phoenix-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Geoffrey Jacoby (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (PHOENIX-5020) PhoenixMRJobSubmitter should use a long timeout when getting candidate jobs
Date Tue, 28 May 2019 16:34:00 GMT

    [ https://issues.apache.org/jira/browse/PHOENIX-5020?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16849910#comment-16849910
] 

Geoffrey Jacoby commented on PHOENIX-5020:
------------------------------------------

Not sure why that got moved to Blocker; I created it as Minor. It only applies if you have
a giant System.Catalog, which would be alleviated by the splittable system catalog anyway.

Depending on whether we get to this before PHOENIX-5212, which will deprecate the PhoenixMRJobSubmitter,
this might be Wont Fixed later. 

> PhoenixMRJobSubmitter should use a long timeout when getting candidate jobs
> ---------------------------------------------------------------------------
>
>                 Key: PHOENIX-5020
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-5020
>             Project: Phoenix
>          Issue Type: Improvement
>    Affects Versions: 5.0.0, 4.14.1
>            Reporter: Geoffrey Jacoby
>            Assignee: Geoffrey Jacoby
>            Priority: Blocker
>              Labels: SFDC
>             Fix For: 5.1.0, 4.15.1
>
>
> If an environment has a huge System.Catalog (such as having many views) the query in
getCandidateJobs can timeout. Because of PHOENIX-4936, this looks like there are no indexes
that need an async rebuild. In addition to fixing PHOENIX-4936, we should extend the timeout.




--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message