828 Commits (88516ab4bd4fefc278cbc696cad1e6cb2d9cdd67)

Author	SHA1	Message	Date
Vitaliy Filippov	19426aa4c5	Add JSON output for stress-tester	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	08f586bcec	Print total stats	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	f1cd87473a	Do not send more than op_count operations (fix segfault on finish)	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	1bd8d2da56	Add some more resiliency to serialize()	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	a7396d2baf	Invalidate blocks being updated too	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	e98a38810d	Change new block allocation method: make each writer choose multiple empty PG blocks and place blocks in them	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	28c4324c36	Remove blocks from cache on unsuccessful updates	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	31ec3fa8f5	Allow to track multiple updates per block (it should never happen though)	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	e4fa26f60a	Do not call stop_updating after failed write_new_block and after clear_block (both delete the item)	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	59ae27f9e5	Track versions of parent blocks and recheck if changed during update	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	2c6a301d9b	Fix resume_split condition (key_lt can also be "")	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	01558349f8	Experiment: transform offsets for better sharding	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	36f4717d0d	More post-stress-test fixes ... - Prevent _split types of new blocks - Stop updating new blocks only after the whole update, otherwise pointers may become invalid - Use recheck_none for updates initially - Use UINT64_MAX as initial block version when postponing ops, otherwise the check fails when the block is initially empty. This for example leads to writing both leaf items & block pointers (which is incorrect) into the root block when starting stress-test with --parallelism 32 - Fix -EINTR comparison	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	babaf2a0ce	Print operation statistics	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	5773f1a375	K/V fixes after stress-test :-) ... - track block versions correctly - per inode block (128kb) instead of tree block (4kb) - prevent multiple parallel CAS writes of the same inode block - add logging for EILSEQ which means invalid data in the tree - fix get_block updated flag which was true for blocks already in cache and was leading to infinite loops on "unrelated block" errors - apply changes to blocks in cache only after successful writes (using "virtual changes") - do not replace cached block with an older version from disk - recheck "unrelated blocks" (read/update collisions) until data stops changing - track tree path correctly - do not treat split block as parent of its right half - correctly move blocks when finding new empty place on disk - restart updates from the beginning when one of blocks is changed by a parallel update - fix delete using SET opcode and setting key to the empty value instead - prevent changing the same key more than 1 time in parallel - fix listing verification - resume continue_updates in update_find (required because it uses continue_update itself) - add allow_old_cached parameter to get()	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	57222a9f79	Implement K/V DB stress tester	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	61ef000c6e	Evict blocks based on memory limit & block usage	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	7d5e1cc393	Track blocks per level	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	5e7f27a02d	Track block level	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	fd1d8a8520	Experimental B-Tree Vitastor embedded K/V database implementation!	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	c364e14c40	Stop then retry, not retry then stop	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	3ebbfa0428	Fix another rare OSD hang on zeroing out entries on start	2024-03-16 13:24:36 +03:00
Vitaliy Filippov	aa79d1db1c	Fix incorrect "changing scheme" message in modify-pool	2024-03-06 00:41:35 +03:00
