle_target_mbs = 100;
    }
    if (!throttle_target_parallelism)
    {
        throttle_target_parallelism = 1;
    }
    if (!throttle_threshold_us)
    {
        throttle_threshold_us = 50;
    }
    if (!init)
    {
        return;
    }
    // Offline-configurable options:
    // Common disk options
    dsk.parse_config(config);
    // Parse
@@ -44,29 +90,7 @@ void blockstore_impl_t::parse_config(blockstore_config_t & config)
    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";
    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);
    min_flusher_count = strtoull(config["min_flusher_count"].c_str(), NULL, 10);
    max_write_iodepth = strtoull(config["max_write_iodepth"].c_str(), NULL, 10);
    throttle_small_writes = config["throttle_small_writes"] == "true" || config["throttle_small_writes"] == "1" || config["throttle_small_writes"] == "yes";
    throttle_target_iops = strtoull(config["throttle_target_iops"].c_str(), NULL, 10);
    throttle_target_mbs = strtoull(config["throttle_target_mbs"].c_str(), NULL, 10);
    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 (!max_flusher_count)
    {
        max_flusher_count = 256;
    }
    if (!min_flusher_count || journal.flush_journal)
    {
        min_flusher_count = 1;
    }
    if (!max_write_iodepth)
    {
        max_write_iodepth = 128;
    }
    if (journal.sector_count < 2)
    {
        journal.sector_count = 32;
@@ -91,22 +115,6 @@ void blockstore_impl_t::parse_config(blockstore_config_t & config)
    {
        throw std::runtime_error("immediate_commit=all requires disable_journal_fsync and disable_data_fsync");
    }
    if (!throttle_target_iops)
    {
        throttle_target_iops = 100;
    }
    if (!throttle_target_mbs)
    {
        throttle_target_mbs = 100;
    }
    if (!throttle_target_parallelism)
    {
        throttle_target_parallelism = 1;
    }
    if (!throttle_threshold_us)
    {
        throttle_threshold_us = 50;
    }
    // init some fields
    journal.block_size = dsk.journal_block_size;
    journal.next_free = dsk.journal_block_size;
--- a/src/blockstore_rollback.cpp
+++ b/src/blockstore_rollback.cpp
@@ -9,48 +9,39 @@ int blockstore_impl_t::dequeue_rollback(blockstore_op_t *op)
    {
        return continue_rollback(op);
    }
-    obj_ver_id *v, *nv;
+    int r = split_stab_op(op, [this](obj_ver_id ov)
    int i, todo = op->len;
    for (i = 0, v = (obj_ver_id*)op->buf, nv = (obj_ver_id*)op->buf; i < op->len; i++, v++, nv++)
    {
        if (nv != v)
        {
            *nv = *v;
        }
        // Check that there are some versions greater than v->version (which may be zero),
        // check that they're unstable, synced, and not currently written to
        auto dirty_it = dirty_db.lower_bound((obj_ver_id){
-            .oid = v->oid,
+            .oid = ov.oid,
            .version = UINT64_MAX,
        });
        if (dirty_it == dirty_db.begin())
        {
 skip_ov:
            // Already rolled back, skip this object version
-            todo--;
+            return STAB_SPLIT_DONE;
            nv--;
            continue;
        }
        else
        {
            dirty_it--;
-            if (dirty_it->first.oid != v->oid || dirty_it->first.version < v->version)
+            if (dirty_it->first.oid != ov.oid || dirty_it->first.version < ov.version)
            {
-                goto skip_ov;
+                // Already rolled back, skip this object version
                return STAB_SPLIT_DONE;
            }
-            while (dirty_it->first.oid == v->oid && dirty_it->first.version > v->version)
+            while (dirty_it->first.oid == ov.oid && dirty_it->first.version > ov.version)
            {
                if (IS_IN_FLIGHT(dirty_it->second.state))
                {
                    // Object write is still in progress. Wait until the write request completes
-                    return 0;
+                    return STAB_SPLIT_WAIT;
                }
                else if (!IS_SYNCED(dirty_it->second.state) ||
                    IS_STABLE(dirty_it->second.state))
                {
-                    op->retval = -EBUSY;
+                    // Sync the object
-                    FINISH_OP(op);
+                    return STAB_SPLIT_SYNC;
                    return 2;
                }
                if (dirty_it == dirty_db.begin())
                {
@@ -58,19 +49,16 @@ skip_ov:
                }
                dirty_it--;
            }
            return STAB_SPLIT_TODO;
        }
-    }
+    });
-    op->len = todo;
+    if (r != 1)
    if (!todo)
    {
-        // Already rolled back
+        return r;
        op->retval = 0;
        FINISH_OP(op);
        return 2;
    }
    // Check journal space
    blockstore_journal_check_t space_check(this);
-    if (!space_check.check_available(op, todo, sizeof(journal_entry_rollback), 0))
+    if (!space_check.check_available(op, op->len, sizeof(journal_entry_rollback), 0))
    {
        return 0;
    }
@@ -78,7 +66,8 @@ skip_ov:
    BS_SUBMIT_CHECK_SQES(space_check.sectors_to_write);
    // Prepare and submit journal entries
    int s = 0;
-    for (i = 0, v = (obj_ver_id*)op->buf; i < op->len; i++, v++)
+    auto v = (obj_ver_id*)op->buf;
    for (int i = 0; i < op->len; i++, v++)
    {
        if (!journal.entry_fits(sizeof(journal_entry_rollback)) &&
            journal.sector_info[journal.cur_sector].dirty)
@@ -212,6 +201,11 @@ void blockstore_impl_t::erase_dirty(blockstore_dirty_db_t::iterator dirty_start,
    }
    while (1)
    {
        if ((IS_BIG_WRITE(dirty_it->second.state) || IS_DELETE(dirty_it->second.state)) &&
            IS_STABLE(dirty_it->second.state))
        {
            big_to_flush--;
        }
        if (IS_BIG_WRITE(dirty_it->second.state) && dirty_it->second.location != clean_loc &&
            dirty_it->second.location != UINT64_MAX)
        {
--- a/src/blockstore_stable.cpp
+++ b/src/blockstore_stable.cpp
@@ -41,60 +41,309 @@
 // 4) after a while it takes his synced object list and sends stabilize requests
 //    to peers and to its own blockstore, thus freeing the old version
-int blockstore_impl_t::dequeue_stable(blockstore_op_t *op)
+struct ver_vector_t
 {
-    if (PRIV(op)->op_state)
+    obj_ver_id *items = NULL;
    uint64_t alloc = 0, size = 0;
 };
 static void init_versions(ver_vector_t & vec, obj_ver_id *start, obj_ver_id *end, uint64_t len)
 {
    if (!vec.items)
    {
-        return continue_stable(op);
+        vec.alloc = len;
        vec.items = (obj_ver_id*)malloc_or_die(sizeof(obj_ver_id) * vec.alloc);
        for (auto sv = start; sv < end; sv++)
        {
            vec.items[vec.size++] = *sv;
        }
    }
 }
 static void append_version(ver_vector_t & vec, obj_ver_id ov)
 {
    if (vec.size >= vec.alloc)
    {
        vec.alloc = !vec.alloc ? 4 : vec.alloc*2;
        vec.items = (obj_ver_id*)realloc_or_die(vec.items, sizeof(obj_ver_id) * vec.alloc);
    }
    vec.items[vec.size++] = ov;
 }
 static bool check_unsynced(std::vector<obj_ver_id> & check, obj_ver_id ov, std::vector<obj_ver_id> & to, int *count)
 {
    bool found = false;
    int j = 0, k = 0;
    while (j < check.size())
    {
        if (check[j] == ov)
            found = true;
        if (check[j].oid == ov.oid && check[j].version <= ov.version)
        {
            to.push_back(check[j++]);
            if (count)
                (*count)--;
        }
        else
            check[k++] = check[j++];
    }
    check.resize(k);
    return found;
 }
 blockstore_op_t* blockstore_impl_t::selective_sync(blockstore_op_t *op)
 {
    unsynced_big_write_count -= unsynced_big_writes.size();
    unsynced_big_writes.swap(PRIV(op)->sync_big_writes);
    unsynced_big_write_count += unsynced_big_writes.size();
    unsynced_small_writes.swap(PRIV(op)->sync_small_writes);
    // Create a sync operation, insert into the end of the queue
    // And move ourselves into the end too!
    // Rather hacky but that's what we need...
    blockstore_op_t *sync_op = new blockstore_op_t;
    sync_op->opcode = BS_OP_SYNC;
    sync_op->buf = NULL;
    sync_op->callback = [this](blockstore_op_t *sync_op)
    {
        delete sync_op;
    };
    init_op(sync_op);
    int sync_res = continue_sync(sync_op);
    if (sync_res != 2)
    {
        // Put SYNC into the queue if it's not finished yet
        submit_queue.push_back(sync_op);
    }
    // Restore unsynced_writes
    unsynced_small_writes.swap(PRIV(op)->sync_small_writes);
    unsynced_big_write_count -= unsynced_big_writes.size();
    unsynced_big_writes.swap(PRIV(op)->sync_big_writes);
    unsynced_big_write_count += unsynced_big_writes.size();
    if (sync_res == 2)
    {
        // Sync is immediately completed
        return NULL;
    }
    return sync_op;
 }
 // Returns: 2 = stop processing and dequeue, 0 = stop processing and do not dequeue, 1 = proceed with op itself
 int blockstore_impl_t::split_stab_op(blockstore_op_t *op, std::function<int(obj_ver_id v)> decider)
 {
    bool add_sync = false;
    ver_vector_t good_vers, bad_vers;
    obj_ver_id* v;
    int i, todo = 0;
    for (i = 0, v = (obj_ver_id*)op->buf; i < op->len; i++, v++)
    {
-        auto dirty_it = dirty_db.find(*v);
+        int action = decider(*v);
-        if (dirty_it == dirty_db.end())
+        if (action < 0)
        {
-            auto & clean_db = clean_db_shard(v->oid);
+            // Rollback changes
-            auto clean_it = clean_db.find(v->oid);
+            for (auto & ov: PRIV(op)->sync_big_writes)
            if (clean_it == clean_db.end() || clean_it->second.version < v->version)
            {
-                // No such object version
+                unsynced_big_writes.push_back(ov);
-                op->retval = -ENOENT;
+                unsynced_big_write_count++;
                FINISH_OP(op);
                return 2;
            }
-            else
+            for (auto & ov: PRIV(op)->sync_small_writes)
            {
-                // Already stable
+                unsynced_small_writes.push_back(ov);
            }
-        }
+            free(good_vers.items);
-        else if (IS_IN_FLIGHT(dirty_it->second.state))
+            good_vers.items = NULL;
-        {
+            free(bad_vers.items);
-            // Object write is still in progress. Wait until the write request completes
+            bad_vers.items = NULL;
-            return 0;
+            // Error
-        }
+            op->retval = action;
        else if (!IS_SYNCED(dirty_it->second.state))
        {
            // Object not synced yet. Caller must sync it first
            op->retval = -EBUSY;
            FINISH_OP(op);
            return 2;
        }
-        else if (!IS_STABLE(dirty_it->second.state))
+        else if (action == STAB_SPLIT_DONE)
        {
            // Already done
            init_versions(good_vers, (obj_ver_id*)op->buf, v, op->len);
        }
        else if (action == STAB_SPLIT_WAIT)
        {
            // Already in progress, we just have to wait until it finishes
            init_versions(good_vers, (obj_ver_id*)op->buf, v, op->len);
            append_version(bad_vers, *v);
        }
        else if (action == STAB_SPLIT_SYNC)
        {
            // Needs a SYNC, we have to send a SYNC if not already in progress
            //
            // If the object is not present in unsynced_(big|small)_writes then
            // it's currently being synced. If it's present then we can initiate
            // its sync ourselves.
