[
https://issues.apache.org/jira/browse/COUCHDB-1569?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13481364#comment-13481364
]
Nils Breunese commented on COUCHDB-1569:
----------------------------------------
We are in fact monitoring disk space with dedicated monitoring tools, but
CouchDB apparently required at least two times the current data_size for
compaction and that can be a high number for large databases which could be
compacted just fine with less free diskspace available.
Anyway, we're just suspecting this is the threshold we're hitting in
production, since the message is currently logged at info level and the
instance is only logging at error level.
> Low diskspace warning for compaction should be logged at error level
> --------------------------------------------------------------------
>
> Key: COUCHDB-1569
> URL: https://issues.apache.org/jira/browse/COUCHDB-1569
> Project: CouchDB
> Issue Type: Improvement
> Components: Database Core
> Affects Versions: 1.2
> Reporter: Nils Breunese
> Priority: Trivial
>
> In production we're running CouchDB at log level 'error', because running at
> 'info' is too verbose for production. We suspect automatic compaction might
> not be running due to low diskspace.
> http://wiki.apache.org/couchdb/Compaction#Automatic_Compaction says: "When
> there's not enough free disk space to compact a particular database or view
> index, a warning message is logged." But there is no warning log level in
> CouchDB as far as I know and the can_db_compact function in
> src/couchdb/couch_compaction_daemon.erl seems to log the message about
> diskspace being too low at info level. Could this be changed to error level?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