Vitaliy Filippov	a1fecb7eff	Move callback away when calling it in cluster_client	2024-03-06 00:41:35 +03:00
Vitaliy Filippov	ff74b19423	Fix rare OSD hang on zeroing out bad entries on start	2024-03-06 00:41:35 +03:00
Vitaliy Filippov	4cf6dceed7	Merge branch 'rel-1.4'	2024-02-29 09:59:01 +03:00
Vitaliy Filippov	38b8963330	Release 1.4.8 ... - Do not use \r if output is not a terminal (should fix unexpected job output in proxmox) - Fix rm/rm-data error return code, add --down-ok option to bypass the error - Add EIO retry timeout and allow to disable these retries, rename up_wait_retry_interval to client_retry_interval - Add ubuntu jammy build - Wait for blockstore initialisation before starting OSD (prevent timeouts when init takes time) - Fix a rare use-after-free in automatic sync after delete in blockstore	2024-02-29 09:58:34 +03:00
Vitaliy Filippov	77167e2920	Do not use \r if output is not a terminal	2024-02-29 00:21:17 +03:00
Vitaliy Filippov	5af23672d0	Fix rm/rm-data error return code, add --down-ok option to bypass the error	2024-02-29 00:20:10 +03:00
Vitaliy Filippov	6bf1f539a6	Add EIO retry timeout and allow to disable these retries, rename up_wait_retry_interval to client_retry_interval	2024-02-28 13:10:02 +03:00
Vitaliy Filippov	4eab26f968	Add documentation and a very basic test for pool management commands	2024-02-28 13:08:04 +03:00
Vitaliy Filippov	86243b7101	Rework & fix pool-create / pool-modify / pool-ls	2024-02-28 13:08:04 +03:00
idelson	dc92851322	vitastor-cli: add commands to control pools: pool-create, pool-ls, pool-modify, pool-rm ... PR #59 - https://github.com/vitalif/vitastor/pull/58/commits By MIND Software LLC By submitting this pull request, I accept Vitastor CLA	2024-02-28 13:08:04 +03:00
Vitaliy Filippov	fc413038d1	Wait for blockstore initialisation before starting OSD	2024-02-27 02:20:04 +03:00
Vitaliy Filippov	1bc0b5aab3	Fix a rare use-after-free in automatic sync after delete in blockstore ... ASan report: [0] READ of size 16 at operator() /root/vitastor/src/blockstore_write.cpp:100 ...[5] blockstore_impl_t::ack_sync(blockstore_op_t*) /root/vitastor/src/blockstore_sync.cpp:232	2024-02-24 00:06:36 +03:00
Vitaliy Filippov	5e934264cf	Release 1.4.7 ... - Fix another old "BUG: Attempt to overwrite used offset" in a very simple case: bs=4k rw=write iodepth=16 from OSD start; add this case to tests - Fix a rare crash with "unexpected state during flush: 0x51" possible with EC since 1.4.2 during rebalance and OSD outages - Fix a rare write stall with EC & immediate_commit=none caused by sync operations reserving unneeded space in the journal - Fix 32-bit build warnings, most in printf/scanf format strings	2024-02-22 12:45:52 +03:00
Vitaliy Filippov	f20564b44b	Fix 32-bit build warnings (99.9% in printf)	2024-02-22 12:22:16 +03:00
Vitaliy Filippov	b3c15db331	32M journal by default in simple-offsets	2024-02-21 15:25:02 +03:00
Vitaliy Filippov	685bcd6ef9	Do not reserve extra space for big_writes during sync - sync itself is needed to commit and clear them	2024-02-21 13:00:14 +03:00
Vitaliy Filippov	3eb389b321	Supposed fix for "unexpected state during flush: 0x51" with EC	2024-02-21 01:32:06 +03:00
Vitaliy Filippov	3d16cde23c	Fix assertions, add small sequential write test	2024-02-20 19:41:48 +03:00
Vitaliy Filippov	c6406d67fc	Fix journal space_check incorrectly checking for space at the beginning	2024-02-20 19:40:56 +03:00
Vitaliy Filippov	f87964861d	Release 1.4.6 ... Unwavering stabilization of 1.4.x, continued :-) - Include the accidentally lost part of 1.4.5 journal trimming fix - Fix a possible OSD crash with "BUG: Attempt to overwrite used offset" which was probably present for long time, but became apparent after fixing flapping tests in CI - Fix remaining flapping tests in CI. It was the first time when tests actually passed without retries :-)	2024-02-20 17:01:26 +03:00
Vitaliy Filippov	7048228678	Supposed fix for "BUG: Attempt to overwrite used offset"	2024-02-20 15:56:48 +03:00
Vitaliy Filippov	ea73857450	Add asserts to catch "BUG: Attempt to overwrite used offset"	2024-02-20 15:56:48 +03:00
Vitaliy Filippov	6cfe38ec04	Followup to empty cur.oid as stop condition for forced trim fix	2024-02-20 15:56:38 +03:00