            init_versions(good_vers, (obj_ver_id*)op->buf, v, op->len);
            append_version(bad_vers, *v);
            if (!add_sync)
            {
                PRIV(op)->sync_big_writes.clear();
                PRIV(op)->sync_small_writes.clear();
                add_sync = true;
            }
            check_unsynced(unsynced_small_writes, *v, PRIV(op)->sync_small_writes, NULL);
            check_unsynced(unsynced_big_writes, *v, PRIV(op)->sync_big_writes, &unsynced_big_write_count);
        }
        else /* if (action == STAB_SPLIT_TODO) */
        {
            if (good_vers.items)
            {
                // If we're selecting versions then append it
                // Main idea is that 99% of the time all versions passed to BS_OP_STABLE are synced
                // And we don't want to select/allocate anything in that optimistic case
                append_version(good_vers, *v);
            }
            todo++;
        }
    }
-    if (!todo)
+    // In a pessimistic scenario, an operation may be split into 3:
    // - Stabilize synced entries
    // - Sync unsynced entries
    // - Continue for unsynced entries after sync
    add_sync = add_sync && (PRIV(op)->sync_big_writes.size() || PRIV(op)->sync_small_writes.size());
    if (!todo && !bad_vers.size)
    {
        // Already stable
        op->retval = 0;
        FINISH_OP(op);
        return 2;
    }
    op->retval = 0;
    if (!todo && !add_sync)
    {
        // Only wait for inflight writes or current in-progress syncs
        return 0;
    }
    blockstore_op_t *sync_op = NULL, *split_stab_op = NULL;
    if (add_sync)
    {
        // Initiate a selective sync for PRIV(op)->sync_(big|small)_writes
        sync_op = selective_sync(op);
    }
    if (bad_vers.size)
    {
        // Split part of the request into a separate operation
        split_stab_op = new blockstore_op_t;
        split_stab_op->opcode = op->opcode;
        split_stab_op->buf = bad_vers.items;
        split_stab_op->len = bad_vers.size;
        init_op(split_stab_op);
        submit_queue.push_back(split_stab_op);
    }
    if (sync_op || split_stab_op || good_vers.items)
    {
        void *orig_buf = op->buf;
        if (good_vers.items)
        {
            op->buf = good_vers.items;
            op->len = good_vers.size;
        }
        // Make a wrapped callback
        int *split_op_counter = (int*)malloc_or_die(sizeof(int));
        *split_op_counter = (sync_op ? 1 : 0) + (split_stab_op ? 1 : 0) + (todo ? 1 : 0);
        auto cb = [this, op, good_items = good_vers.items,
            bad_items = bad_vers.items, split_op_counter,
            orig_buf, real_cb = op->callback](blockstore_op_t *split_op)
        {
            if (split_op->retval != 0)
                op->retval = split_op->retval;
            (*split_op_counter)--;
            assert((*split_op_counter) >= 0);
            if (op != split_op)
                delete split_op;
            if (!*split_op_counter)
            {
                free(good_items);
                free(bad_items);
                free(split_op_counter);
                op->buf = orig_buf;
                real_cb(op);
            }
        };
        if (sync_op)
        {
            sync_op->callback = cb;
        }
        if (split_stab_op)
        {
            split_stab_op->callback = cb;
        }
        op->callback = cb;
    }
    if (!todo)
    {
        // All work is postponed
        op->callback = NULL;
        return 2;
    }
    return 1;
 }
 int blockstore_impl_t::dequeue_stable(blockstore_op_t *op)
 {
    if (PRIV(op)->op_state)
    {
        return continue_stable(op);
    }
    int r = split_stab_op(op, [this](obj_ver_id ov)
    {
        auto dirty_it = dirty_db.find(ov);
        if (dirty_it == dirty_db.end())
        {
            auto & clean_db = clean_db_shard(ov.oid);
            auto clean_it = clean_db.find(ov.oid);
            if (clean_it == clean_db.end() || clean_it->second.version < ov.version)
            {
                // No such object version
                printf("Error: %lx:%lx v%lu not found while stabilizing\n", ov.oid.inode, ov.oid.stripe, ov.version);
                return -ENOENT;
            }
            else
            {
                // Already stable
                return STAB_SPLIT_DONE;
            }
        }
        else if (IS_IN_FLIGHT(dirty_it->second.state))
        {
            // Object write is still in progress. Wait until the write request completes
            return STAB_SPLIT_WAIT;
        }
        else if (!IS_SYNCED(dirty_it->second.state))
        {
            // Object not synced yet - sync it
            // In previous versions we returned EBUSY here and required
            // the caller (OSD) to issue a global sync first. But a global sync
            // waits for all writes in the queue including inflight writes. And
            // inflight writes may themselves be blocked by unstable writes being
            // still present in the journal and not flushed away from it.
            // So we must sync specific objects here.
            //
            // Even more, we have to process "stabilize" request in parts. That is,
            // we must stabilize all objects which are already synced. Otherwise
            // they may block objects which are NOT synced yet.
            return STAB_SPLIT_SYNC;
        }
        else if (IS_STABLE(dirty_it->second.state))
        {
            // Already stable
            return STAB_SPLIT_DONE;
        }
        else
        {
            return STAB_SPLIT_TODO;
        }
    });
    if (r != 1)
    {
        return r;
    }
    // Check journal space
    blockstore_journal_check_t space_check(this);
-    if (!space_check.check_available(op, todo, sizeof(journal_entry_stable), 0))
+    if (!space_check.check_available(op, op->len, sizeof(journal_entry_stable), 0))
    {
        return 0;
    }
@@ -102,9 +351,9 @@ int blockstore_impl_t::dequeue_stable(blockstore_op_t *op)
    BS_SUBMIT_CHECK_SQES(space_check.sectors_to_write);
    // Prepare and submit journal entries
    int s = 0;
-    for (i = 0, v = (obj_ver_id*)op->buf; i < op->len; i++, v++)
+    auto v = (obj_ver_id*)op->buf;
    for (int i = 0; i < op->len; i++, v++)
    {
        // FIXME: Only stabilize versions that aren't stable yet
        if (!journal.entry_fits(sizeof(journal_entry_stable)) &&
            journal.sector_info[journal.cur_sector].dirty)
        {
@@ -197,6 +446,7 @@ void blockstore_impl_t::mark_stable(const obj_ver_id & v, bool forget_dirty)
                    {
                        inode_space_stats[dirty_it->first.oid.inode] += dsk.data_block_size;
                    }
                    big_to_flush++;
                }
                else if (IS_DELETE(dirty_it->second.state))
                {
@@ -205,6 +455,7 @@ void blockstore_impl_t::mark_stable(const obj_ver_id & v, bool forget_dirty)
                        sp -= dsk.data_block_size;
                    else
                        inode_space_stats.erase(dirty_it->first.oid.inode);
                    big_to_flush++;
                }
            }
            if (forget_dirty && (IS_BIG_WRITE(dirty_it->second.state) ||
--- a/src/blockstore_sync.cpp
+++ b/src/blockstore_sync.cpp
@@ -12,7 +12,7 @@
 #define SYNC_JOURNAL_SYNC_SENT 7
 #define SYNC_DONE 8
-int blockstore_impl_t::continue_sync(blockstore_op_t *op, bool queue_has_in_progress_sync)
+int blockstore_impl_t::continue_sync(blockstore_op_t *op)
 {
    if (immediate_commit == IMMEDIATE_ALL)
    {
@@ -145,7 +145,7 @@ int blockstore_impl_t::continue_sync(blockstore_op_t *op, bool queue_has_in_prog
            PRIV(op)->op_state = SYNC_DONE;
        }
    }
-    if (PRIV(op)->op_state == SYNC_DONE && !queue_has_in_progress_sync)
+    if (PRIV(op)->op_state == SYNC_DONE)
    {
        ack_sync(op);
        return 2;
--- a/src/blockstore_write.cpp
+++ b/src/blockstore_write.cpp
@@ -271,6 +271,13 @@ int blockstore_impl_t::dequeue_write(blockstore_op_t *op)
        if (loc == UINT64_MAX)
        {
            // no space
            if (big_to_flush > 0)
            {
                // hope that some space will be available after flush
                flusher->request_trim();
                PRIV(op)->wait_for = WAIT_FREE;
                return 0;
            }
            cancel_all_writes(op, dirty_it, -ENOSPC);
            return 2;
        }
--- a/src/cli_df.cpp
+++ b/src/cli_df.cpp
@@ -121,8 +121,7 @@ resume_1:
            }
            if (pool_cfg.scheme != POOL_SCHEME_REPLICATED)
            {
-                uint64_t pg_real_size = pool_stats[pool_cfg.id]["pg_real_size"].uint64_value();
+                pool_avail *= (pool_cfg.pg_size - pool_cfg.parity_chunks);
                pool_avail = pg_real_size > 0 ? pool_avail * (pool_cfg.pg_size - pool_cfg.parity_chunks) / pg_real_size : 0;
            }
            pool_stats[pool_cfg.id] = json11::Json::object {
                { "name", pool_cfg.name },
--- a/src/cli_rm_data.cpp
+++ b/src/cli_rm_data.cpp
@@ -92,6 +92,7 @@ struct rm_inode_t
    void send_ops(rm_pg_t *cur_list)
    {
        parent->cli->init_msgr();
        if (parent->cli->msgr.osd_peer_fds.find(cur_list->rm_osd_num) ==
            parent->cli->msgr.osd_peer_fds.end())
        {
--- a/src/cli_rm_osd.cpp
+++ b/src/cli_rm_osd.cpp
@@ -278,7 +278,7 @@ struct rm_osd_t
            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();
+                retry_wait = parent->cli->config["mon_change_timeout"].uint64_value();
                if (!retry_wait)
                    retry_wait = 1000;
                retry_wait += etcd_tx_retry_ms;
--- a/src/cli_status.cpp
+++ b/src/cli_status.cpp
@@ -198,9 +198,9 @@ resume_2:
            }
            pgs_by_state_str += std::to_string(kv.second)+" "+kv.first;
        }
-        bool readonly = json_is_true(parent->cli->merged_config["readonly"]);
+        bool readonly = json_is_true(parent->cli->config["readonly"]);
-        bool no_recovery = json_is_true(parent->cli->merged_config["no_recovery"]);
+        bool no_recovery = json_is_true(parent->cli->config["no_recovery"]);
-        bool no_rebalance = json_is_true(parent->cli->merged_config["no_rebalance"]);
+        bool no_rebalance = json_is_true(parent->cli->config["no_rebalance"]);
        if (parent->json_output)
        {
            // JSON output
--- a/src/cluster_client.cpp
+++ b/src/cluster_client.cpp
@@ -18,11 +18,12 @@
 cluster_client_t::cluster_client_t(ring_loop_t *ringloop, timerfd_manager_t *tfd, json11::Json & config)
 {
-    config = osd_messenger_t::read_config(config);
+    cli_config = config.object_items();
    file_config = osd_messenger_t::read_config(config);
    config = osd_messenger_t::merge_configs(cli_config, file_config, etcd_global_config, {});
    this->ringloop = ringloop;
    this->tfd = tfd;
    this->config = config;
    msgr.osd_num = 0;
    msgr.tfd = tfd;
@@ -58,8 +59,7 @@ cluster_client_t::cluster_client_t(ring_loop_t *ringloop, timerfd_manager_t *tfd
        msgr.stop_client(op->peer_fd);
        delete op;
    };
-    msgr.parse_config(this->config);
+    msgr.parse_config(config);
    msgr.init();
    st_cli.tfd = tfd;
    st_cli.on_load_config_hook = [this](json11::Json::object & cfg) { on_load_config_hook(cfg); };
@@ -73,17 +73,6 @@ cluster_client_t::cluster_client_t(ring_loop_t *ringloop, timerfd_manager_t *tfd
    scrap_buffer_size = SCRAP_BUFFER_SIZE;
    scrap_buffer = malloc_or_die(scrap_buffer_size);
    if (ringloop)
    {
        consumer.loop = [this]()
        {
            msgr.read_requests();
            msgr.send_replies();
            this->ringloop->submit();
        };
        ringloop->register_consumer(&consumer);
    }
 }
 cluster_client_t::~cluster_client_t()
@@ -115,6 +104,24 @@ cluster_op_t::~cluster_op_t()
    }
 }
 void cluster_client_t::init_msgr()
 {
    if (msgr_initialized)
        return;
    msgr.init();
    msgr_initialized = true;
    if (ringloop)
    {
        consumer.loop = [this]()
        {
            msgr.read_requests();
            msgr.send_replies();
            this->ringloop->submit();
        };
        ringloop->register_consumer(&consumer);
    }
 }
 void cluster_client_t::calc_wait(cluster_op_t *op)
 {
    op->prev_wait = 0;
@@ -223,11 +230,14 @@ 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_FLUSH_BUFFER)
        std::function<void(cluster_op_t*)>(op->callback)(op);
    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 (!(flags & OP_FLUSH_BUFFER))
        std::function<void(cluster_op_t*)>(op->callback)(op);
 }
 void cluster_client_t::continue_ops(bool up_retry)
@@ -267,13 +277,10 @@ restart:
    continuing_ops = 0;
 }
-void cluster_client_t::on_load_config_hook(json11::Json::object & config)
+void cluster_client_t::on_load_config_hook(json11::Json::object & etcd_global_config)
 {
-    this->merged_config = config;
+    this->etcd_global_config = etcd_global_config;
-    for (auto & kv: this->config.object_items())
+    config = osd_messenger_t::merge_configs(cli_config, file_config, etcd_global_config, {});
    {
        this->merged_config[kv.first] = kv.second;
    }
    if (config.find("client_max_dirty_bytes") != config.end())
    {
        client_max_dirty_bytes = config["client_max_dirty_bytes"].uint64_value();
@@ -283,14 +290,13 @@ void cluster_client_t::on_load_config_hook(json11::Json::object & config)
        // Old name
        client_max_dirty_bytes = config["client_dirty_limit"].uint64_value();
    }
-    if (config.find("client_max_dirty_ops") != config.end())
+    else
-    {
+        client_max_dirty_bytes = 0;
        client_max_dirty_ops = config["client_max_dirty_ops"].uint64_value();
    }
    if (!client_max_dirty_bytes)
    {
        client_max_dirty_bytes = DEFAULT_CLIENT_MAX_DIRTY_BYTES;
    }
    client_max_dirty_ops = config["client_max_dirty_ops"].uint64_value();
    if (!client_max_dirty_ops)
    {
        client_max_dirty_ops = DEFAULT_CLIENT_MAX_DIRTY_OPS;
@@ -305,7 +311,7 @@ void cluster_client_t::on_load_config_hook(json11::Json::object & config)
        up_wait_retry_interval = 50;
    }
    msgr.parse_config(config);
-    msgr.parse_config(this->config);
+    st_cli.parse_config(config);
    st_cli.load_pgs();
 }
@@ -737,15 +743,16 @@ resume_3:
        erase_op(op);
        return 1;
    }
-    else if (op->retval != 0 && op->retval != -EPIPE)
+    else if (op->retval != 0 && op->retval != -EPIPE && op->retval != -EIO && op->retval != -ENOSPC)
    {
-        // Fatal error (not -EPIPE)
+        // Fatal error (neither -EPIPE, -EIO nor -ENOSPC)
        // FIXME: Add a parameter to allow to not wait for EIOs (incomplete or corrupted objects) to heal
        erase_op(op);
        return 1;
    }
    else
    {
-        // -EPIPE - clear the error and retry
+        // Non-fatal error - clear the error and retry
        op->retval = 0;
        if (op->needs_reslice)
        {
@@ -757,7 +764,10 @@ resume_3:
        {
            for (int i = 0; i < op->parts.size(); i++)
            {
-                op->parts[i].flags = PART_RETRY;
+                if (!(op->parts[i].flags & PART_DONE))
                {
