hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ramkrishna.s.vasudevan (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-17035) Check why we roll a wal writer at 10MB when the configured roll size is 120M+ with AsyncFSWAL
Date Mon, 07 Nov 2016 08:39:59 GMT

    [ https://issues.apache.org/jira/browse/HBASE-17035?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15643512#comment-15643512

ramkrishna.s.vasudevan commented on HBASE-17035:

I think with HBASE-17021 the accounting of the entries is corrected
2016-11-07 18:59:33,465 INFO  [regionserver/stobdtserver5/] wal.AbstractFSWAL:
Rolled WAL /hbase1/WALs/stobdtserver5,16041,1478525130117/stobdtserver5%2C16041%2C1478525130117.1478525364568
with entries=14376, filesize=522.53 MB; new WAL /hbase1/WALs/stobdtserver5,16041,1478525130117/stobdtserver5%2C16041%2C1478525130117.1478525373221
2016-11-07 18:59:33,822 INFO  [regionserver/stobdtserver5/] wal.AbstractFSWAL:
Rolled WAL /hbase1/WALs/stobdtserver5,16041,1478525130117/stobdtserver5%2C16041%2C1478525130117.1478525373221
with entries=488, filesize=17.62 MB; new WAL /hbase1/WALs/stobdtserver5,16041,1478525130117/stobdtserver5%2C16041%2C1478525130117.1478525373466

On the part you say on waitForSafePoint. I will again check from logs.

But my point is (even after HBASE-17021). The waitingRoll is set inside 'consumeLock.lock();'
So by the time the waitForSafePoint tries to acquire the lock we could have added new entries
and did a sync call which calls syncCompleted()??
Also I can see sometimes we get requestLogroll from syncCompleted for the same length
Requesting log roll 537902596
Requesting log roll 537902596

> Check why we roll a wal writer at 10MB when the configured roll size is 120M+ with AsyncFSWAL
> ---------------------------------------------------------------------------------------------
>                 Key: HBASE-17035
>                 URL: https://issues.apache.org/jira/browse/HBASE-17035
>             Project: HBase
>          Issue Type: Sub-task
>          Components: wal
>    Affects Versions: 2.0.0
>            Reporter: Duo Zhang
>            Assignee: Duo Zhang
>             Fix For: 2.0.0
> Found this when addressing HBASE-16890. It is one of the possible reason that why AsyncFSWAL
performs worse than FSHLog when running PE tool.
> https://issues.apache.org/jira/browse/HBASE-16890?focusedCommentId=15636688&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15636688

This message was sent by Atlassian JIRA

View raw message