[HBASE-7006] [MTTR] Improve Region Server Recovery Time - Distributed Log Replay

• Type: New Feature
• Status: Closed
• Priority: Critical
• Resolution: Fixed
• Fix Version/s: 0.98.0, 0.95.1
• Component/s: MTTR
• Hadoop Flags: Reviewed

Release Note

Distributed Log Replay Description:

After a region server fails, we firstly assign a failed region to another region server with recovering state marked in ZooKeeper. Then a SplitLogWorker directly replays edits from WAL(Write-Ahead-Log)s of the failed region server to the region after it's re-opened in the new location. When a region is in recovering state, it can also accept writes but no reads(including Append and Increment), region split or merge.

The feature piggybacks on existing distributed log splitting framework and directly replay WAL edits to another region server instead of creating recovered.edits files.

The advantages over existing log splitting recovered edits implementation:
1) Eliminate the steps to write and read recovered.edits files. There could be thousands of recovered.edits files are created and written concurrently during a region server recovery. Many small random writes could degrade the overall system performance.
2) Allow writes even when a region is in recovering state. It only takes seconds for a failed over region to accept writes again.

The feature can be enabled by setting hbase.master.distributed.log.replay to true (by default is false)

• Assignee: Jeffrey Zhong
• Reporter: Michael Stack
• Created: 17/Oct/12 23:47
• Updated: 30/Oct/17 22:49
• Resolved: 15/May/13 04:28

Description

Just saw interesting issue where a cluster went down hard and 30 nodes had 1700 WALs to replay. Replay took almost an hour. It looks like it could run faster that much of the time is spent zk'ing and nn'ing.

Putting in 0.96 so it gets a look at least. Can always punt.

Issue Links

• depends upon: HBASE-14028 DistributedLogReplay drops edits when ITBLL 125M
• is related to: HBASE-8560 TestMasterShutdown failing in trunk 0.95/trunk -- "Unable to get data of znode /hbase/meta-region-server because node does not exist (not an error)"
• is related to: HBASE-8567 TestDistributedLogSplitting#testLogReplayForDisablingTable fails on hadoop 2.0
• is related to: HBASE-7825 Retire non distributed log splitting related code
• is required by: HBASE-5843 Improve HBase MTTR - Mean Time To Recover
• relates to: HBASE-11280 Document distributed log replay and distributed log splitting
• relates to: HBASE-8701 distributedLogReplay need to apply wal edits in the receiving order of those edits
• relates to: HBASE-8729 distributedLogReplay may hang during chained region server failure
• relates to: HBASE-8568 Test case TestDistributedLogSplitting#testWorkerAbort failed intermittently
• relates to: HBASE-8573 Store last flushed sequence id for each store of region for Distributed Log Replay
• relates to: HBASE-8617 Introducing a new config to disable writes during recovering
• relates to: HBASE-8575 TestDistributedLogSplitting#testMarkRegionsRecoveringInZK fails intermittently due to lack of online region
• supercedes: HBASE-6984 Serve writes during log split

Sub-Tasks

• Test for split edit file ordering using failover scenario: Closed
• Implementation of the log splitting without creating intermediate files: Closed
• Create a new "replay" command so that recovered edits won't mess up normal coprocessing & metrics: Closed
• Add new metrics to better monitor recovery process: Closed
• Introducing "recovering" region state in AM to mark a region in recovering status used in distributedLogReplay: Closed