                    op->parts[i].flags = PART_RETRY;
                }
            }
            goto resume_2;
        }
@@ -915,6 +925,10 @@ bool cluster_client_t::affects_osd(uint64_t inode, uint64_t offset, uint64_t len
 bool cluster_client_t::try_send(cluster_op_t *op, int i)
 {
    if (!msgr_initialized)
    {
        init_msgr();
    }
    auto part = &op->parts[i];
    auto & pool_cfg = st_cli.pool_config.at(INODE_POOL(op->cur_inode));
    auto pg_it = pool_cfg.pg_config.find(part->pg_num);
@@ -1035,7 +1049,7 @@ resume_1:
                uw_it->second.state = CACHE_DIRTY;
            }
        }
-        if (op->retval == -EPIPE)
+        if (op->retval == -EPIPE || op->retval == -EIO || op->retval == -ENOSPC)
        {
            // Retry later
            op->parts.clear();
@@ -1105,33 +1119,52 @@ void cluster_client_t::handle_op_part(cluster_op_part_t *part)
    if (part->op.reply.hdr.retval != expected)
    {
        // Operation failed, retry
-        if (part->op.reply.hdr.retval == -EPIPE)
+        part->flags |= PART_ERROR;
        if (!op->retval || op->retval == -EPIPE || part->op.reply.hdr.retval == -EIO)
        {
-            // Mark op->up_wait = true before stopping the client
+            // Error priority: EIO > ENOSPC > EPIPE
            op->up_wait = true;
            if (!retry_timeout_id)
            {
                retry_timeout_id = tfd->set_timer(up_wait_retry_interval, false, [this](int)
                {
                    retry_timeout_id = 0;
                    continue_ops(true);
                });
            }
        }
        if (!op->retval || op->retval == -EPIPE)
        {
            // Don't overwrite other errors with -EPIPE
            op->retval = part->op.reply.hdr.retval;
        }
-        if (op->retval != -EINTR && op->retval != -EIO)
+        int stop_fd = -1;
        if (op->retval != -EINTR && op->retval != -EIO && op->retval != -ENOSPC)
        {
            stop_fd = part->op.peer_fd;
            fprintf(
                stderr, "%s operation failed on OSD %lu: retval=%ld (expected %d), dropping connection\n",
                osd_op_names[part->op.req.hdr.opcode], part->osd_num, part->op.reply.hdr.retval, expected
            );
            msgr.stop_client(part->op.peer_fd);
        }
-        part->flags |= PART_ERROR;
+        else
        {
            fprintf(
                stderr, "%s operation failed on OSD %lu: retval=%ld (expected %d)\n",
                osd_op_names[part->op.req.hdr.opcode], part->osd_num, part->op.reply.hdr.retval, expected
            );
        }
        // All next things like timer, continue_sync/rw and stop_client may affect the operation again
        // So do all these things after modifying operation state, otherwise we may hit reenterability bugs
        // FIXME postpone such things to set_immediate here to avoid bugs
        // Mark op->up_wait = true to retry operation after a short pause (not immediately)
        op->up_wait = true;
        if (!retry_timeout_id)
        {
            retry_timeout_id = tfd->set_timer(up_wait_retry_interval, false, [this](int)
            {
                retry_timeout_id = 0;
                continue_ops(true);
            });
        }
        if (op->inflight_count == 0)
        {
            if (op->opcode == OSD_OP_SYNC)
                continue_sync(op);
            else
                continue_rw(op);
        }
        if (stop_fd >= 0)
        {
            msgr.stop_client(stop_fd);
        }
    }
    else
    {
@@ -1145,13 +1178,13 @@ void cluster_client_t::handle_op_part(cluster_op_part_t *part)
            copy_part_bitmap(op, part);
            op->version = op->parts.size() == 1 ? part->op.reply.rw.version : 0;
        }
-    }
+        if (op->inflight_count == 0)
-    if (op->inflight_count == 0)
+        {
-    {
+            if (op->opcode == OSD_OP_SYNC)
-        if (op->opcode == OSD_OP_SYNC)
+                continue_sync(op);
-            continue_sync(op);
+            else
-        else
+                continue_rw(op);
-            continue_rw(op);
+        }
    }
 }
--- a/src/cluster_client.h
+++ b/src/cluster_client.h
@@ -104,12 +104,16 @@ class cluster_client_t
    std::vector<std::function<void(void)>> on_ready_hooks;
    std::vector<inode_list_t*> lists;
    int continuing_ops = 0;
    bool msgr_initialized = false;
 public:
    etcd_state_client_t st_cli;
    osd_messenger_t msgr;
-    json11::Json config;
+    void init_msgr();
-    json11::Json::object merged_config;
+
    json11::Json::object cli_config, file_config, etcd_global_config;
    json11::Json::object config;
    cluster_client_t(ring_loop_t *ringloop, timerfd_manager_t *tfd, json11::Json & config);
    ~cluster_client_t();
--- a/src/cluster_client_list.cpp
+++ b/src/cluster_client_list.cpp
@@ -43,6 +43,7 @@ struct inode_list_t
 inode_list_t* cluster_client_t::list_inode_start(inode_t inode,
    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)
 {
    init_msgr();
    int skipped_pgs = 0;
    pool_id_t pool_id = INODE_POOL(inode);
    if (!pool_id || st_cli.pool_config.find(pool_id) == st_cli.pool_config.end())
--- a/src/disk_tool_journal.cpp
+++ b/src/disk_tool_journal.cpp
@@ -281,7 +281,7 @@ void disk_tool_t::dump_journal_entry(int num, journal_entry *je, bool json)
        if (je->big_write.size > sizeof(journal_entry_big_write))
        {
            printf(json ? ",\"bitmap\":\"" : " (bitmap: ");
-            for (int i = sizeof(journal_entry_big_write); i < je->small_write.size; i++)
+            for (int i = sizeof(journal_entry_big_write); i < je->big_write.size; i++)
            {
                printf("%02x", ((uint8_t*)je)[i]);
            }
--- a/src/disk_tool_meta.cpp
+++ b/src/disk_tool_meta.cpp
@@ -26,7 +26,7 @@ int disk_tool_t::process_meta(std::function<void(blockstore_meta_header_v1_t *)>
        buf_size = dsk.meta_len;
    void *data = memalign_or_die(MEM_ALIGNMENT, buf_size);
    lseek64(dsk.meta_fd, dsk.meta_offset, 0);
-    read_blocking(dsk.meta_fd, data, buf_size);
+    read_blocking(dsk.meta_fd, data, dsk.meta_block_size);
    // Check superblock
    blockstore_meta_header_v1_t *hdr = (blockstore_meta_header_v1_t *)data;
    if (hdr->zero == 0 &&
@@ -41,8 +41,11 @@ int disk_tool_t::process_meta(std::function<void(blockstore_meta_header_v1_t *)>
            if (buf_size % dsk.meta_block_size)
            {
                buf_size = 8*dsk.meta_block_size;
                void *new_data = memalign_or_die(MEM_ALIGNMENT, buf_size);
                memcpy(new_data, data, dsk.meta_block_size);
                free(data);
-                data = memalign_or_die(MEM_ALIGNMENT, buf_size);
+                data = new_data;
                hdr = (blockstore_meta_header_v1_t *)data;
            }
        }
        dsk.bitmap_granularity = hdr->bitmap_granularity;
--- a/src/disk_tool_prepare.cpp
+++ b/src/disk_tool_prepare.cpp
@@ -621,6 +621,11 @@ int disk_tool_t::prepare(std::vector<std::string> devices)
                }
                // Treat all disks as SSDs if not in the hybrid mode
                prepare_one(options, hybrid && dev.is_hdd ? 1 : 0);
                if (hybrid)
                {
                    options.erase("journal_device");
                    options.erase("meta_device");
                }
            }
        }
    }
--- a/src/disk_tool_utils.cpp
+++ b/src/disk_tool_utils.cpp
@@ -305,10 +305,10 @@ int write_zero(int fd, uint64_t offset, uint64_t size)
 json11::Json read_parttable(std::string dev)
 {
    std::string part_dump;
-    int r = shell_exec({ "sfdisk", "--dump", dev, "--json" }, "", &part_dump, NULL);
+    int r = shell_exec({ "sfdisk", "--json", dev }, "", &part_dump, NULL);
    if (r == 255)
    {
-        fprintf(stderr, "Error running sfdisk --dump %s --json\n", dev.c_str());
+        fprintf(stderr, "Error running sfdisk --json %s\n", dev.c_str());
        return json11::Json(false);
    }
    // Decode partition table
@@ -319,7 +319,7 @@ json11::Json read_parttable(std::string dev)
        pt = json11::Json::parse(part_dump, err);
        if (err != "")
        {
-            fprintf(stderr, "sfdisk --dump %s --json returned bad JSON: %s\n", dev.c_str(), part_dump.c_str());
+            fprintf(stderr, "sfdisk --json %s returned bad JSON: %s\n", dev.c_str(), part_dump.c_str());
            return json11::Json(false);
        }
        pt = pt["partitiontable"];
--- a/src/epoll_manager.cpp
+++ b/src/epoll_manager.cpp
@@ -54,6 +54,13 @@ void epoll_manager_t::set_fd_handler(int fd, bool wr, std::function<void(int, in
        ev.events = (wr ? EPOLLOUT : 0) | EPOLLIN | EPOLLRDHUP | EPOLLET;
        if (epoll_ctl(epoll_fd, exists ? EPOLL_CTL_MOD : EPOLL_CTL_ADD, fd, &ev) < 0)
        {
            if (errno == ENOENT)
            {
                // The FD is probably already closed
                epoll_ctl(epoll_fd, EPOLL_CTL_DEL, fd, NULL);
                epoll_handlers.erase(fd);
                return;
            }
            throw std::runtime_error(std::string("epoll_ctl: ") + strerror(errno));
        }
        epoll_handlers[fd] = handler;
--- a/src/etcd_state_client.cpp
+++ b/src/etcd_state_client.cpp
@@ -18,12 +18,8 @@ etcd_state_client_t::~etcd_state_client_t()
    }
    watches.clear();
    etcd_watches_initialised = -1;
    if (ws_keepalive_timer >= 0)
    {
        tfd->clear_timer(ws_keepalive_timer);
        ws_keepalive_timer = -1;
    }
 #ifndef __MOCK__
    stop_ws_keepalive();
    if (etcd_watch_ws)
    {
        http_close(etcd_watch_ws);
@@ -245,6 +241,7 @@ void etcd_state_client_t::parse_config(const json11::Json & config)
        if (this->etcd_keepalive_timeout < 30)
            this->etcd_keepalive_timeout = 30;
    }
    auto old_etcd_ws_keepalive_interval = this->etcd_ws_keepalive_interval;
    this->etcd_ws_keepalive_interval = config["etcd_ws_keepalive_interval"].uint64_value();
    if (this->etcd_ws_keepalive_interval <= 0)
    {
@@ -265,6 +262,13 @@ void etcd_state_client_t::parse_config(const json11::Json & config)
    {
        this->etcd_quick_timeout = 1000;
    }
    if (this->etcd_ws_keepalive_interval != old_etcd_ws_keepalive_interval && ws_keepalive_timer >= 0)
    {
 #ifndef __MOCK__
        stop_ws_keepalive();
        start_ws_keepalive();
 #endif
    }
 }
 void etcd_state_client_t::pick_next_etcd()
@@ -478,6 +482,20 @@ void etcd_state_client_t::start_etcd_watcher()
    {
        on_start_watcher_hook(etcd_watch_ws);
    }
    start_ws_keepalive();
 }
 void etcd_state_client_t::stop_ws_keepalive()
 {
    if (ws_keepalive_timer >= 0)
    {
        tfd->clear_timer(ws_keepalive_timer);
        ws_keepalive_timer = -1;
    }
 }
 void etcd_state_client_t::start_ws_keepalive()
 {
    if (ws_keepalive_timer < 0)
    {
        ws_keepalive_timer = tfd->set_timer(etcd_ws_keepalive_interval*1000, true, [this](int)
--- a/src/etcd_state_client.h
+++ b/src/etcd_state_client.h
@@ -132,6 +132,8 @@ public:
    void etcd_txn(json11::Json txn, int timeout, int retries, int interval, std::function<void(std::string, json11::Json)> callback);
    void etcd_txn_slow(json11::Json txn, std::function<void(std::string, json11::Json)> callback);
    void start_etcd_watcher();
    void stop_ws_keepalive();
    void start_ws_keepalive();
    void load_global_config();
    void load_pgs();
    void parse_state(const etcd_kv_t & kv);
--- a/src/messenger.cpp
+++ b/src/messenger.cpp
@@ -157,10 +157,10 @@ void osd_messenger_t::parse_config(const json11::Json & config)
        this->rdma_max_sge = 128;
    this->rdma_max_send = config["rdma_max_send"].uint64_value();
    if (!this->rdma_max_send)
-        this->rdma_max_send = 1;
+        this->rdma_max_send = 8;
    this->rdma_max_recv = config["rdma_max_recv"].uint64_value();
    if (!this->rdma_max_recv)
-        this->rdma_max_recv = 128;
+        this->rdma_max_recv = 16;
    this->rdma_max_msg = config["rdma_max_msg"].uint64_value();
    if (!this->rdma_max_msg || this->rdma_max_msg > 128*1024*1024)
        this->rdma_max_msg = 129*1024;
@@ -534,8 +534,9 @@ bool osd_messenger_t::is_rdma_enabled()
 }
 #endif
-json11::Json osd_messenger_t::read_config(const json11::Json & config)
+json11::Json::object osd_messenger_t::read_config(const json11::Json & config)
 {
    json11::Json::object file_config;
    const char *config_path = config["config_path"].string_value() != ""
        ? config["config_path"].string_value().c_str() : VITASTOR_CONFIG_PATH;
    int fd = open(config_path, O_RDONLY);
@@ -543,14 +544,14 @@ json11::Json osd_messenger_t::read_config(const json11::Json & config)
    {
        if (errno != ENOENT)
            fprintf(stderr, "Error reading %s: %s\n", config_path, strerror(errno));
-        return config;
+        return file_config;
    }
    struct stat st;
    if (fstat(fd, &st) != 0)
    {
        fprintf(stderr, "Error reading %s: %s\n", config_path, strerror(errno));
        close(fd);
-        return config;
+        return file_config;
    }
    std::string buf;
    buf.resize(st.st_size);
@@ -562,23 +563,125 @@ json11::Json osd_messenger_t::read_config(const json11::Json & config)
        {
            fprintf(stderr, "Error reading %s: %s\n", config_path, strerror(errno));
            close(fd);
-            return config;
+            return file_config;
        }
        done += r;
    }
    close(fd);
    std::string json_err;
-    json11::Json::object file_config = json11::Json::parse(buf, json_err).object_items();
+    file_config = json11::Json::parse(buf, json_err).object_items();
    if (json_err != "")
    {
        fprintf(stderr, "Invalid JSON in %s: %s\n", config_path, json_err.c_str());
        return config;
    }
    file_config.erase("config_path");
    file_config.erase("osd_num");
    for (auto kv: config.object_items())
    {
        file_config[kv.first] = kv.second;
    }
    return file_config;
 }
 static const char* cli_only_params[] = {
    // The list has to be sorted
    "bitmap_granularity",
    "block_size",
    "data_device",
    "data_offset",
    "data_size",
    "disable_data_fsync",
    "disable_device_lock",
    "disable_journal_fsync",
    "disable_meta_fsync",
    "disk_alignment",
    "flush_journal",
    "immediate_commit",
    "inmemory_journal",
    "inmemory_metadata",
    "journal_block_size",
    "journal_device",
    "journal_no_same_sector_overwrites",
    "journal_offset",
    "journal_sector_buffer_count",
    "journal_size",
    "meta_block_size",
    "meta_buf_size",
    "meta_device",
    "meta_offset",
    "osd_num",
    "readonly",
 };
 static const char **cli_only_end = cli_only_params + (sizeof(cli_only_params)/sizeof(cli_only_params[0]));
 static const char* local_only_params[] = {
    // The list has to be sorted
    "config_path",
    "rdma_device",
    "rdma_gid_index",
    "rdma_max_msg",
    "rdma_max_recv",
    "rdma_max_send",
    "rdma_max_sge",
    "rdma_mtu",
    "rdma_port_num",
    "tcp_header_buffer_size",
    "use_rdma",
    "use_sync_send_recv",
 };
 static const char **local_only_end = local_only_params + (sizeof(local_only_params)/sizeof(local_only_params[0]));
 // Basically could be replaced by std::lower_bound()...
