[ https://issues.apache.org/jira/browse/HBASE-17928?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15984336#comment-15984336 ]
Guangxu Cheng commented on HBASE-17928: --------------------------------------- bq.Is this any good if currently we have a long-running compaction currently in operation hogging all i/o with no end in sight? [~stack] Maybe the scene is different.We just need to clear the compaction queues in our production cluster. In our production cluster, we will run major compaction on a table at 1:00 am every day(1:00 to 5:00 is off-peak times) If the compaction has not done before 5:00 am, we will clear compaction queues regardless of the running compactions. Cancel long-running compactions is also a useful feature. Maybe we can create a new issue to implement the functions as I listed earlier. I'd like to work on that. > Shell tool to clear compact queues > ---------------------------------- > > Key: HBASE-17928 > URL: https://issues.apache.org/jira/browse/HBASE-17928 > Project: HBase > Issue Type: New Feature > Reporter: Guangxu Cheng > Assignee: Guangxu Cheng > Fix For: 2.0.0 > > Attachments: HBASE-17928-branch-1-v1.patch, HBASE-17928-v1.patch, > HBASE-17928-v2.patch, HBASE-17928-v3.patch, HBASE-17928-v4.patch > > > scenarioļ¼ > 1. Compact a table by mistake > 2. Compact is not completed within the specified time period > In this case, clearing the queue is a better choice, so as not to affect the > stability of the cluster -- This message was sent by Atlassian JIRA (v6.3.15#6346)