[jira] Updated: (HBASE-1013) Add debugging around commit log cleanup

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

     [ https://issues.apache.org/jira/browse/HBASE-1013?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack updated HBASE-1013:
-------------------------

    Attachment: 1013.patch

Patch that adds detail to HLog logging.  Compounds a few log lines and adds detail on count of logs and who has oldest sequence id.

{code}
2008-11-20 17:22:26,955 INFO org.apache.hadoop.hbase.regionserver.HLog: Closed file:/tmp/hbase-stack/hbase/log_127.0.0.1_1227230509150_59629/hlog.dat.0, entries=100004. New log writer: /tmp/hbase-stack/hbase/log_127.0.0.1_1227230509150_59629/hlog.dat.1227230546926
2008-11-20 17:22:26,958 DEBUG org.apache.hadoop.hbase.regionserver.HLog: Found 0 logs to remove  out of total 1; oldest outstanding seqnum is 2 from region -ROOT-,,0
{code}

Refactored HLog a little too making smaller methods out of big ones.

> Add debugging around commit log cleanup
> ---------------------------------------
>
>                 Key: HBASE-1013
>                 URL: https://issues.apache.org/jira/browse/HBASE-1013
>             Project: Hadoop HBase
>          Issue Type: Bug
>            Reporter: stack
>             Fix For: 0.19.0
>
>         Attachments: 1013.patch
>
>
> Yesterday, streamy replayed 1000 logs.  This seems too many for any one regionserver to be carrying.  Its hard to tell if the 1000 logs were legit though because our logging isn't detailed enough.  This issue is about adding detail around log cleaning so we can see how many logs are to be cleaned and which region the last edit belongs to.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.