Vitaliy Filippov	f882c7dd87	Release 1.4.5 ... - Fix a write stall caused by incorrect journal trimming introduced in 1.4.4 :) - Fix PGs sometimes hanging in "starting" state on mass OSD restarts - Fix a rare crash with "map::at" during OSD pings - Use new defaults for non-capacitor (desktop) SSDs - improves T1Q256 random write from ~6k iops to ~45k iops - Make journal_trim_interval configurable	2024-02-16 10:13:33 +03:00
Vitaliy Filippov	26dd863c8d	Fix sometimes possible crash on clients.at() during pings	2024-02-16 10:13:33 +03:00
Vitaliy Filippov	2ae859fbc6	Use min/max_flusher_count=32/256, 128M journal and autosync_writes=512 for non-capacitor SSDs by default	2024-02-16 10:13:33 +03:00
Vitaliy Filippov	8389c0f33b	Fix PGs sometimes hanging in "starting" state on mass OSD restarts	2024-02-15 23:38:52 +03:00
Vitaliy Filippov	9db2196aef	Make journal_trim_interval configurable	2024-02-15 23:38:51 +03:00
Vitaliy Filippov	8d6ae662fe	Use empty cur.oid as stop condition for forced trim, not journal_trim_counter	2024-02-15 23:27:17 +03:00
Vitaliy Filippov	c777a0041a	Release 1.4.4 ... A couple of fixes for EC pools - Fix a segfault possible on partial EC overwrite in 1234 -> 5030 rebalance scenario - Fix two problems leading to EC pools stalling on rebalance & parallel sudden stops of OSDs, for example during a sudden poweroff of a host: - Recovery auto-tuning (1.4.0 feature) could apply too large delays and stall the EC journal - fixed by limiting delays with a new recovery_tune_sleep_cutoff_us parameter (10 seconds by default) and applying recovery pauses before write operations, not after them, to not occupy space in the journal for long time - Dynamic journal space reservation (1.3.0 feature) wasn't accounting new writes when checking the limit so OSDs could still fill the journal fully and stall - fixed by including new writes into the limit - Print etcd dbSize instead of dbSizeInUse in status	2024-02-11 16:23:08 +03:00
Vitaliy Filippov	978bdc128a	Apply recovery pause before writes, after commits, and do not apply it to syncs to not block EC pools from functioning	2024-02-11 16:13:52 +03:00
Vitaliy Filippov	bb2f395f1e	Add cutoff threshold for recovery auto-tuning	2024-02-11 16:13:52 +03:00
Vitaliy Filippov	b127da40f7	Add a FIXME about incomplete PGs	2024-02-11 13:42:51 +03:00
Vitaliy Filippov	ca34a6047a	Fix dynamic journal space reservation: include the new write itself, too	2024-02-11 13:42:51 +03:00
Vitaliy Filippov	38ba76e893	Fix flusher sometimes being unable to trim journal when the flush queue is empty	2024-02-11 13:42:51 +03:00
Vitaliy Filippov	1e3c4edea0	Print etcd dbSize instead of dbSizeInUse in status	2024-02-11 13:42:51 +03:00
Vitaliy Filippov	e7ac855b07	Fix that EC segfault (1234 -> 5030 partial overwrite)	2024-02-11 13:42:51 +03:00
Vitaliy Filippov	c53357ac45	Add a test for EC segfault with partial overwrite in 1234 -> 5030 rebalance scenario	2024-02-11 13:42:51 +03:00
Vitaliy Filippov	27e9f244ec	Release 1.4.3 ... Hotfix for hotfix O:-) - "Write stall fix" was incomplete and EC write stalls could continue even on 1.4.2. Now they're finally fixed O:-) - Make monitor ignore statistics of stopped OSDs. Previously if you stopped all OSDs the last total I/O numbers would remain the same indefinitely	2024-02-09 00:29:31 +03:00
Vitaliy Filippov	5d3317e4f2	Followup to 1.4.2 write stall fix - sadly, the previous version was not working correctly :)	2024-02-08 19:34:29 +03:00
Vitaliy Filippov	016115c0d4	Release 1.4.2 ... - Log to systemd by default - Fix excessive autosyncs after every operation with disabled immediate_commit (introduced in 1.1.0) - Fix a possible write stall with EC due to the lack of OSD wakeup after stabilizing previous writes - Change sync operation semantics as a final fix to possible write stalls with EC and disabled immediate_commit - Sync after deleting data in CLI rm / rm-data if immediate_commit is disabled - Fix OSDs ignoring syncs & autosyncs for delete operations - Fix OSD space reporting sometimes adding garbage zeros for deleted inodes (causing extra pool/stats etcd keys for deleted pools) - Speed up monitor failover - change default etcd_mon_ttl from 30 to 5 seconds - Speed up operation retries - change default up_wait_retry_interval to 50 ms - Add patch for libvirt 9.10	2024-02-04 02:23:49 +03:00
