[
https://issues.apache.org/jira/browse/HBASE-14450?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15428455#comment-15428455
]
Ted Yu commented on HBASE-14450:
--------------------------------
The following code is not capable of obtaining highest filenum among WAL files
for the region server:
{code}
hlog = (FSHLog) rss.getWAL(null);
long filenum = hlog.getFilenum();
{code}
This was why sometimes TestIncrementalBackup passed and sometimes it failed.
{code}
2016-08-18 19:12:33,355 INFO
[rs(192.168.0.14,52286,1471572715945)-backup-pool19-thread-1]
regionserver.LogRollBackupSubprocedure$RSRollLogTask(80): Trying to roll log in
backup subprocedure, current log number: 1471572720163 highest:
1471572748368 on 192.168.0.14,52286,1471572715945
{code}
1471572720163 was the filenum for the WAL returned by getWAL(null).
1471572748368 was the highest filenum obtained by iterating all outstanding WAL
files.
> HBase Backup/Restore Phase 3: Multiwal support
> ----------------------------------------------
>
> Key: HBASE-14450
> URL: https://issues.apache.org/jira/browse/HBASE-14450
> Project: HBase
> Issue Type: Task
> Affects Versions: 2.0.0
> Reporter: Vladimir Rodionov
> Assignee: Ted Yu
> Labels: backup
> Fix For: 2.0.0
>
> Attachments: 14450-HBASE-7912.v1.txt, 14450-branch-7912.v1.txt,
> 14450-roll.v2.txt, 14450.HBASE-7912.v1.txt, 14450.roll-wait.txt,
> 14450.v2.txt,
> org.apache.hadoop.hbase.backup.TestIncrementalBackup-output.txt,
> testIncrementalBackup-8-15.txt, testIncrementalBackup-8-18.txt,
> testIncrementalBackup-succ.txt
>
>
> We need to support multiwal configurations.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
