hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Shuai Lin <linshuai2...@gmail.com>
Subject Some regions never get online after a region server crashes
Date Thu, 05 May 2016 01:07:33 GMT
Hi list,

Last weekend I got a region server crashed, but some regions never got
online again on other RSes. I've gone through the logs, and here is the
timeline about some of the events:

* 13:03:50 on of the region server, rs-node7, died because of a disk
failure. Master started to split rs-node7's WALs


2016-04-30 13:03:50,953 INFO
org.apache.hadoop.hbase.master.handler.ServerShutdownHandler: Splitting
logs for rs-node7.example.com,60020,1458724695128 before assignment; region
count=133
2016-04-30 13:03:50,966 DEBUG
org.apache.hadoop.hbase.master.SplitLogManager: Scheduling batch of logs to
split
2016-04-30 13:03:50,966 INFO
org.apache.hadoop.hbase.master.SplitLogManager: started splitting 33 logs
in [hdfs://nameservice1/hbase/WALs/rs-node7.example.com,60020,1458724695128-splitting]
for [rs-node7.example.com,60020,1458724695128]

* 13:10:47 WAL splits done, master began to re-assign regions

2016-04-30 13:10:47,655 INFO
org.apache.hadoop.hbase.master.handler.ServerShutdownHandler: Reassigning
133 region(s) that rs-node7.example.com,60020,1458724695128 was carrying
(and 0 regions(s) that were opening on this server)
2016-04-30 13:10:47,665 INFO
org.apache.hadoop.hbase.master.AssignmentManager: Bulk assigning 133
region(s) across 6 server(s), round-robin=true
2016-04-30 13:10:47,667 INFO
org.apache.hadoop.hbase.master.AssignmentManager: Assigning 22 region(s) to
rs-node1.example.com,60020,1458720625688
2016-04-30 13:10:47,667 INFO
org.apache.hadoop.hbase.master.AssignmentManager: Assigning 22 region(s) to
rs-node2.example.com,60020,1458721110988
2016-04-30 13:10:47,667 INFO
org.apache.hadoop.hbase.master.AssignmentManager: Assigning 22 region(s) to
rs-node3.example.com,60020,1458721713906
2016-04-30 13:10:47,679 INFO
org.apache.hadoop.hbase.master.AssignmentManager: Assigning 23 region(s) to
rs-node4.example.com,60020,1458722335527
2016-04-30 13:10:47,691 INFO
org.apache.hadoop.hbase.master.AssignmentManager: Assigning 22 region(s) to
rs-node5.example.com,60020,1458722992296
2016-04-30 13:10:47,702 INFO
org.apache.hadoop.hbase.master.AssignmentManager: Assigning 22 region(s) to
rs-node6.example.com,60020,1458723856883

* 13:11:47 the opening regions rpc sent by master to region servers timed
out after 1 minutes

2016-04-30 13:11:47,780 INFO
org.apache.hadoop.hbase.master.AssignmentManager: Unable to communicate
with rs-node3.example.com,60020,1458721713906 in order to assign regions
java.io.IOException: Call to rs-node3.example.com/172.16.6.3:60020 failed
on local exception: org.apache.hadoop.hbase.ipc.CallTimeoutException: Call
id=4, waitTime=60001, operationTimeout=60000 expired.
2016-04-30 13:11:47,782 INFO
org.apache.hadoop.hbase.master.GeneralBulkAssigner: Failed assigning 22
regions to server rs-node6.example.com,60020,1458723856883, reassigning them
2016-04-30 13:11:47,782 INFO
org.apache.hadoop.hbase.master.GeneralBulkAssigner: Failed assigning 23
regions to server rs-node4.example.com,60020,1458722335527, reassigning them
2016-04-30 13:11:47,782 INFO
org.apache.hadoop.hbase.master.GeneralBulkAssigner: Failed assigning 22
regions to server rs-node3.example.com,60020,1458721713906, reassigning them
2016-04-30 13:11:47,783 DEBUG
org.apache.hadoop.hbase.master.AssignmentManager: Force region state
offline {a65660e421f114e93862194f7cc35644 state=OPENING, ts=1462021907753,
server=rs-node6.example.com,60020,1458723856883}


* After that, part of the regions (40 out of 130 regions) never got online,
and the following lines were logged repeatly in master log:

2016-04-30 13:12:37,188 DEBUG
org.apache.hadoop.hbase.master.AssignmentManager: update
{d1c7f3f455f2529da82a2f713b5ee067 state=PENDING_OPEN, ts=1462021957087,
server=rs-node6.example.com,60020,1458723856883} the timestamp.
2016-04-30 13:12:37,188 DEBUG
org.apache.hadoop.hbase.master.AssignmentManager: Region is already in
transition; waiting up to 10668ms

$ grep 'AssignmentManager: update {d1c7f3f455f2529da82a2f713b5ee067'
/var/log/hbase/hbase-cmf-hbase-MASTER-head.example.com.log.out.1|wc -l
484


I've searched in mailing list archive and hbase JIRA, but didn't find any
similar situations like this one. The most similar one is HBASE-14407
<https://issues.apache.org/jira/browse/HBASE-14407> , but after reading its
discussion I don't think that's the same problem.

Anyone have a clue? Thanks!

Regards,
Shuai

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message