[ https://issues.apache.org/jira/browse/HBASE-21415?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Josh Elser updated HBASE-21415: ------------------------------- Resolution: Fixed Hadoop Flags: Reviewed Status: Resolved (was: Patch Available) Committed this to master (and only master). My attempt to cherry pick this to branch-2 showed multiple areas where the master patch was insufficient. I've no objection to getting this backported to other branches. I've just spent enough time caressing this patch to make it acceptable and correct :) Thanks for your initial work, Philippe! > Admin#snapshot method documentation is misleading. > -------------------------------------------------- > > Key: HBASE-21415 > URL: https://issues.apache.org/jira/browse/HBASE-21415 > Project: HBase > Issue Type: Improvement > Components: documentation > Reporter: Philippe Laflamme > Assignee: Philippe Laflamme > Priority: Trivial > Fix For: 3.0.0 > > Attachments: HBASE-21415.002.patch, HBASE-21415.003.patch, > HBASE-21415.patch, HBASE-21415.patch > > > The documentation for the {{Admin#snapshot}} function is misleading in > regards to snapshot concurrency. > It currently states the following: > {quote}Only a single snapshot should be taken at a time for an instance of > HBase, or results may be undefined (you can tell multiple HBase clusters to > snapshot at the same time, but only one at a time for a single cluster). > {quote} > This seems to indicate that it's dangerous to run snapshots concurrently when > in fact the HBase master will simply run them sequentially (as per this Slack > thread [https://apache-hbase.slack.com/archives/C13K8NVAM/p1540994948005900)] > The suggested fix is to simply remove this entire sentence. -- This message was sent by Atlassian JIRA (v7.6.3#76005)