 static int find_str_array(const char **start, const char **end, const std::string & s)
 {
    int min = 0, max = end-start;
    while (max-min >= 2)
    {
        int mid = (min+max)/2;
        int r = strcmp(s.c_str(), start[mid]);
        if (r < 0)
            max = mid;
        else if (r > 0)
            min = mid;
        else
            return mid;
    }
    if (min < end-start && !strcmp(s.c_str(), start[min]))
        return min;
    return -1;
 }
 json11::Json::object osd_messenger_t::merge_configs(const json11::Json::object & cli_config,
    const json11::Json::object & file_config,
    const json11::Json::object & etcd_global_config,
    const json11::Json::object & etcd_osd_config)
 {
    // Priority: most important -> less important:
    // etcd_osd_config -> cli_config -> etcd_global_config -> file_config
    json11::Json::object res = file_config;
    for (auto & kv: file_config)
    {
        int cli_only = find_str_array(cli_only_params, cli_only_end, kv.first);
        if (cli_only < 0)
        {
            res[kv.first] = kv.second;
        }
    }
    for (auto & kv: etcd_global_config)
    {
        int local_only = find_str_array(local_only_params, local_only_end, kv.first);
        if (local_only < 0)
        {
            res[kv.first] = kv.second;
        }
    }
    for (auto & kv: cli_config)
    {
        res[kv.first] = kv.second;
    }
    for (auto & kv: etcd_osd_config)
    {
        int local_only = find_str_array(local_only_params, local_only_end, kv.first);
        if (local_only < 0)
        {
            res[kv.first] = kv.second;
        }
    }
    return res;
 }
--- a/src/messenger.h
+++ b/src/messenger.h
@@ -138,6 +138,7 @@ protected:
    std::vector<int> read_ready_clients;
    std::vector<int> write_ready_clients;
    // We don't use ringloop->set_immediate here because we may have no ringloop in client :)
    std::vector<std::function<void()>> set_immediate;
 public:
@@ -165,7 +166,11 @@ public:
    void accept_connections(int listen_fd);
    ~osd_messenger_t();
-    static json11::Json read_config(const json11::Json & config);
+    static json11::Json::object read_config(const json11::Json & config);
    static json11::Json::object merge_configs(const json11::Json::object & cli_config,
        const json11::Json::object & file_config,
        const json11::Json::object & etcd_global_config,
        const json11::Json::object & etcd_osd_config);
 #ifdef WITH_RDMA
    bool is_rdma_enabled();
--- a/src/mock/messenger.cpp
+++ b/src/mock/messenger.cpp
@@ -43,7 +43,15 @@ void osd_messenger_t::send_replies()
 {
 }
-json11::Json osd_messenger_t::read_config(const json11::Json & config)
+json11::Json::object osd_messenger_t::read_config(const json11::Json & config)
 {
-    return config;
+    return json11::Json::object();
 }
 json11::Json::object osd_messenger_t::merge_configs(const json11::Json::object & cli_config,
    const json11::Json::object & file_config,
    const json11::Json::object & etcd_global_config,
    const json11::Json::object & etcd_osd_config)
 {
    return cli_config;
 }
--- a/src/msgr_rdma.cpp
+++ b/src/msgr_rdma.cpp
@@ -368,9 +368,8 @@ static void try_send_rdma_wr(osd_client_t *cl, ibv_sge *sge, int op_sge)
 bool osd_messenger_t::try_send_rdma(osd_client_t *cl)
 {
    auto rc = cl->rdma_conn;
-    if (!cl->send_list.size() || rc->cur_send > 0)
+    if (!cl->send_list.size() || rc->cur_send >= rc->max_send)
    {
        // Only send one batch at a time
        return true;
    }
    uint64_t op_size = 0, op_sge = 0;
@@ -380,6 +379,7 @@ bool osd_messenger_t::try_send_rdma(osd_client_t *cl)
        iovec & iov = cl->send_list[rc->send_pos];
        if (op_size >= rc->max_msg || op_sge >= rc->max_sge)
        {
            rc->send_sizes.push_back(op_size);
            try_send_rdma_wr(cl, sge, op_sge);
            op_sge = 0;
            op_size = 0;
@@ -405,18 +405,24 @@ bool osd_messenger_t::try_send_rdma(osd_client_t *cl)
    }
    if (op_sge > 0)
    {
        rc->send_sizes.push_back(op_size);
        try_send_rdma_wr(cl, sge, op_sge);
    }
    return true;
 }
-static void try_recv_rdma_wr(osd_client_t *cl, ibv_sge *sge, int op_sge)
+static void try_recv_rdma_wr(osd_client_t *cl, void *buf)
 {
    ibv_sge sge = {
        .addr = (uintptr_t)buf,
        .length = (uint32_t)cl->rdma_conn->max_msg,
        .lkey = cl->rdma_conn->ctx->mr->lkey,
    };
    ibv_recv_wr *bad_wr = NULL;
    ibv_recv_wr wr = {
        .wr_id = (uint64_t)(cl->peer_fd*2),
-        .sg_list = sge,
+        .sg_list = &sge,
-        .num_sge = op_sge,
+        .num_sge = 1,
    };
    int err = ibv_post_recv(cl->rdma_conn->qp, &wr, &bad_wr);
    if (err || bad_wr)
@@ -434,12 +440,7 @@ bool osd_messenger_t::try_recv_rdma(osd_client_t *cl)
    {
        void *buf = malloc_or_die(rc->max_msg);
        rc->recv_buffers.push_back(buf);
-        ibv_sge sge = {
+        try_recv_rdma_wr(cl, buf);
            .addr = (uintptr_t)buf,
            .length = (uint32_t)rc->max_msg,
            .lkey = rc->ctx->mr->lkey,
        };
        try_recv_rdma_wr(cl, &sge, 1);
    }
    return true;
 }
@@ -476,6 +477,7 @@ void osd_messenger_t::handle_rdma_events()
                continue;
            }
            osd_client_t *cl = cl_it->second;
            auto rc = cl->rdma_conn;
            if (wc[i].status != IBV_WC_SUCCESS)
            {
                fprintf(stderr, "RDMA work request failed for client %d", client_id);
@@ -489,44 +491,59 @@ void osd_messenger_t::handle_rdma_events()
            }
            if (!is_send)
            {
-                cl->rdma_conn->cur_recv--;
+                rc->cur_recv--;
-                if (!handle_read_buffer(cl, cl->rdma_conn->recv_buffers[0], wc[i].byte_len))
+                if (!handle_read_buffer(cl, rc->recv_buffers[rc->next_recv_buf], wc[i].byte_len))
                {
                    // handle_read_buffer may stop the client
                    continue;
                }
-                free(cl->rdma_conn->recv_buffers[0]);
+                try_recv_rdma_wr(cl, rc->recv_buffers[rc->next_recv_buf]);
-                cl->rdma_conn->recv_buffers.erase(cl->rdma_conn->recv_buffers.begin(), cl->rdma_conn->recv_buffers.begin()+1);
+                rc->next_recv_buf = (rc->next_recv_buf+1) % rc->recv_buffers.size();
                try_recv_rdma(cl);
            }
            else
            {
-                cl->rdma_conn->cur_send--;
+                rc->cur_send--;
-                if (!cl->rdma_conn->cur_send)
+                uint64_t sent_size = rc->send_sizes.at(0);
                rc->send_sizes.erase(rc->send_sizes.begin(), rc->send_sizes.begin()+1);
                int send_pos = 0, send_buf_pos = 0;
                while (sent_size > 0)
                {
-                    // Wait for the whole batch
+                    if (sent_size >= cl->send_list.at(send_pos).iov_len)
                    for (int i = 0; i < cl->rdma_conn->send_pos; i++)
                    {
-                        if (cl->outbox[i].flags & MSGR_SENDP_FREE)
+                        sent_size -= cl->send_list[send_pos].iov_len;
-                        {
+                        send_pos++;
                            // Reply fully sent
                            delete cl->outbox[i].op;
                        }
                    }
-                    if (cl->rdma_conn->send_pos > 0)
+                    else
                    {
-                        cl->send_list.erase(cl->send_list.begin(), cl->send_list.begin()+cl->rdma_conn->send_pos);
+                        send_buf_pos = sent_size;
-                        cl->outbox.erase(cl->outbox.begin(), cl->outbox.begin()+cl->rdma_conn->send_pos);
+                        sent_size = 0;
                        cl->rdma_conn->send_pos = 0;
                    }
                    if (cl->rdma_conn->send_buf_pos > 0)
                    {
                        cl->send_list[0].iov_base = (uint8_t*)cl->send_list[0].iov_base + cl->rdma_conn->send_buf_pos;
                        cl->send_list[0].iov_len -= cl->rdma_conn->send_buf_pos;
                        cl->rdma_conn->send_buf_pos = 0;
                    }
                    try_send_rdma(cl);
                }
                assert(rc->send_pos >= send_pos);
                if (rc->send_pos == send_pos)
                {
                    rc->send_buf_pos -= send_buf_pos;
                }
                rc->send_pos -= send_pos;
                for (int i = 0; i < send_pos; i++)
                {
                    if (cl->outbox[i].flags & MSGR_SENDP_FREE)
                    {
                        // Reply fully sent
                        delete cl->outbox[i].op;
                    }
                }
                if (send_pos > 0)
                {
                    cl->send_list.erase(cl->send_list.begin(), cl->send_list.begin()+send_pos);
                    cl->outbox.erase(cl->outbox.begin(), cl->outbox.begin()+send_pos);
                }
                if (send_buf_pos > 0)
                {
                    cl->send_list[0].iov_base = (uint8_t*)cl->send_list[0].iov_base + send_buf_pos;
                    cl->send_list[0].iov_len -= send_buf_pos;
                }
                try_send_rdma(cl);
            }
        }
    } while (event_count > 0);
--- a/src/msgr_rdma.h
+++ b/src/msgr_rdma.h
@@ -49,8 +49,9 @@ struct msgr_rdma_connection_t
    uint64_t max_msg = 0;
    int send_pos = 0, send_buf_pos = 0;
-    int recv_pos = 0, recv_buf_pos = 0;
+    int next_recv_buf = 0;
    std::vector<void*> recv_buffers;
    std::vector<uint64_t> send_sizes;
    ~msgr_rdma_connection_t();
    static msgr_rdma_connection_t *create(msgr_rdma_context_t *ctx, uint32_t max_send, uint32_t max_recv, uint32_t max_sge, uint32_t max_msg);
--- a/src/msgr_receive.cpp
+++ b/src/msgr_receive.cpp
@@ -103,7 +103,10 @@ bool osd_messenger_t::handle_read(int result, osd_client_t *cl)
            cl->recv_list.eat(result);
            if (cl->recv_list.done >= cl->recv_list.count)
            {
-                handle_finished_read(cl);
+                if (!handle_finished_read(cl))
                {
                    goto fin;
                }
            }
        }
        if (result >= cl->read_iov.iov_len)
@@ -313,17 +316,18 @@ bool osd_messenger_t::handle_reply_hdr(osd_client_t *cl)
            stop_client(cl->peer_fd);
            return false;
        }
-        if (op->reply.hdr.retval >= 0 && bmp_len > 0)
+        if (bmp_len > 0)
        {
            assert(op->bitmap);
            cl->recv_list.push_back(op->bitmap, bmp_len);
            cl->read_remaining += bmp_len;
        }
        if (op->reply.hdr.retval > 0)
        {
            assert(op->iov.count > 0);
            cl->recv_list.append(op->iov);
            cl->read_remaining += op->reply.hdr.retval;
        }
        cl->read_remaining = op->reply.hdr.retval + bmp_len;
        if (cl->read_remaining == 0)
        {
            goto reuse;
--- a/src/object_id.h
+++ b/src/object_id.h
@@ -39,6 +39,11 @@ struct __attribute__((__packed__)) obj_ver_id
    uint64_t version;
 };
 inline bool operator == (const obj_ver_id & a, const obj_ver_id & b)
 {
    return a.oid == b.oid && a.version == b.version;
 }
 inline bool operator < (const obj_ver_id & a, const obj_ver_id & b)
 {
    return a.oid < b.oid || a.oid == b.oid && a.version < b.version;
--- a/src/osd.cpp
+++ b/src/osd.cpp
