Releases: mantechnology/wdrbd
1.3.5
Bug
[DW-1234] - fix NULL dereference case for Waiter.
[DW-1236] - fix wait_for_completion is not completed for down / down case
[DW-1239] - fix local i/o hang, for peer memory allocator's bug .
[DW-1240] - fix crash for peer disconnect
[DW-1243] - fix crash Netlink's global variables
1.3.2
Bug
[DW-1202] - fix released secondary lock
[DW-1203] - fix Permission denied error for drbdadm status
[DW-1205] - fix SyncSource 100%, resync stop,
Improvement
[DW-1201] - increase performance for local I/O
[DW-1204] - disconnect process is long time for low latency network
1.3
We got Fast Sync.
The previous initial sync time was very long when disk size is big.
we add a Fast (initial) Sync to WDRBD 1.3. Fast Sync perform to resync only used disk space based on File systems(NTFS/ReFS).
New Feature
[DW-844] - Fast Sync
Bug
[DW-1152] - fix inconsistency while reconnecting
[DW-1163] - fix split-brain while resync progress and switch-over
[DW-1173] - fix 2pc timeout while initial sync node switch-over and disconnect
[DW-1181] - while primary node shutdown, drbd_md_sync_page_io error occurred
[DW-1191] - fix inconsistency on Ahead mode
[DW-1198] - fix resync stop that change to Established on Ahead
[DW-1166] - fix disconnect on Async protocol
[DW-1171] - fix spinlock recursion on multiple resources env.
[DW-1200] - fix memory overflow on low latency network
Task
[DW-1170] - LDRBD patch 6c5ae94 to 24f8aaa
1.2
Finally, WDRBD provide 1:2 replications, but we restrict functions to WDRBD 1.2.
Only Primary node can be sync source, because original drbd9 core engine is not good to operate in 1:N node environment. (for examples, it will be occurred that the replication is broken after secondary sync source go to primary (change role) and replication I/O is proceed.)
then, we decided that only primary node can be sync source in WDRBD policy.
we call this as a restricted mesh topology replication.
New Features
[DW-1142] - Only Primary node can be sync source
Bug
[DW-960] - fix oos remaining after disk detach
[DW-1114] - fix outdate and oos remaining after network fail
[DW-1127] - fix WFBitMapS continue.
[DW-1131] - fix remaining WFBitMapT/WFBitMapS status
[DW-1139] - fix disconnecting after inittial full sync and switch-over
[DW-1140] - drbdcon /info insufficient memory error
[DW-1144] - Meta disk IO error after reconfig.
[DW-1146] - shutdown hang
[DW-1151] - primary auto resync feature's bug(cannot full sync)
[DW-1155] - limit to change primary role for Outdated node.
[DW-1159] - remaining Behind after go to secondary in Ahead
[DW-1160] - dual primary occured in 1:2 star topology
1.1.11
[DW-1111] - fix resync stop while replication I/O processing
[DW-1113] - go to outdated for reconnecting
[DW-1116] - adjust forget bitmap logic
[DW-1121] - 1:1 Stop to Secondary syncSource 100%
[DW-1134] - fix crash bugcheck that invalidate from a incorrect node id
[DW-1136] - clear bitmap uuid when resync is done
1.1.10
Bug
[DW-777] - fix cstate output mispatch in multiple resources
[DW-955] - OOS remaining for P_PEERS_IN_SYNC
[DW-982] - after Split Brain , resync doesn't finished
[DW-1009] - while Invalidate, resync stop in PausedSyncSource
[DW-1034] - [1:2]when connect with Outdated node, full sync progess
[DW-1044] - crash while disk detach
[DW-1073] - while disconnect, secondary node hang in 1:2 Env.
[DW-1117] - patch flush io memory leak
Task
[DW-1083] - LDRBD patch d91a174 to a14cb9c
[DW-1084] - drbd-utils LDRBD patch bc01a46 to 7ea0389
[DW-1088] - get wdrbd volume info
1.1.9
Bug
[DW-1057] - fix crash while down-up repeatedly
[DW-1059] - [DV] accessing freed-memory area while drbd attach
[DW-1067] - [DV] fix crash while SyncTarget meta disk is removed
[DW-1074] - fix crash while drbd detach/ attach
[DW-1078] - fix WskCloseSocket DRIVER_PAGE_FAULT_IN_FREED_SPECIAL_POOL
[DW-1080] - fix issue that Primary-WFBitMapS status is not changed.
1.1.8
1.1.8
Bug
[DW-1063] - DV violation wait at dispatch level in drbd_send_sizes
1.1.7
Bug
[DW-1046] - fix hang while drbdsetup down
[DW-1049] - fix crash while drbd_adm_attach
[DW-1051] - fix crash . loglink memory overflow
[DW-1061] - DV Violation wait at dispatch level
Improvement
[DW-996] - Improve connection fail in case of Disk size's differency
1.1.6
1.1.6
Bug
[DW-926] - fix call SB Handler
[DW-956] - after intial config, leave outdated
[DW-979] - fix sync freeze while secondary node's up/down repeatedly
[DW-980] - fix sync freeze after secondary reconnect
[DW-1010] - fix install fail issue on unicode user account name
[DW-1011] - fix OOS remaining after Secondary power reset in 1:1 Async
[DW-1012] - fix primary OOS remaining after io stress on AHEAD mode
[DW-1015] - fix crash(WskSend NULL dereference) while reboot
[DW-1017] - fix switch-over fail on Mirror Disk resource
[DW-1024] - fix replication problem on Shared volume config
[DW-1029] - fix loglink crash
[DW-1030] - fix resync problem in case of 1:1 Outdated node
[DW-1036] - fix Loglink wait at completion routine
[DW-1039] - adjust c-plan-ahead default value
Task
[DW-1016] - LDRBD patch 2de4880 to 605195a
1.1.2
1.1.2
Bug
[DW-885] - fix 1:2 resync problem(after network is disconnected on Sync/Replication, resync progress is not done in Ahead mode).
[DW-886] - fix crash issue while disk is removing from disk manager.
[DW-892] - fix outdated issue. (after 1:2 sync is done, one node's state is Outdated).
[DW-897] - fix resync issue (while 1:2 invalidate-remote on primary, resync problem occurred on reconnect progress).
[DW-900] - fix system hang after primary power reset
[DW-903] - fix spinlock recurion in drbd_free_peer_reqs.
[DW-907] - fix SyncTarget resync problem issue.
[DW-917] - fix oos issue(negative value).
[DW-918] - fix "Failed to RegCreateKeyEx( ). Status(0x5)" error
[DW-927] - fix crash on up-down-up
[DW-934] - adjust drbd.conf's Environment variables.
[DW-950] - fix timeout hang while down
Task
[DW-933] - LDRBD patch a49b0a7 to 14d2eea