Csaba Ringhofer has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/10745 )

Change subject: IMPALA-7175: deflake check for failed impalad
......................................................................


Patch Set 3: Code-Review+1

Sorry, I gave it a bit more thought, and I would actually prefer someone else 
to look at this too. The code is absolutely ok for me, but I am unsure about 
the approach. +1 running impalad that shouldn't be there sounds like a real 
problem to me, for example it may lead to exceeding the memory limits of the 
test environment.

My first bet on the way to avoid this situation would be to make 
impala_cluster.py/kill() synchronous somehow. This is generally not easy, but 
it should be possible here, as no one else should start/stop impalad processes 
during the tests, so kill() could wait for the process with the pid to 
disappear or be replaced with a different process name.


--
To view, visit http://gerrit.cloudera.org:8080/10745
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I97c6b398e43c6abb1df2b1783c26159137f14fa4
Gerrit-Change-Number: 10745
Gerrit-PatchSet: 3
Gerrit-Owner: Vuk Ercegovac <[email protected]>
Gerrit-Reviewer: Csaba Ringhofer <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Vuk Ercegovac <[email protected]>
Gerrit-Comment-Date: Thu, 21 Jun 2018 15:25:38 +0000
Gerrit-HasComments: No

Reply via email to