@@ -35,18 +35,18 @@ osd_t::osd_t(const json11::Json & config, ring_loop_t *ringloop)
    this->ringloop = ringloop;
-    this->config = msgr.read_config(config).object_items();
+    this->cli_config = config.object_items();
-    if (this->config.find("log_level") == this->config.end())
+    this->file_config = msgr.read_config(this->cli_config);
-        this->config["log_level"] = 1;
+    parse_config(true);
    parse_config(this->config, true);
    epmgr = new epoll_manager_t(ringloop);
    // FIXME: Use timerfd_interval based directly on io_uring
    this->tfd = epmgr->tfd;
-    auto bs_cfg = json_to_bs(this->config);
+    if (!json_is_true(this->config["disable_blockstore"]))
    this->bs = new blockstore_t(bs_cfg, ringloop, tfd);
    {
        auto bs_cfg = json_to_bs(this->config);
        this->bs = new blockstore_t(bs_cfg, ringloop, tfd);
        // Autosync based on the number of unstable writes to prevent stalls due to insufficient journal space
        uint64_t max_autosync = bs->get_journal_size() / bs->get_block_size() / 2;
        if (autosync_writes > max_autosync)
@@ -67,11 +67,11 @@ osd_t::osd_t(const json11::Json & config, ring_loop_t *ringloop)
        }
    }
-    this->tfd->set_timer(print_stats_interval*1000, true, [this](int timer_id)
+    print_stats_timer_id = this->tfd->set_timer(print_stats_interval*1000, true, [this](int timer_id)
    {
        print_stats();
    });
-    this->tfd->set_timer(slow_log_interval*1000, true, [this](int timer_id)
+    slow_log_timer_id = this->tfd->set_timer(slow_log_interval*1000, true, [this](int timer_id)
    {
        print_slow();
    });
@@ -91,18 +91,42 @@ osd_t::osd_t(const json11::Json & config, ring_loop_t *ringloop)
 osd_t::~osd_t()
 {
    if (slow_log_timer_id >= 0)
    {
        tfd->clear_timer(slow_log_timer_id);
        slow_log_timer_id = -1;
    }
    if (print_stats_timer_id >= 0)
    {
        tfd->clear_timer(print_stats_timer_id);
        print_stats_timer_id = -1;
    }
    if (autosync_timer_id >= 0)
    {
        tfd->clear_timer(autosync_timer_id);
        autosync_timer_id = -1;
    }
    ringloop->unregister_consumer(&consumer);
    delete epmgr;
-    delete bs;
+    if (bs)
        delete bs;
    close(listen_fd);
    free(zero_buffer);
 }
-void osd_t::parse_config(const json11::Json & config, bool allow_disk_params)
+void osd_t::parse_config(bool init)
 {
    config = msgr.merge_configs(cli_config, file_config, etcd_global_config, etcd_osd_config);
    if (config.find("log_level") == this->config.end())
        config["log_level"] = 1;
    if (bs)
    {
        auto bs_cfg = json_to_bs(config);
        bs->parse_config(bs_cfg);
    }
    st_cli.parse_config(config);
    msgr.parse_config(config);
-    if (allow_disk_params)
+    if (init)
    {
        // OSD number
        osd_num = config["osd_num"].uint64_value();
@@ -124,24 +148,27 @@ void osd_t::parse_config(const json11::Json & config, bool allow_disk_params)
            immediate_commit = IMMEDIATE_SMALL;
        else
            immediate_commit = IMMEDIATE_NONE;
        // Bind address
        bind_address = config["bind_address"].string_value();
        if (bind_address == "")
            bind_address = "0.0.0.0";
        bind_port = config["bind_port"].uint64_value();
        if (bind_port <= 0 || bind_port > 65535)
            bind_port = 0;
        // OSD configuration
        etcd_report_interval = config["etcd_report_interval"].uint64_value();
        if (etcd_report_interval <= 0)
            etcd_report_interval = 5;
        readonly = json_is_true(config["readonly"]);
        run_primary = !json_is_false(config["run_primary"]);
        allow_test_ops = json_is_true(config["allow_test_ops"]);
    }
    // Bind address
    bind_address = config["bind_address"].string_value();
    if (bind_address == "")
        bind_address = "0.0.0.0";
    bind_port = config["bind_port"].uint64_value();
    if (bind_port <= 0 || bind_port > 65535)
        bind_port = 0;
    // OSD configuration
    log_level = config["log_level"].uint64_value();
-    etcd_report_interval = config["etcd_report_interval"].uint64_value();
+    auto old_no_rebalance = no_rebalance;
    if (etcd_report_interval <= 0)
        etcd_report_interval = 5;
    readonly = json_is_true(config["readonly"]);
    run_primary = !json_is_false(config["run_primary"]);
    no_rebalance = json_is_true(config["no_rebalance"]);
    auto old_no_recovery = no_recovery;
    no_recovery = json_is_true(config["no_recovery"]);
-    allow_test_ops = json_is_true(config["allow_test_ops"]);
+    auto old_autosync_interval = autosync_interval;
    if (!config["autosync_interval"].is_null())
    {
        // Allow to set it to 0
@@ -169,15 +196,46 @@ void osd_t::parse_config(const json11::Json & config, bool allow_disk_params)
    recovery_sync_batch = config["recovery_sync_batch"].uint64_value();
    if (recovery_sync_batch < 1 || recovery_sync_batch > MAX_RECOVERY_QUEUE)
        recovery_sync_batch = DEFAULT_RECOVERY_BATCH;
    auto old_print_stats_interval = print_stats_interval;
    print_stats_interval = config["print_stats_interval"].uint64_value();
    if (!print_stats_interval)
        print_stats_interval = 3;
    auto old_slow_log_interval = slow_log_interval;
    slow_log_interval = config["slow_log_interval"].uint64_value();
    if (!slow_log_interval)
        slow_log_interval = 10;
    inode_vanish_time = config["inode_vanish_time"].uint64_value();
    if (!inode_vanish_time)
        inode_vanish_time = 60;
    if ((old_no_rebalance && !no_rebalance || old_no_recovery && !no_recovery) &&
        !(peering_state & (OSD_RECOVERING | OSD_FLUSHING_PGS)))
    {
        peering_state = peering_state | OSD_RECOVERING;
    }
    if (old_autosync_interval != autosync_interval && autosync_timer_id >= 0)
    {
        this->tfd->clear_timer(autosync_timer_id);
        autosync_timer_id = this->tfd->set_timer(autosync_interval*1000, true, [this](int timer_id)
        {
            autosync();
        });
    }
    if (old_print_stats_interval != print_stats_interval && print_stats_timer_id >= 0)
    {
        tfd->clear_timer(print_stats_timer_id);
        print_stats_timer_id = this->tfd->set_timer(print_stats_interval*1000, true, [this](int timer_id)
        {
            print_stats();
        });
    }
    if (old_slow_log_interval != slow_log_interval && slow_log_timer_id >= 0)
    {
        tfd->clear_timer(slow_log_timer_id);
        slow_log_timer_id = this->tfd->set_timer(slow_log_interval*1000, true, [this](int timer_id)
        {
            print_slow();
        });
    }
 }
 void osd_t::bind_socket()
@@ -402,7 +460,7 @@ void osd_t::print_slow()
                int l = sizeof(alloc), n;
                char *buf = alloc;
 #define bufprintf(s, ...) { n = snprintf(buf, l, s, __VA_ARGS__); n = n < 0 ? 0 : n; buf += n; l -= n; }
-                bufprintf("[OSD %lu] Slow op", osd_num);
+                bufprintf("[OSD %lu] Slow op %lx", osd_num, (unsigned long)op);
                if (kv.second->osd_num)
                {
                    bufprintf(" from peer OSD %lu (client %d)", kv.second->osd_num, kv.second->peer_fd);
@@ -475,7 +533,7 @@ void osd_t::print_slow()
            }
        }
    }
-    if (has_slow)
+    if (has_slow && bs)
    {
        bs->dump_diagnostics();
    }
--- a/src/osd.h
+++ b/src/osd.h
@@ -90,7 +90,7 @@ class osd_t
 {
    // config
-    json11::Json::object config;
+    json11::Json::object cli_config, file_config, etcd_global_config, etcd_osd_config, config;
    int etcd_report_interval = 5;
    bool readonly = false;
@@ -126,6 +126,7 @@ class osd_t
    bool pg_config_applied = false;
    bool etcd_reporting_pg_state = false;
    bool etcd_reporting_stats = false;
    int autosync_timer_id = -1, print_stats_timer_id = -1, slow_log_timer_id = -1;
    // peers and PGs
@@ -152,7 +153,7 @@ class osd_t
    bool stopping = false;
    int inflight_ops = 0;
-    blockstore_t *bs;
+    blockstore_t *bs = NULL;
    void *zero_buffer = NULL;
    uint64_t zero_buffer_size = 0;
    uint32_t bs_block_size, bs_bitmap_granularity, clean_entry_bitmap_size;
@@ -173,7 +174,7 @@ class osd_t
    uint64_t recovery_stat_bytes[2][2] = {};
    // cluster connection
-    void parse_config(const json11::Json & config, bool allow_disk_params);
+    void parse_config(bool init);
    void init_cluster();
    void on_change_osd_state_hook(osd_num_t peer_osd);
    void on_change_pg_history_hook(pool_id_t pool_id, pg_num_t pg_num);
--- a/src/osd_cluster.cpp
+++ b/src/osd_cluster.cpp
@@ -75,7 +75,7 @@ void osd_t::init_cluster()
    }
    if (run_primary && autosync_interval > 0)
    {
-        this->tfd->set_timer(autosync_interval*1000, true, [this](int timer_id)
+        autosync_timer_id = this->tfd->set_timer(autosync_interval*1000, true, [this](int timer_id)
        {
            autosync();
        });
@@ -182,10 +182,10 @@ json11::Json osd_t::get_statistics()
    char time_str[50] = { 0 };
    sprintf(time_str, "%ld.%03ld", ts.tv_sec, ts.tv_nsec/1000000);
    st["time"] = time_str;
    st["blockstore_ready"] = bs->is_started();
    st["data_block_size"] = (uint64_t)bs->get_block_size();
    if (bs)
    {
        st["blockstore_ready"] = bs->is_started();
        st["data_block_size"] = (uint64_t)bs->get_block_size();
        st["size"] = bs->get_block_count() * bs->get_block_size();
        st["free"] = bs->get_free_block_count() * bs->get_block_size();
    }
@@ -233,7 +233,8 @@ void osd_t::report_statistics()
    json11::Json::object inode_space;
    json11::Json::object last_stat;
    pool_id_t last_pool = 0;
-    auto & bs_inode_space = bs->get_inode_space_stats();
+    std::map<uint64_t, uint64_t> bs_empty_space;
    auto & bs_inode_space = bs ? bs->get_inode_space_stats() : bs_empty_space;
    for (auto kv: bs_inode_space)
    {
        pool_id_t pool_id = INODE_POOL(kv.first);
@@ -374,7 +375,11 @@ void osd_t::on_change_osd_state_hook(osd_num_t peer_osd)
 void osd_t::on_change_etcd_state_hook(std::map<std::string, etcd_kv_t> & changes)
 {
-    // FIXME apply config changes in runtime (maybe, some)
+    if (changes.find(st_cli.etcd_prefix+"/config/global") != changes.end())
    {
        etcd_global_config = changes[st_cli.etcd_prefix+"/config/global"].value.object_items();
        parse_config(false);
    }
    if (run_primary)
    {
        apply_pg_count();
@@ -384,11 +389,8 @@ void osd_t::on_change_etcd_state_hook(std::map<std::string, etcd_kv_t> & changes
 void osd_t::on_load_config_hook(json11::Json::object & global_config)
 {
-    json11::Json::object osd_config = this->config;
+    etcd_global_config = global_config;
-    for (auto & kv: global_config)
+    parse_config(true);
        if (osd_config.find(kv.first) == osd_config.end())
            osd_config[kv.first] = kv.second;
    parse_config(osd_config, false);
    bind_socket();
    acquire_lease();
 }
@@ -683,7 +685,7 @@ void osd_t::apply_pg_config()
                auto vec_all_peers = std::vector<osd_num_t>(all_peers.begin(), all_peers.end());
                if (currently_taken)
                {
-                    if (pg_it->second.state & (PG_ACTIVE | PG_INCOMPLETE | PG_PEERING | PG_REPEERING | PG_PEERED))
+                    if (pg_it->second.state & (PG_ACTIVE | PG_INCOMPLETE | PG_PEERING | PG_REPEERING))
                    {
                        if (pg_it->second.target_set == pg_cfg.target_set &&
                            pg_it->second.target_history == pg_cfg.target_history &&
