mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jie Yu" <yujie....@gmail.com>
Subject Re: Review Request 42096: Fixed race in persistent volume tests.
Date Mon, 11 Jan 2016 22:40:34 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/42096/#review113864
-----------------------------------------------------------

Ship it!


Ship It!

- Jie Yu


On Jan. 9, 2016, 1:19 a.m., Greg Mann wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/42096/
> -----------------------------------------------------------
> 
> (Updated Jan. 9, 2016, 1:19 a.m.)
> 
> 
> Review request for mesos and Jie Yu.
> 
> 
> Bugs: MESOS-4318
>     https://issues.apache.org/jira/browse/MESOS-4318
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Fixed race in persistent volume tests.
> 
> 
> Diffs
> -----
> 
>   src/tests/persistent_volume_tests.cpp 0a9a2bbd3b3a9c1bacf361750fae7a1970a622b7 
> 
> Diff: https://reviews.apache.org/r/42096/diff/
> 
> 
> Testing
> -------
> 
> I was able to reproduce this failure by doing
> `MESOS_VERBOSE=1 GTEST_FILTER="PersistentVolumeTest.BadACLNoPrincipal:PersistentVolumeTest.BadACLDropCreateAndDestroy"
bin/mesos-tests.sh --gtest_repeat=1000 --gtest_break_on_failure=1`
> on Ubuntu 14.04. Unfortunately, the tests in question were relying on the `suppressOffers`
and `reviveOffers` calls being reliably completed in a specified order, which is not the case.
Inserting `Clock::settle()` after the appropriate `suppressOffers` calls allows those messages
to be processed before `driver1` calls `reviveOffers`, ensuring that `driver1` can receive
the resulting offers.
> 
> After applying the patch, I ran the above command on both Ubuntu 14.04 and OSX and it
was successful.
> 
> 
> Thanks,
> 
> Greg Mann
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message