You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@hbase.apache.org by "stack (JIRA)" <ji...@apache.org> on 2018/02/20 20:14:00 UTC

[jira] [Resolved] (HBASE-20029) @Ignore TestQuotaThrottle and TestReplicasClient#testCancelOfMultiGet

     [ https://issues.apache.org/jira/browse/HBASE-20029?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack resolved HBASE-20029.
---------------------------
       Resolution: Fixed
         Assignee: stack
    Fix Version/s: 2.0.0-beta-2

Pushed to master and branch-2. HBASE-20030 is about fixing and reenabling these flakies.

> @Ignore TestQuotaThrottle and TestReplicasClient#testCancelOfMultiGet
> ---------------------------------------------------------------------
>
>                 Key: HBASE-20029
>                 URL: https://issues.apache.org/jira/browse/HBASE-20029
>             Project: HBase
>          Issue Type: Sub-task
>            Reporter: stack
>            Assignee: stack
>            Priority: Major
>             Fix For: 2.0.0-beta-2
>
>         Attachments: HBASE-20029.branch-2.001.patch
>
>
> We are down to the last few flakies on branch-2. These two fail mostly on a constrained GCE machine (4cpus, 15G). The TestQuotaThrottling fails about 30% of the time. It needs someone to dive in and add counting of RPCs and study the difference when it fails. The TestReplicas#testCancleOfMultiGet fails occasionally on Apache and about %17 of the time on GCE machine. 
> I'm disabling them because I'm not going to try and figure them. If someone wants to take up the cause that'd be great but meantime I don't want them failing our nightlies.



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