@@ -734,7 +736,7 @@ void osd_t::apply_pg_config()
                    .pg_cursize = 0,
                    .pg_size = pool_item.second.pg_size,
                    .pg_minsize = pool_item.second.pg_minsize,
-                    .pg_data_size = pg.scheme == POOL_SCHEME_REPLICATED
+                    .pg_data_size = pool_item.second.scheme == POOL_SCHEME_REPLICATED
                         ? 1 : pool_item.second.pg_size - pool_item.second.parity_chunks,
                    .pool_id = pool_id,
                    .pg_num = pg_num,
@@ -810,11 +812,21 @@ void osd_t::report_pg_states()
                    pg_it->second.cur_state != 0)
                {
                    pg_state_exists = true;
                    if (pg.state == PG_OFFLINE && pg_it->second.cur_primary != this->osd_num)
                    {
                        // Nothing to check or report, PG is already taken over by another OSD
                        continue;
                    }
                }
            }
        }
        if (!pg_state_exists)
        {
            if (pg.state == PG_OFFLINE)
            {
                // Nothing to check or report, PG is already stopped
                continue;
            }
            // Check that the PG key does not exist
            // Failed check indicates an unsuccessful PG lock attempt in this case
            checks.push_back(json11::Json::object {
@@ -899,6 +911,15 @@ void osd_t::report_pg_states()
    {
        etcd_reporting_pg_state = false;
        if (!data["succeeded"].bool_value())
        {
            std::string rpgnames = "";
            for (auto pp: reporting_pgs)
            {
                rpgnames += (rpgnames.size() ? ", " : "")+std::to_string(pp.pool_pg_num.pool_id)+"/"+std::to_string(pp.pool_pg_num.pg_num);
            }
            printf("Error reporting PG %s states, will repeat the attempt: %s\n", rpgnames.c_str(), err.c_str());
        }
        if (!data["succeeded"].bool_value())
        {
            // One of PG state updates failed, put dirty flags back
            for (auto pp: reporting_pgs)
@@ -963,13 +984,6 @@ void osd_t::report_pg_states()
                        }
                        this->pgs.erase(pg_it);
                    }
                    else if (pg_it->second.state & PG_PEERED)
                    {
                        // Activate PG after PG PEERED state is reported along with history
                        // (if the state wasn't changed again)
                        pg_it->second.state = pg_it->second.state & ~PG_PEERED | PG_ACTIVE;
                        report_pg_state(pg_it->second);
                    }
                }
            }
            // Push other PG state updates, if any
--- a/src/osd_flush.cpp
+++ b/src/osd_flush.cpp
@@ -64,6 +64,11 @@ void osd_t::submit_pg_flush_ops(pg_t & pg)
 void osd_t::handle_flush_op(bool rollback, pool_id_t pool_id, pg_num_t pg_num, pg_flush_batch_t *fb, osd_num_t peer_osd, int retval)
 {
    if (log_level > 2)
    {
        printf("[PG %u/%u] flush batch %lx completed on OSD %lu with result %d\n",
            pool_id, pg_num, (uint64_t)fb, peer_osd, retval);
    }
    pool_pg_num_t pg_id = { .pool_id = pool_id, .pg_num = pg_num };
    if (pgs.find(pg_id) == pgs.end() || pgs[pg_id].flush_batch != fb)
    {
@@ -99,10 +104,9 @@ void osd_t::handle_flush_op(bool rollback, pool_id_t pool_id, pg_num_t pg_num, p
        std::vector<osd_op_t*> continue_ops;
        auto & pg = pgs.at(pg_id);
        auto it = pg.flush_actions.begin(), prev_it = it;
        auto erase_start = it;
        while (1)
        {
-            if (it == pg.flush_actions.end() ||
+            if (it == pg.flush_actions.end() || !it->second.submitted ||
                it->first.oid.inode != prev_it->first.oid.inode ||
                (it->first.oid.stripe & ~STRIPE_MASK) != (prev_it->first.oid.stripe & ~STRIPE_MASK))
            {
@@ -116,29 +120,23 @@ void osd_t::handle_flush_op(bool rollback, pool_id_t pool_id, pg_num_t pg_num, p
                });
                if (wr_it != pg.write_queue.end())
                {
                    if (log_level > 2)
                    {
                        printf("[PG %u/%u] continuing write %lx to object %lx:%lx after flush\n",
                            pool_id, pg_num, (uint64_t)wr_it->second, wr_it->first.inode, wr_it->first.stripe);
                    }
                    continue_ops.push_back(wr_it->second);
                    pg.write_queue.erase(wr_it);
                }
            }
-            if ((it == pg.flush_actions.end() || !it->second.submitted) &&
+            if (it == pg.flush_actions.end() || !it->second.submitted)
                erase_start != it)
            {
                pg.flush_actions.erase(erase_start, it);
            }
            if (it == pg.flush_actions.end())
            {
                if (it != pg.flush_actions.begin())
                {
                    pg.flush_actions.erase(pg.flush_actions.begin(), it);
                }
                break;
            }
-            prev_it = it;
+            prev_it = it++;
            if (!it->second.submitted)
            {
                it++;
                erase_start = it;
            }
            else
            {
                it++;
            }
        }
        delete fb;
        pg.flush_batch = NULL;
@@ -168,6 +166,18 @@ bool osd_t::submit_flush_op(pool_id_t pool_id, pg_num_t pg_num, pg_flush_batch_t
    // Copy buffer so it gets freed along with the operation
    op->buf = malloc_or_die(sizeof(obj_ver_id) * count);
    memcpy(op->buf, data, sizeof(obj_ver_id) * count);
    if (log_level > 2)
    {
        printf(
            "[PG %u/%u] flush batch %lx on OSD %lu: %s objects: ",
            pool_id, pg_num, (uint64_t)fb, peer_osd, rollback ? "rollback" : "stabilize"
        );
        for (int i = 0; i < count; i++)
        {
            printf(i > 0 ? ", %lx:%lx v%lu" : "%lx:%lx v%lu", data[i].oid.inode, data[i].oid.stripe, data[i].version);
        }
        printf("\n");
    }
    if (peer_osd == this->osd_num)
    {
        // local
@@ -304,9 +314,10 @@ void osd_t::submit_recovery_op(osd_recovery_op_t *op)
            {
                // PG is stopped or one of the OSDs is gone, error is harmless
                printf(
-                    "Recovery operation failed with object %lx:%lx (PG %u/%u)\n",
+                    "[PG %u/%u] Recovery operation failed with object %lx:%lx\n",
-                    op->oid.inode, op->oid.stripe, INODE_POOL(op->oid.inode),
+                    INODE_POOL(op->oid.inode),
-                    map_to_pg(op->oid, st_cli.pool_config.at(INODE_POOL(op->oid.inode)).pg_stripe_size)
+                    map_to_pg(op->oid, st_cli.pool_config.at(INODE_POOL(op->oid.inode)).pg_stripe_size),
                    op->oid.inode, op->oid.stripe
                );
            }
            else
--- a/src/osd_ops.h
+++ b/src/osd_ops.h
@@ -191,7 +191,7 @@ struct __attribute__((__packed__)) osd_op_rw_t
    uint64_t inode;
    // offset
    uint64_t offset;
-    // length
+    // length. 0 means to read all bitmaps of the specified range, but no data.
    uint32_t len;
    // flags (for future)
    uint32_t flags;
--- a/src/osd_peering.cpp
+++ b/src/osd_peering.cpp
@@ -50,10 +50,6 @@ void osd_t::handle_peers()
                    still = true;
                }
            }
            else if (p.second.state & PG_PEERED)
            {
                still = true;
            }
        }
        if (!still)
        {
@@ -74,17 +70,13 @@ void osd_t::handle_peers()
                }
                still = true;
            }
            else if (p.second.state & PG_PEERED)
            {
                still = true;
            }
        }
        if (!still)
        {
            peering_state = peering_state & ~OSD_FLUSHING_PGS | OSD_RECOVERING;
        }
    }
-    if ((peering_state & OSD_RECOVERING) && !readonly)
+    if (!(peering_state & OSD_FLUSHING_PGS) && (peering_state & OSD_RECOVERING) && !readonly)
    {
        if (!continue_recovery())
        {
@@ -100,7 +92,7 @@ void osd_t::repeer_pgs(osd_num_t peer_osd)
    {
        auto & pg = p.second;
        bool repeer = false;
-        if (pg.state & (PG_PEERING | PG_PEERED | PG_ACTIVE | PG_INCOMPLETE))
+        if (pg.state & (PG_PEERING | PG_ACTIVE | PG_INCOMPLETE))
        {
            for (osd_num_t pg_osd: pg.all_peers)
            {
@@ -491,6 +483,10 @@ void osd_t::report_pg_state(pg_t & pg)
        pg.all_peers = pg.target_set;
        std::sort(pg.all_peers.begin(), pg.all_peers.end());
        pg.cur_peers = pg.target_set;
        // Change pg_config at the same time, otherwise our PG reconciling loop may try to apply the old metadata
        auto & pg_cfg = st_cli.pool_config[pg.pool_id].pg_config[pg.pg_num];
        pg_cfg.target_history = pg.target_history;
        pg_cfg.all_peers = pg.all_peers;
    }
    else if (pg.state == (PG_ACTIVE|PG_LEFT_ON_DEAD))
    {
@@ -530,6 +526,9 @@ void osd_t::report_pg_state(pg_t & pg)
                pg.cur_peers.push_back(pg_osd);
            }
        }
        auto & pg_cfg = st_cli.pool_config[pg.pool_id].pg_config[pg.pg_num];
        pg_cfg.target_history = pg.target_history;
        pg_cfg.all_peers = pg.all_peers;
    }
    if (pg.state == PG_OFFLINE && !this->pg_config_applied)
    {
--- a/src/osd_peering_pg.cpp
+++ b/src/osd_peering_pg.cpp
@@ -88,14 +88,10 @@ void pg_obj_state_check_t::walk()
    {
        // Activate as degraded
        // Current OSD set will be added into target_history on first write
-        pg->state |= PG_DEGRADED | PG_PEERED;
+        pg->state |= PG_DEGRADED;
    }
-    else
+    pg->state |= PG_ACTIVE;
-    {
+    if (pg->cur_peers.size() < pg->all_peers.size())
        // Just activate
        pg->state |= PG_ACTIVE;
    }
    if (pg->state == PG_ACTIVE && pg->cur_peers.size() < pg->all_peers.size())
    {
        pg->state |= PG_LEFT_ON_DEAD;
    }
@@ -460,11 +456,10 @@ void pg_t::calc_object_states(int log_level)
 void pg_t::print_state()
 {
    printf(
-        "[PG %u/%u] is %s%s%s%s%s%s%s%s%s%s%s%s%s%s%s (%lu objects)\n", pool_id, pg_num,
+        "[PG %u/%u] is %s%s%s%s%s%s%s%s%s%s%s%s%s%s (%lu objects)\n", pool_id, pg_num,
        (state & PG_STARTING) ? "starting" : "",
        (state & PG_OFFLINE) ? "offline" : "",
        (state & PG_PEERING) ? "peering" : "",
        (state & PG_PEERED) ? "peered" : "",
        (state & PG_INCOMPLETE) ? "incomplete" : "",
        (state & PG_ACTIVE) ? "active" : "",
        (state & PG_REPEERING) ? "repeering" : "",
--- a/src/osd_peering_pg_test.cpp
+++ b/src/osd_peering_pg_test.cpp
@@ -54,5 +54,6 @@ int main(int argc, char *argv[])
    {
        printf("dev: state=%lx\n", it.second.state);
    }
    delete pg.peering_state;
    return 0;
 }
--- a/src/osd_primary.cpp
+++ b/src/osd_primary.cpp
@@ -186,10 +186,22 @@ void osd_t::continue_primary_read(osd_op_t *cur_op)
    cur_op->reply.rw.bitmap_len = 0;
    {
        auto & pg = pgs.at({ .pool_id = INODE_POOL(op_data->oid.inode), .pg_num = op_data->pg_num });
-        for (int role = 0; role < op_data->pg_data_size; role++)
+        if (cur_op->req.rw.len == 0)
        {
-            op_data->stripes[role].read_start = op_data->stripes[role].req_start;
+            // len=0 => bitmap read
-            op_data->stripes[role].read_end = op_data->stripes[role].req_end;
+            for (int role = 0; role < op_data->pg_data_size; role++)
            {
