You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@hbase.apache.org by "Enis Soztutar (JIRA)" <ji...@apache.org> on 2015/04/20 23:57:59 UTC

[jira] [Created] (HBASE-13515) Handle FileNotFoundException in region replica replay for flush/compaction events

Enis Soztutar created HBASE-13515:
-------------------------------------

             Summary: Handle FileNotFoundException in region replica replay for flush/compaction events
                 Key: HBASE-13515
                 URL: https://issues.apache.org/jira/browse/HBASE-13515
             Project: HBase
          Issue Type: Sub-task
            Reporter: Enis Soztutar
            Assignee: Enis Soztutar
             Fix For: 2.0.0, 1.1.0


I had this patch laying around that somehow dropped from my plate. We should skip replaying compaction / flush and region open event markers if the files (from flush or compaction) can no longer be found from the secondary. If we do not skip, the replay will be retried forever, effectively blocking the replication further. 

Bulk load already does this, we just need to do it for flush / compaction and region open events as well. 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)