[jira] [Reopened] (HBASE-17049) Find out why AsyncFSWAL issues much more syncs than FSHLog

["Duo Zhang (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HBASE-17049?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Duo Zhang reopened HBASE-17049:
-------------------------------
      Assignee: Duo Zhang

Talked with [~chancelq] and [~carp84] offline, and we found a problem that may cause AsyncFSWAL issues more syncs than FSHLog.

For AsyncFSWAL, we will first take all the entries out of the ringbuffer, and then do append and sync, but while we do append and sync, there may be more entries inserted to the ringbuffer, but we just ignore them and issue a sync at last even if we have only buffered very small data.

For FSHLog, there is no such problem. We just do take, append, take, append.

> Find out why AsyncFSWAL issues much more syncs than FSHLog
> ----------------------------------------------------------
>
>                 Key: HBASE-17049
>                 URL: https://issues.apache.org/jira/browse/HBASE-17049
>             Project: HBase
>          Issue Type: Sub-task
>          Components: wal
>    Affects Versions: 2.0.0
>            Reporter: Duo Zhang
>            Assignee: Duo Zhang
>            Priority: Blocker
>             Fix For: 2.0.0
>
>         Attachments: delay-sync.patch
>
>
> https://issues.apache.org/jira/browse/HBASE-16890?focusedCommentId=15647590&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15647590



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)