mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mesos Reviewbot Windows <revi...@mesos.apache.org>
Subject Re: Review Request 65348: Fixed flakyness of MasterTest.UpdateSlaveMessageWithPendingOffers.
Date Mon, 12 Feb 2018 12:38:09 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/65348/#review197253
-----------------------------------------------------------



PASS: Mesos patch 65348 was successfully built and tested.

Reviews applied: `['65348']`

All the build artifacts available at: http://dcos-win.westus.cloudapp.azure.com/mesos-build/review/65348

- Mesos Reviewbot Windows


On Feb. 12, 2018, 10:56 a.m., Jan Schlicht wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/65348/
> -----------------------------------------------------------
> 
> (Updated Feb. 12, 2018, 10:56 a.m.)
> 
> 
> Review request for mesos, Alexander Rukletsov and Benjamin Bannier.
> 
> 
> Bugs: MESOS-8490
>     https://issues.apache.org/jira/browse/MESOS-8490
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Under rare circumstances, the clock would have to be advanced to trigger
> a new allocation cycle after 'UPDATE_SLAVE' has been handled in the
> master. We also wait for old offers to be rescinded before waiting for
> updated offers as sometimes the old offer could be received erroneously.
> 
> 
> Diffs
> -----
> 
>   src/tests/master_tests.cpp 76c86f8fb10339af263b6831ac1e58bc54f88c55 
> 
> 
> Diff: https://reviews.apache.org/r/65348/diff/3/
> 
> 
> Testing
> -------
> 
> make check
> 
> `src/mesos-tests --gtest_filter=MasterTest.UpdateSlaveMessageWithPendingOffers --gtest_repeat=10000
--gtest_break_on_failure` while running `stress -c 4 -i 8 -m 2 -d 6` on a 4 core machine.
It usually takes a few thousand iterations to trigger the flakyness in the old code. And only
when the host is under heavy load.
> 
> 
> Thanks,
> 
> Jan Schlicht
> 
>


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