[
https://issues.apache.org/jira/browse/PHOENIX-3950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16637546#comment-16637546
]
Vincent Poon edited comment on PHOENIX-3950 at 10/3/18 9:54 PM:
----------------------------------------------------------------
I think what is happening here is the mapreduce job fails, but then while the
index is still in 'b', an index write fails and the index gets disabled. The
rebuilder thread in MetadataRegionObserver then kicks in and rebuilds from the
disabletimestamp to the current time, and marks the index back to 'a' (even
though many of the index entries from before the disabletimestamp may not have
been populated).
FYI [~gjacoby][~kozdemir][~tdsilva]
was (Author: vincentpoon):
I think what is happening here is the mapreduce job fails, but then while the
index is still in 'b', an index write fails and the index gets disabled. The
rebuilder thread in MetadataRegionObserver then kicks in and rebuilds from the
disabletimestamp to the current time, and marks the index back to 'a' (even
though many of the index entries from before the disabletimestamp may not have
been populated).
> IndexTool marks index active even when index building fails
> -----------------------------------------------------------
>
> Key: PHOENIX-3950
> URL: https://issues.apache.org/jira/browse/PHOENIX-3950
> Project: Phoenix
> Issue Type: Bug
> Reporter: James Taylor
> Priority: Major
>
> We've seen cases in which the MR-based index building tool (IndexTool) marks
> an index as active when the index building clearly did not succeed. It might
> be related to a timeout occurring and then not being propagated to the client.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