Vitaliy Filippov	77c10fd1f8	In fact, do not autosync blockstore when autosync_writes=0	2024-02-03 20:37:36 +03:00
Vitaliy Filippov	581d02e581	Mark secondary OSDs with deletions as dirty to not forget to sync & autosync them	2024-02-03 20:31:08 +03:00
Vitaliy Filippov	f03a9db4d9	Fix OSD space reporting sometimes adding garbage zeros for deleted inodes (causing extra pool/stats etcd keys for deleted pools)	2024-02-03 20:31:08 +03:00
Vitaliy Filippov	cb9c30bc31	Sync after sending all deletes to each PG in cli rm-data	2024-02-03 20:31:08 +03:00
Vitaliy Filippov	a86a380d20	Fix invalid parsing of autosync_writes in blockstore leading to autosyncs after every operation with disabled immediate_commit :D	2024-02-03 20:31:08 +03:00
Vitaliy Filippov	1cec62d25d	Sync only completed writes ... Should be a final remaining fix to EC + non-capacitor (non-immediate-commit) write hangs :). First it was breaking non-EC ("instantly stable") writes because they sometimes complete out of order which was leading to the following error: terminate called after throwing an instance of 'std::runtime_error' what(): BUG: Unexpected dirty_entry 1000000000001:29480000 v65540 unstable state during flush: 0x151 But it is easily fixed by scanning previous and next dirty_entries in mark_stable.	2024-01-27 15:17:22 +03:00
Vitaliy Filippov	1c322b33ed	Change default up_wait_retry_interval to 50 ms	2024-01-26 01:51:08 +03:00
Vitaliy Filippov	ba55f91409	Release 1.4.1 ... - Fix a monitor crash on primary OSD switching introduced in 1.4.0 - Fix "partly outside array bounds" warnings for GCC 12 in cpp-btree - Fix a realloc memory leak in theory possible with too large listings (OSD_OP_LIST)	2024-01-18 02:31:42 +03:00
Vitaliy Filippov	d00d4dbac0	Initialize mod_revision field in etcd_state_client	2024-01-13 01:30:28 +03:00
Vitaliy Filippov	5d9d6f32a0	Fix common realloc memory leak mistakes found by cppcheck	2024-01-13 01:30:28 +03:00
Vitaliy Filippov	5280d1d561	Release 1.4.0 ... New features: - Intelligent recovery/rebalance speed auto-tuning to reduce its impact on clients (see README -> Features) - Auto-restoration of dead VDUSE daemons in CSI plugin - Add vitastor-disk update-sb command - Update QEMU for Debian Bookworm to 8.1 and use it for CSI plugin Bug fixes: - Fix pools SOMETIMES staying inactive after stopping a node due to OSDs not reacting to PG state changes caused by incorrect full reload of state from etcd on reconnection - Make monitors retry pool configuration changes quickier which fixes them being unable to apply changes when an ongoing rebalance is quickly making a lot of PGs clean - Fix CSI plugin not accepting array of strings as etcd address in /etc/vitastor/vitastor.conf - Allow multiple interfaces with the same IP address, for "simple routed" full mesh network - Do not ignore loopback addresses for OSD network (to make ECMP setups with frr possible) - Fix a rare client crash during OSD reconnections - Only treat data partitions as existing OSDs in vitastor-disk prepare - Remove etcd parameter from default command examples - Fix reported free space sometimes changing non-immediately after deletion of data from OSDs - Fix a possible OSD crash on print_slow when bs_op is NULL - Use the same etcd_ws_keepalive_interval in mon as in OSD - Fix mon not using values from config when /config/global is not present - Remove pve-storage-portal-dns-list format for vitastor_etcd_address - Parse log_level in cluster_client - Fix vitastor-nbd image existence check not working because of non-zeroed inode_watch fields - Do not warn on EPIPE in client unless log_level is raised explicitly - Fix incorrect error in CSI when searching for the device in /sys - Remove 2 last prints to stdout in etcd_state_client - Fix a possible OSD crash when checking corrupted journal entries	2024-01-12 01:28:33 +03:00
Vitaliy Filippov	2f228fa96a	Only treat data partitions as existing OSDs in vitastor-disk prepare	2023-12-31 11:46:47 +03:00
Vitaliy Filippov	a6ab54b1ba	Do not allow negative util_low/high	2023-12-31 01:23:17 +03:00
Vitaliy Filippov	99ee8596ea	Rename min/max_util to util_low/high	2023-12-31 01:23:17 +03:00
