[
https://issues.apache.org/jira/browse/MESOS-9014?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16518661#comment-16518661
]
Meng Zhu commented on MESOS-9014:
---------------------------------
After discussing with [~greggomann], looks like the segfault happens due to a
race during the test tear-down.
The test looks for a particular task status update. A segfault would happen
when the scheduler receives another task status update and tries to acknowledge
the update, and in between, the scheduler got tear-down.
A possible fix is to kill the task and wait for the task kill status update
explicitly before test tear-down.
> MasterAPITest.SubscribersReceiveHealthUpdates is flaky
> ------------------------------------------------------
>
> Key: MESOS-9014
> URL: https://issues.apache.org/jira/browse/MESOS-9014
> Project: Mesos
> Issue Type: Bug
> Components: test
> Reporter: Meng Zhu
> Priority: Major
> Labels: flaky-test
> Attachments: SubscribersReceiveHealthUpdates_BadRun.txt
>
>
> This test fails flaky on CI. Log attached.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
