hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HBASE-774) Optional flush provokes periodic large spike in cluster traffic
Date Fri, 25 Jul 2008 17:55:31 GMT

    [ https://issues.apache.org/jira/browse/HBASE-774?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12616976#action_12616976

stack commented on HBASE-774:

This cluster has many tables each of many families.   Cycling through flush, at least one
or two of the many familes per table will dump a small file into the filesystem and trigger

> Optional flush provokes periodic large spike in cluster traffic
> ---------------------------------------------------------------
>                 Key: HBASE-774
>                 URL: https://issues.apache.org/jira/browse/HBASE-774
>             Project: Hadoop HBase
>          Issue Type: Bug
>            Reporter: stack
>            Priority: Critical
>         Attachments: network.gif
> Looking at our 70-node cluster, whose current state is mostly read-only but for a service
that is trickling in pages at about 5/600 an hour, network traffic goes through the roof every
30 minutes.  This is 0.1.3.  Regionserver logs are silent but for the optional flush that
runs every 30 minutes.  The optional flush will put a few small files out on the filesystem
tripping compactions of small files.

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

View raw message