[
https://issues.apache.org/jira/browse/ACCUMULO-2711?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13976857#comment-13976857
]
Mike Drob commented on ACCUMULO-2711:
-------------------------------------
Looking at logs of other tservers on the cluster, here is my reconstructed
timeline:
# Some tserver dies, due to agitation.
# Its WAL is shared amongst the masses as recovery occurs.
# Original tserver comes back online, tries to delete the WAL, but it is long
gone.
> Tserver failed to delete WAL
> ----------------------------
>
> Key: ACCUMULO-2711
> URL: https://issues.apache.org/jira/browse/ACCUMULO-2711
> Project: Accumulo
> Issue Type: Bug
> Components: tserver
> Affects Versions: 1.6.0
> Environment: 1.6.0-RC3, cdh4.5.0, physical cluster
> Reporter: Mike Drob
>
> Running CI with agitation, saw the following in one of the tserver logs:
> {noformat}
> 2014-04-22 07:02:02,288 [tserver.TabletServer] INFO : Deleting walog
> hdfs://nameservice1/accumulo/wal/tserver1.example.com+10011/fe91611c-8e54-4527-87d2-9003a07c5566
> 2014-04-22 07:02:02,389 [tserver.TabletServer] WARN : Error attempting to
> delete write-ahead log
> hdfs://nameservice1/accumulo/wal/tserver1.example.com+10011/fe91611c-8e54-4527-87d2-9003a07c5566:
> java.io.IOException: Failed to move to trash:
> hdfs://nameservice1/accumulo/recovery/fe91611c-8e54-4527-87d2-9003a07c5566
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.2#6252)
