[
https://issues.apache.org/jira/browse/PHOENIX-2154?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14704332#comment-14704332
]
James Taylor edited comment on PHOENIX-2154 at 8/20/15 5:50 AM:
----------------------------------------------------------------
Did you try it, Ravi? In theory, the mapper would call the HBase APIs, so when
the reducer runs, all the mapper tasks have been completed. We don't want
Phoenix to send the work to HBase and we definitely don't want auto commit on,
as that would cause an RPC for every row.
was (Author: jamestaylor):
Did you try it, Ravi? In theory, the mapper would call the HBase APIs, so when
the reducer runs, all the mapper tasks have been completed.
> Failure of one mapper should not affect other mappers in MR index build
> -----------------------------------------------------------------------
>
> Key: PHOENIX-2154
> URL: https://issues.apache.org/jira/browse/PHOENIX-2154
> Project: Phoenix
> Issue Type: Bug
> Reporter: James Taylor
> Assignee: maghamravikiran
> Attachments: IndexTool.java, PHOENIX-2154-WIP.patch
>
>
> Once a mapper in the MR index job succeeds, it should not need to be re-done
> in the event of the failure of one of the other mappers. The initial
> population of an index is based on a snapshot in time, so new rows getting
> *after* the index build has started and/or failed do not impact it.
> Also, there's a 1:1 correspondence between index rows and table rows, so
> there's really no need to dedup. However, the index rows will have a
> different row key than the data table, so I'm not sure how the HFiles are
> split. Will they potentially overlap and is this an issue?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
