[
https://issues.apache.org/jira/browse/PHOENIX-2154?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14705258#comment-14705258
]
Ravi Kishore Valeti commented on PHOENIX-2154:
----------------------------------------------
[~maghamraviki...@gmail.com]
I think the major task for the subject line, "Failure of one mapper should not
affect other mappers in MR index build", is missing from the task list :).
Even a single map task eventual failure (after retires) will mark the job as
"failed" and I think there is no way to resume ONLY a specific task in the
next run (unless we write custom logic?). Next run will be a fresh job
execution.
[~jamestaylor] [~lhofhansl] please correct me if I'm wrong.
> 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)