                op_data->stripes[role].read_start = 0;
                op_data->stripes[role].read_end = UINT32_MAX;
            }
        }
        else
        {
            for (int role = 0; role < op_data->pg_data_size; role++)
            {
                op_data->stripes[role].read_start = op_data->stripes[role].req_start;
                op_data->stripes[role].read_end = op_data->stripes[role].req_end;
            }
        }
        // Determine version
        auto vo_it = pg.ver_override.find(op_data->oid);
--- a/src/osd_primary_chain.cpp
+++ b/src/osd_primary_chain.cpp
@@ -219,7 +219,7 @@ int osd_t::submit_bitmap_subops(osd_op_t *cur_op, pg_t & pg)
            op_data->n_subops++;
        }
    }
-    if (op_data->n_subops)
+    if (op_data->n_subops > 0)
    {
        op_data->fact_ver = 0;
        op_data->done = op_data->errors = 0;
@@ -297,7 +297,7 @@ int osd_t::submit_bitmap_subops(osd_op_t *cur_op, pg_t & pg)
                    // Fail it immediately
                    subop->peer_fd = -1;
                    subop->reply.hdr.retval = -EPIPE;
-                    subop->callback(subop);
+                    ringloop->set_immediate([subop]() { std::function<void(osd_op_t*)>(subop->callback)(subop); });
                }
                subop_idx++;
            }
--- a/src/osd_primary_subops.cpp
+++ b/src/osd_primary_subops.cpp
@@ -53,7 +53,10 @@ void osd_t::finish_op(osd_op_t *cur_op, int retval)
        inode_stats[cur_op->req.rw.inode].op_count[inode_st_op]++;
        inode_stats[cur_op->req.rw.inode].op_sum[inode_st_op] += usec;
        if (cur_op->req.hdr.opcode == OSD_OP_DELETE)
-            inode_stats[cur_op->req.rw.inode].op_bytes[inode_st_op] += cur_op->op_data->pg_data_size * bs_block_size;
+        {
            if (cur_op->op_data)
                inode_stats[cur_op->req.rw.inode].op_bytes[inode_st_op] += cur_op->op_data->pg_data_size * bs_block_size;
        }
        else
            inode_stats[cur_op->req.rw.inode].op_bytes[inode_st_op] += cur_op->req.rw.len;
    }
@@ -148,6 +151,13 @@ int osd_t::submit_primary_subop_batch(int submit_type, inode_t inode, uint64_t o
        {
            int stripe_num = rep ? 0 : role;
            osd_op_t *subop = op_data->subops + i;
            uint32_t subop_len = wr
                ? stripes[stripe_num].write_end - stripes[stripe_num].write_start
                : stripes[stripe_num].read_end - stripes[stripe_num].read_start;
            if (!wr && stripes[stripe_num].read_end == UINT32_MAX)
            {
                subop_len = 0;
            }
            if (role_osd_num == this->osd_num)
            {
                clock_gettime(CLOCK_REALTIME, &subop->tv_begin);
@@ -166,7 +176,7 @@ int osd_t::submit_primary_subop_batch(int submit_type, inode_t inode, uint64_t o
                    },
                    .version = op_version,
                    .offset = wr ? stripes[stripe_num].write_start : stripes[stripe_num].read_start,
-                    .len = wr ? stripes[stripe_num].write_end - stripes[stripe_num].write_start : stripes[stripe_num].read_end - stripes[stripe_num].read_start,
+                    .len = subop_len,
                    .buf = wr ? stripes[stripe_num].write_buf : stripes[stripe_num].read_buf,
                    .bitmap = stripes[stripe_num].bmp_buf,
                });
@@ -196,7 +206,7 @@ int osd_t::submit_primary_subop_batch(int submit_type, inode_t inode, uint64_t o
                    },
                    .version = op_version,
                    .offset = wr ? stripes[stripe_num].write_start : stripes[stripe_num].read_start,
-                    .len = wr ? stripes[stripe_num].write_end - stripes[stripe_num].write_start : stripes[stripe_num].read_end - stripes[stripe_num].read_start,
+                    .len = subop_len,
                    .attr_len = wr ? clean_entry_bitmap_size : 0,
                };
 #ifdef OSD_DEBUG
@@ -215,9 +225,9 @@ int osd_t::submit_primary_subop_batch(int submit_type, inode_t inode, uint64_t o
                }
                else
                {
-                    if (stripes[stripe_num].read_end > stripes[stripe_num].read_start)
+                    if (subop_len > 0)
                    {
-                        subop->iov.push_back(stripes[stripe_num].read_buf, stripes[stripe_num].read_end - stripes[stripe_num].read_start);
+                        subop->iov.push_back(stripes[stripe_num].read_buf, subop_len);
                    }
                }
                subop->callback = [cur_op, this](osd_op_t *subop)
@@ -235,7 +245,7 @@ int osd_t::submit_primary_subop_batch(int submit_type, inode_t inode, uint64_t o
                    // Fail it immediately
                    subop->peer_fd = -1;
                    subop->reply.hdr.retval = -EPIPE;
-                    subop->callback(subop);
+                    ringloop->set_immediate([subop]() { std::function<void(osd_op_t*)>(subop->callback)(subop); });
                }
            }
            i++;
@@ -469,7 +479,7 @@ void osd_t::submit_primary_del_batch(osd_op_t *cur_op, obj_ver_osd_t *chunks_to_
    osd_primary_op_data_t *op_data = cur_op->op_data;
    op_data->n_subops = chunks_to_delete_count;
    op_data->done = op_data->errors = op_data->errcode = 0;
-    if (!op_data->n_subops)
+    if (op_data->n_subops <= 0)
    {
        return;
    }
@@ -520,7 +530,7 @@ void osd_t::submit_primary_del_batch(osd_op_t *cur_op, obj_ver_osd_t *chunks_to_
                // Fail it immediately
                subops[i].peer_fd = -1;
                subops[i].reply.hdr.retval = -EPIPE;
-                subops[i].callback(&subops[i]);
+                ringloop->set_immediate([subop = &subops[i]]() { std::function<void(osd_op_t*)>(subop->callback)(subop); });
            }
        }
    }
@@ -635,7 +645,7 @@ void osd_t::submit_primary_stab_subops(osd_op_t *cur_op)
                // Fail it immediately
                subops[i].peer_fd = -1;
                subops[i].reply.hdr.retval = -EPIPE;
-                subops[i].callback(&subops[i]);
+                ringloop->set_immediate([subop = &subops[i]]() { std::function<void(osd_op_t*)>(subop->callback)(subop); });
            }
        }
    }
--- a/src/osd_primary_sync.cpp
+++ b/src/osd_primary_sync.cpp
@@ -166,7 +166,7 @@ resume_6:
                for (int i = 0; i < unstable_osd.len; i++)
                {
                    // Except those from peered PGs
-                    auto & w = op_data->unstable_writes[i];
+                    auto & w = op_data->unstable_writes[unstable_osd.start + i];
                    pool_pg_num_t wpg = {
                        .pool_id = INODE_POOL(w.oid.inode),
                        .pg_num = map_to_pg(w.oid, st_cli.pool_config.at(INODE_POOL(w.oid.inode)).pg_stripe_size),
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
Vitaliy Filippov	72f0cff79d	WIP Use random_hier_combinations	2023-05-18 17:44:00 +03:00
Vitaliy Filippov	c1d470522c	Replace flatten_tree with extract_tree_levels	2023-05-18 17:44:00 +03:00
Vitaliy Filippov	57feb7f390	Implement multi-level tree extractor for hierarchical failure domains	2023-05-18 17:44:00 +03:00
Vitaliy Filippov	431f780347	Implement a PG generator for hierarchical failure domains	2023-05-18 17:44:00 +03:00
Vitaliy Filippov	98077a1712	Remove unused dependencies from CSI	2023-05-18 11:54:47 +03:00
Vitaliy Filippov	1c7d53996d	Reweight only 2 OSDs to zero in test_rebalance_verify, otherwise the test does not pass with EC 3+2	2023-05-18 00:42:40 +03:00
Vitaliy Filippov	2ca07b1ea7	Raise timeout in test_rebalance_verify	2023-05-17 01:58:01 +03:00
Vitaliy Filippov	022176aa98	Fix NaN during PG optimisation if there are nonexisting OSDs in node_placement	2023-05-17 01:20:30 +03:00
Vitaliy Filippov	120e3fa7bc	Fix pool deletion	2023-05-17 00:45:59 +03:00
Vitaliy Filippov	629999f789	Clear journal_device and meta_device before initialising the next OSD in automatic mode	2023-05-15 23:58:55 +03:00
Vitaliy Filippov	93eca11ba2	Fix rhel 9 installation docs	2023-05-15 13:09:18 +03:00
Vitaliy Filippov	5a9e1ede52	Release 0.8.9 - The tests are now stable and run in a CI system based on Gitea CI - The release includes final bug fixes for EC: - Implement missing EC recovery of allocation bitmap when built with ISA-L - Fix broken snapshot export with EC (allocation bitmap reads were giving incorrect results previously) - Also fixed bugs manifesting under heavy load: - Fix monitor possibly applying incorrect PG history on retries - Fix monitor incorrectly changing PG count when last_clean_pgs contains less PGs than the new number - Allow writes to wait for free space again, but now correctly (previously dropped in 0.8.2) - Fix a rare segfault in client (handle client stop during incoming stream handling in 1 more place) - Make monitor correctly handle etcd connection errors - it could die instead of connecting to another etcd - Fix OSD rarely being unable to report PG states after a PG was taken over by another OSD - Fixed return code for incomplete EC objects (now EIO) and made cluster client retry this error - Made other small changes for tests: timeouts, nice/ionice for etcd, waiting conditions, NBD device checks and so on	2023-05-14 01:25:09 +03:00
Vitaliy Filippov	1c9a188600	Add tests to CI	2023-05-14 00:06:09 +03:00
Vitaliy Filippov	de3e609166	Add a FIXME about QEMU driver thread safety	2023-05-14 00:06:09 +03:00
Vitaliy Filippov	11481170f5	Add a FIXME about ENOSPC	2023-05-13 23:59:44 +03:00
Vitaliy Filippov	e69d459d43	Allow rebalance to start in test_interrupted_rebalance, raise etcd start timeout	2023-05-13 15:16:28 +03:00
Vitaliy Filippov	da82754baa	Wait for conditions in test_move_reappear instead of waiting a fixed amount of time	2023-05-12 23:18:07 +03:00
Vitaliy Filippov	d356aca030	Add missing $NO_SAME OSD argument to test_splitbrain	2023-05-12 23:18:07 +03:00
Vitaliy Filippov	04a273d213	Raise NBD timeout in tests	2023-05-12 23:18:07 +03:00
Vitaliy Filippov	6442010f93	Skip offline PGs during state reporting when the state is already deleted or taken over by another OSD This fixes OSDs being unable to report PG states in rare conditions	2023-05-12 23:17:45 +03:00
Vitaliy Filippov	6f4dc16c59	Handle etcd connection errors correctly in mon (unhandled error events)	2023-05-11 11:02:44 +03:00
Vitaliy Filippov	ce4a8067b5	Handle client stop during incoming stream handling in 1 more place	2023-05-11 01:53:41 +03:00
Vitaliy Filippov	e431ecb715	Make tests more stable in CI	2023-05-11 01:53:41 +03:00
Vitaliy Filippov	8cac795445	Return EIO instead of EINVAL for incomplete EC objects	2023-05-11 01:15:23 +03:00
Vitaliy Filippov	a409598b16	Wait for free space again, but count on big_write flushes instead of just flusher activity	2023-05-10 01:51:02 +03:00
Vitaliy Filippov	f4c6765522	Ignore ENOENT in epoll_ctl	2023-05-08 20:39:20 +03:00
Vitaliy Filippov	ad2916068a	Fix test_add_osd rebalance timeout check	2023-05-08 20:39:20 +03:00
Vitaliy Filippov	321cb435a6	Fix monitor incorrectly changing PG count when last_clean_pgs contains less PGs than the new number	2023-05-08 20:39:20 +03:00
Vitaliy Filippov	cfcf4f4355	Support checking /dev/nbdX nodes in Docker	2023-05-08 20:39:20 +03:00
Vitaliy Filippov	e0fb17bfee	Make etcd more stable in tests (add ionice and raise timeout)	2023-05-08 20:36:00 +03:00
