You are viewing a plain text version of this content. The canonical link for it is here.
Posted to oak-issues@jackrabbit.apache.org by "Thomas Mueller (JIRA)" <ji...@apache.org> on 2014/08/19 16:11:19 UTC
[jira] [Updated] (OAK-2040) Tar compaction: log a warning if there
is more than one checkpoint
[ https://issues.apache.org/jira/browse/OAK-2040?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Thomas Mueller updated OAK-2040:
--------------------------------
Description:
Tar segment compaction usually can't shrink the repository if there are many checkpoints. Checkpoints can be cleaned up, but maybe somebody forgets about this, or possibly checkpoints can accumulate due to a bug.
Before running compaction, the segment node store should check how many checkpoints there are, and if there is more than one, it should log a warning. That way one can more easily analyze problems related to compaction.
In addition to that, maybe compaction should simply not do anything (except for logging warnings) if there are more than 10 checkpoints.
was:
Tar segment compaction usually can't shrink the repository if there are many checkpoints. Checkpoints can be cleaned up, but maybe somebody forgets about this, or possibly checkpoints can accumulate due to a bug.
Before running compaction, the segment node store should check how many checkpoints there are, and if there is more than one, it should log a warning. That way one can more easily analyze problems related to compaction.
> Tar compaction: log a warning if there is more than one checkpoint
> ------------------------------------------------------------------
>
> Key: OAK-2040
> URL: https://issues.apache.org/jira/browse/OAK-2040
> Project: Jackrabbit Oak
> Issue Type: Improvement
> Components: core, segmentmk
> Reporter: Thomas Mueller
> Fix For: 1.0.5
>
>
> Tar segment compaction usually can't shrink the repository if there are many checkpoints. Checkpoints can be cleaned up, but maybe somebody forgets about this, or possibly checkpoints can accumulate due to a bug.
> Before running compaction, the segment node store should check how many checkpoints there are, and if there is more than one, it should log a warning. That way one can more easily analyze problems related to compaction.
> In addition to that, maybe compaction should simply not do anything (except for logging warnings) if there are more than 10 checkpoints.
--
This message was sent by Atlassian JIRA
(v6.2#6252)