[ https://issues.apache.org/jira/browse/CASSANDRA-2666?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13504001#comment-13504001 ]
Idcmp commented on CASSANDRA-2666: ---------------------------------- Is there a different way to perform the required periodic maintenance that "nodetool repair" gives but with a way of knowing if it's successful or not? I understand I can have something parse log files, but that feels quite suboptimal. > nodetool repair silently fails (no non-zero exit code) when repair did not > happen > --------------------------------------------------------------------------------- > > Key: CASSANDRA-2666 > URL: https://issues.apache.org/jira/browse/CASSANDRA-2666 > Project: Cassandra > Issue Type: Bug > Components: Tools > Reporter: Peter Schuller > Assignee: Vladimir Sverzhinsky > Priority: Minor > Labels: lhf > > In this particular example I tested with a neighbor down. system.log > correctly reports: > INFO [manual-repair-65fb7826-77ae-4ad6-99e3-5aa94b03b97d] 2011-05-19 > 14:48:53,476 AntiEntropyService.java (line 767) Could not proceed on repair > because a neighbor (/XX.XX.XX.XX) is dead: > manual-repair-65fb7826-77ae-4ad6-99e3-5aa94b03b97d failed. > But "nodetool -h localhost repair || echo fail" doesn't print fail. -- 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