Vitaliy Filippov	5b9031fecc	Fix monitor possibly applying incorrect PG history under heavy load Monitor could deceive itself by immediately saving PG configuration changes which weren't applied to etcd yet in memory, and apply incorrect PG history changes next time if the first update fails. This usually only happened under heavy load and was caught in CI. :-)	2023-05-07 23:23:00 +03:00
Vitaliy Filippov	5da1d8e1b5	Fix EC just-bitmap reads (len=0) (fixes SCHEME=ec test_snapshot.sh)	2023-05-07 14:00:08 +03:00
Vitaliy Filippov	44f86f1999	Add a basic EC 2+2 recovery test (not really required, but let it be there)	2023-05-07 11:26:27 +03:00
Vitaliy Filippov	2d9a80c6f6	Implement missing bitmap recovery with ISA-L \(°□°)/	2023-05-07 11:25:51 +03:00
Vitaliy Filippov	5e295e346e	Do not make vitastor-mon part of vitastor.target	2023-04-29 00:17:47 +03:00
Vitaliy Filippov	d9c0898b7c	Notes about config and vitastor-disk cache status	2023-04-29 00:08:24 +03:00
Vitaliy Filippov	04cfb48361	Add a note about PVE 7.4	2023-04-28 11:37:11 +03:00
Vitaliy Filippov	ab615849d6	Release 0.8.8 - Fix vitastor-cli rm/rm-data broken in 0.8.6 (missing messenger initialization) - Prepare OSD read handler for upcoming version with scrub - allow "secondary reads" to return errors - Fix OSDs re-peering PGs infinitely with a big number of PGs (reproduced in test_add_osd) - Fix another variant of flusher sync-waiting stall (reproduced in test_write) - Fix other tests in tests/ (will add them to Gitea CI soon) - Add patches for QEMU 6.2-8.0 - Fix QEMU driver compatibility with QEMU 8.0 - Build packages for RHEL 9 clones (based on AlmaLinux 9)	2023-04-28 11:22:00 +03:00
Vitaliy Filippov	38be9a49c0	Add AlmaLinux 9 build to documentation	2023-04-28 02:00:52 +03:00
Vitaliy Filippov	7d6bf84a3e	Add scripts/meson-buildoptions.sh to QEMU patches	2023-04-28 01:43:22 +03:00
Vitaliy Filippov	41a40a4123	Add QEMU spec patch for Alma/Rocky/RH 9	2023-04-28 01:32:06 +03:00
Vitaliy Filippov	b94587ef0e	Fix some build warnings	2023-04-28 00:44:27 +03:00
Vitaliy Filippov	2a2f4f6738	Add Almalinux 9 build	2023-04-28 00:40:50 +03:00
Vitaliy Filippov	c768a9015f	Fix QEMU driver compatibility with QEMU 8.0	2023-04-25 11:20:21 +03:00
Vitaliy Filippov	0d9e10cf96	Add patches for QEMU 6.2-8.0	2023-04-25 11:20:21 +03:00
Vitaliy Filippov	b74ccb613c	Fix another variant of flusher sync-waiting stall	2023-04-24 00:44:41 +03:00
Vitaliy Filippov	5052174918	Fix test_write_no_same (too large image)	2023-04-24 00:44:41 +03:00
Vitaliy Filippov	eec9cf5575	Fix test_snapshot.sh - qemu-img requires explicit backing_fmt	2023-04-24 00:44:41 +03:00
Vitaliy Filippov	a04dab0840	Initialize messenger in cluster_client listings	2023-04-24 00:44:41 +03:00
Vitaliy Filippov	160863f707	Print op pointer values in slow log	2023-04-23 17:54:00 +03:00
Vitaliy Filippov	2f16c32eb4	Fix test_minsize_1 (left_on_dead)	2023-04-23 17:54:00 +03:00
Vitaliy Filippov	2877cd0adb	Allow OP_SEC_READ to return errors (do not hang the connection)	2023-04-23 17:54:00 +03:00
Vitaliy Filippov	480509f5b9	Fix pg_data_size > 1 for replicas (harmless bug)	2023-04-23 01:50:42 +03:00
Vitaliy Filippov	46462da45e	Preload own PG history updates to fix PG state loop possibly applying the old metadata version	2023-04-23 01:50:30 +03:00
Vitaliy Filippov	024c8658f6	Fix missing } in quick start documentation	2023-04-12 18:26:38 +03:00
Vitaliy Filippov	7e958afeda	Release 0.8.7 This release includes a bunch of important bugfixes for erasure-coded setups with disabled immediate_commit. After these fixes, "test_heal" OSD killing test now passes fine with EC: - Fix cluster write stalls with "Error while doing flush on OSD xx: -16 (Device or resource busy)" in OSD logs possible in EC setups with disabled immediate_commit by selectively syncing nonsynced objects on STABILIZE/ROLLBACK (https://github.com/vitalif/vitastor/issues/51) - Fix other EC + disabled immediate_commit problems: - Fix "opcode=5 retval=-2" errors happening on SYNC retries - Fix non-working "pagination" during PG dirty object flushing - Fix write operations not continued correctly after dirty object flushing - Fix incorrect parity read-modify-write calculation when writing into a lost chunk - Fix OSDs losing left_on_dead PG state of non-clean PGs and thus not removing junk data in the cluster - Fix a small memory leak caused by bad indexing of EC recovery matrices - Fix a rare use-after-free in cluster_client caused by a reenterability issue - Fix vitastor-cli create command syntax in the CSI driver - Allow to start OSDs without local store for tests - Fix memory allocation error in disk_tool_meta for non-standard metadata block sizes - Fix delete operations received before loading pool metadata crashing OSDs with "null pointer exception" - Improve "theoretical performance" Russian documentation New features: - Implement online configuration update for some parameters. Documentation is coming soon :)	2023-04-11 02:11:57 +03:00
Vitaliy Filippov	2f5e769a29	Fix a small memory leak caused by bad indexing of EC recovery matrices	2023-04-11 00:30:36 +03:00
Vitaliy Filippov	28d5e53c6c	Add test_heal to run_tests	2023-04-09 02:10:42 +03:00
Vitaliy Filippov	d9f55f11d8	More logs (log_level 10), append to log instead of overwriting on restart in tests	2023-04-09 02:06:10 +03:00
Vitaliy Filippov	3237014608	Fix incorrect parity read-modify-write calculation when writing into a lost chunk	2023-04-09 02:06:10 +03:00
Vitaliy Filippov	baaf8f6f44	Fix write operations not continued correctly after flush	2023-04-09 02:06:10 +03:00
Vitaliy Filippov	1d83fdcd17	Add debug logs to osd_flush	2023-04-09 02:06:10 +03:00
Vitaliy Filippov	0ddd787c38	Fix non-working "pagination" during PG dirty object flushing	2023-04-08 02:44:02 +03:00
Vitaliy Filippov	6eff3a60a5	Do not lose left_on_dead PG state of non-clean PGs	2023-04-08 02:44:02 +03:00
Vitaliy Filippov	888a6975ab	Fix a rare use-after-free in cluster_client caused by a reenterability issue	2023-04-08 02:44:02 +03:00
Vitaliy Filippov	cd1e890bd4	Fix "opcode=5 retval=-2" errors sometimes possible with EC	2023-04-08 02:44:02 +03:00
Vitaliy Filippov	0fbf4c6a08	Selectively sync nonsynced objects on STABILIZE/ROLLBACK (fix for github issue #51 )	2023-04-08 02:44:02 +03:00
Vitaliy Filippov	d06ed2b0e7	Implement online config update	2023-03-26 19:21:50 +03:00
Vitaliy Filippov	3bbc46543d	Fix vitastor-cli create syntax	2023-03-17 11:12:58 +03:00
Vitaliy Filippov	2fb0c85618	Allow to start OSDs without local store (only for tests)	2023-03-15 01:13:59 +03:00
Vitaliy Filippov	d81a6c04fc	Update cmake min version so it does not complain about deprecation	2023-03-15 01:08:23 +03:00
Vitaliy Filippov	7b35801647	Fix possible bad realloc in disk_tool_meta for non-standard metadata block sizes	2023-03-15 01:08:23 +03:00
Vitaliy Filippov	f3228d5c07	Fix typo (did not affect execution though)	2023-03-15 01:08:23 +03:00
Vitaliy Filippov	18366f5055	Fix read/write return type in rw_blocking	2023-03-15 01:08:14 +03:00
Vitaliy Filippov	851507c147	Add missing close() in test stubs	2023-03-15 00:23:56 +03:00
Vitaliy Filippov	9aaad28488	Fix "null pointer exception" for unhandled OSD_OP_DELETEs (when pool is not loaded yet)	2023-03-02 11:16:39 +03:00
Vitaliy Filippov	dd57d086fe	Add a missing part of the "theoretical performance" to the Russian version	2023-03-01 00:24:54 +03:00
Vitaliy Filippov	8810eae8fb	Release 0.8.6 Important fixes: - Fix possibly incorrect EC parity chunk updates with EC n+k, k > 1 and when the first parity chunk is missing Minor fixes and improvements: - Fix incorrect EC free space statistics in vitastor-cli df output - Speedup vitastor-cli startup in clusters with RDMA - Remove unused PG "peered" state (previously used to update PG epoch) - Use sfdisk with just --json in vitastor-disk (--dump --json isn't needed) - Allow trailing comma in sfdisk output (fixes sfdisk 2.36 compatibility) - Slightly improve RDMA send/receive code - Reduce RDMA memory consumption by default (rdma_max_recv/send = 16/8) - Use vitastor-cli instead of direct etcd interaction in the CSI driver	2023-02-28 11:18:48 +03:00
Vitaliy Filippov	c1365f46c9	Use vitastor-cli instead of direct etcd interaction in the CSI driver	2023-02-28 11:02:50 +03:00
Vitaliy Filippov	14d6acbcba	Set default rdma_max_recv/send to 16/8, fix documentation	2023-02-28 11:00:56 +03:00
Vitaliy Filippov	1e307069bc	Fix missing parity chunk calculation for EC n+k, k > 1 and first parity chunk missing	2023-02-28 02:40:19 +03:00
Vitaliy Filippov	c3e80abad7	Allow to send more than 1 operation at a time	2023-02-26 02:01:04 +03:00
Vitaliy Filippov	138ffe4032	Reuse incoming RDMA buffers	2023-02-26 00:55:01 +03:00
Vitaliy Filippov	8139a34e97	Fix json11: allow trailing comma	2023-02-23 01:16:01 +03:00
Vitaliy Filippov	4ab630b44d	Use just sfdisk --json, --dump is not needed	2023-02-23 00:55:47 +03:00
Vitaliy Filippov	2c8241b7db	Remove PG "peered" state	2023-02-21 01:30:42 +03:00
Vitaliy Filippov	36a7dd3671	Move tests to "make test"	2023-02-21 01:30:42 +03:00
Vitaliy Filippov	936122bbcf	Initialize msgr lazily in client to speedup vitastor-cli with RDMA enabled	2023-02-19 18:59:07 +03:00
Vitaliy Filippov	1a1ba0d1e7	Add set_immediate to ringloop and use it for bs/osd ops to prevent reenterability issues	2023-02-09 17:37:26 +03:00
Vitaliy Filippov	3d09c9cec7	Remove unused wait_sqe() from ringloop	2023-02-09 17:37:26 +03:00
Vitaliy Filippov	3d08a1ad6c	Fix cluster_client test after last reenterability fixes	2023-02-05 01:47:32 +03:00
Vitaliy Filippov	499881d81c	Fix typo	2023-01-27 01:52:02 +03:00
Vitaliy Filippov	aba93b951b	Fix incorrect EC free space statistics in vitastor-cli df output	2023-01-26 02:04:29 +03:00
Vitaliy Filippov	d125fb1f30	Release 0.8.5 - Fix a possible "double free" bug in the client library happening on OSD restart - Fix a possible write hang on PG history update when only epoch is changed - Fix incorrect systemd target "local.target" in mon/make-etcd - Allow "content" option in PVE storage plugin to allow to enable containers - Build client library without tcmalloc which fixes "attempt to free invalid pointer" errors when, for example, trying to run QEMU with both Vitastor and Ceph RBD disks	2023-01-25 01:43:49 +03:00
Vitaliy Filippov	9d3fd72298	Require liburing < 2 in rpm specs	2023-01-25 01:43:49 +03:00
Vitaliy Filippov	8b552a01f9	Do not retry successful operation parts in client (could lead to "double free" bugs)	2023-01-25 01:30:36 +03:00
Vitaliy Filippov	0385b2f9e8	Fix write hangs on PG epoch update - always set pg.history_changed to true	2023-01-25 01:30:15 +03:00
Vitaliy Filippov	749c837045	Replace non-existing local.target with multi-user.target	2023-01-25 01:29:31 +03:00
Vitaliy Filippov	98001d845b	Remove version from vitastor-release.rpm links	2023-01-23 14:03:33 +03:00
Vitaliy Filippov	c96bcae74b	Allow "content" option in PVE storage plugin to allow to enable containers	2023-01-16 18:14:45 +03:00
Vitaliy Filippov	9f4e34a8cc	Build client library without tcmalloc Fixes "[src/tcmalloc.cc:332] Attempt to free invalid pointer ..." when trying to run QEMU with both Vitastor and Ceph RBD disks and other possible allocator collisions.	2023-01-15 00:01:11 +03:00
`@@ -1,4 +1,4 @@`
	`VERSION ?= v0.8.4`	`VERSION ?= v0.8.9`

	`all: build push`	`all: build push`
`@@ -1,4 +1,4 @@`
	`[Documentation](../../README.md#documentation) → Usage → Disk Tool`	`[Documentation](../../README.md#documentation) → Usage → Disk management tool`

	`-----`	`-----`
`@@ -1,4 +1,4 @@`
	`[Документация](../../README-ru.md#документация) → Использование → Управление дисками`	`[Документация](../../README-ru.md#документация) → Использование → Инструмент управления дисками`

	`-----`	`-----`