Vitaliy Filippov	c4928e6ecd	Protect from try_send completing the operation immediately ... Fixes a possible use-after-free in case of continue_ops() calling try_send(), then connect_peer() -> set_timer() -> trigger_nearest() -> handle_op_part() -> continue_ops() again	2023-12-31 01:23:17 +03:00
Vitaliy Filippov	8b8c1179a7	Use a separate used_blocks counter for free space stats to hide possibly delayed on-flush deallocation	2023-12-31 01:23:17 +03:00
Vitaliy Filippov	d5a6fa6dd7	Fix possible crash on print_slow when bs_op is NULL	2023-12-31 01:23:17 +03:00
Vitaliy Filippov	1edf86ed26	Aggregate recovery delay using simple mean over last 10 observations (EWMA is shit)	2023-12-31 01:23:17 +03:00
Vitaliy Filippov	5ca7cde612	Experiment/WIP: Try to track "secondary" recovery ops separately	2023-12-31 01:23:17 +03:00
Vitaliy Filippov	751935ddd8	WIP Auto-tune recovery speed	2023-12-31 01:23:17 +03:00
Vitaliy Filippov	d84dee7098	Track recovery op latencies + refactor into a structure	2023-12-31 01:23:17 +03:00
Vitaliy Filippov	f72f14e6a7	Clear old PG states, history, and OSD states on etcd state reload ... Also add protection from etcd watcher messages being split into multiple websocket messages - I'm not sure if etcd actually does that, but it's better to have extra protection anyway. Also check that all etcd watchers are started in the keepalive routine, otherwise it sometimes tries to revive etcd watchers starting with revision=1 which obviously always fails because this revision is nearly always compacted. All these changes should fix an old rarely reproduced bug where SOMETIMES OSDs didn't react to PG config changes which was leading to offline pools on node reboot. It happened on the full reload of state from etcd.	2023-12-24 02:02:13 +03:00
Vitaliy Filippov	178bb0e701	Prevent re-entry into timerfd set_nearest	2023-12-22 02:32:40 +03:00
Vitaliy Filippov	7239cfb91a	Parse log_level in cluster_client	2023-12-20 02:21:23 +03:00
Vitaliy Filippov	7cea642f4a	Fix vitastor-nbd image existence check not working because of non-zeroed inode_watch fields	2023-12-19 01:11:37 +03:00
Vitaliy Filippov	dc615403d9	Do not warn on EPIPE in client unless log_level is raised explicitly	2023-12-17 13:42:26 +03:00
Vitaliy Filippov	1a704e06ab	Allow multiple interfaces with the same IP address, for "simple routed" full mesh network	2023-12-17 13:25:56 +03:00
Vitaliy Filippov	575475de71	Do not ignore loopback addresses for OSD network (to make ECMP setups with frr possible)	2023-12-17 11:55:13 +03:00
Vitaliy Filippov	aca2bef15f	Add vitastor-disk update-sb command	2023-12-14 01:11:42 +03:00
Vitaliy Filippov	691ebd991a	Move 2 last log printfs to stderr from stdout in etcd_state_client	2023-12-08 00:01:52 +03:00
Vitaliy Filippov	6d5df908a3	Fix possible out of bounds when checking invalid journal entries	2023-12-08 00:01:07 +03:00
Vitaliy Filippov	a1c7cc3d8d	Release 1.3.1 ... Hotfix to 1.3.0 - new "journal space reservation" had a bug which caused OSDs to crash with EC and without immediate_commit.	2023-12-04 18:35:09 +03:00
Vitaliy Filippov	a5e3dfbc5a	Oops, 1.3.0 needs a hotfix	2023-12-04 13:45:54 +03:00
Vitaliy Filippov	7972502eaf	Release 1.3.0 ... New features: - RDMA without ODP - much faster and all cards are now supported, not just Mellanox - VDUSE in CSI - faster, more stable and can even recover after CSI pod restart! - Reserve journal space for stabilize requests dynamically to prevent stalls under load with EC - Raise default NBD timeout from 30 to 300 seconds and allow to take it from /etc/vitastor/vitastor.conf - Remove explicit etcdUrl/etcdPrefix K8S storage class parameter support to prevent etcd migration issues for volumes created with these parameters - Support QEMU 8.1 and pve-qemu 8.1 Bug fixes: - Fix RDMA connection (and thus memory) leak - Fix rare crashes under load due to incorrect io_uring queue size tracking - Fix monitor statistics aggregation in case of empty /osd/stats keys - Fix crash on unknown long argument to vitastor-disk - Allow trailing comma in JSONs again - Fix crash on attempts to dump a long listing of objects "to stabilize" or "to rollback" in a slow op	2023-12-04 02:36:43 +03:00
Vitaliy Filippov	845454742d	Fix warning with QEMU 8.1	2023-12-04 01:59:07 +03:00
Vitaliy Filippov	628aa59574	Raise default NBD timeout from 30 to 300 seconds and allow to take it from /etc/vitastor/vitastor.conf	2023-12-02 14:11:14 +03:00