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 57469: Made the default executor populate volume mappings for disk resources.
Date Thu, 16 Mar 2017 16:29:41 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/57469/#review169170
-----------------------------------------------------------


Fix it, then Ship it!





src/tests/default_executor_tests.cpp
Lines 1631 (patched)
<https://reviews.apache.org/r/57469/#comment241488>

    I would test the persistent volume functionality. It's possbile that the directory exists
but does not link to the persistent volume. Take a look at PersistentVolumeTest.AccessPersistentVolume.


- Jie Yu


On March 11, 2017, 11:32 p.m., Anand Mazumdar wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/57469/
> -----------------------------------------------------------
> 
> (Updated March 11, 2017, 11:32 p.m.)
> 
> 
> Review request for mesos, Gilbert Song, Jie Yu, and Vinod Kone.
> 
> 
> Bugs: MESOS-7225
>     https://issues.apache.org/jira/browse/MESOS-7225
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> A task can now access any volumes specified in disk resources from
> its own sandbox owing to the default executor populating the mapping
> from the child container sandbox to the parent. Previously, without
> such a mapping it was not possible for the child container to access
> those volumes mounted on the parent container.
> 
> 
> Diffs
> -----
> 
>   src/launcher/default_executor.cpp 0ed436faa68e984d0be4e5186138f738bc7f1b52 
>   src/tests/default_executor_tests.cpp a09ae11f5fcad949fd84e4288f46675eba2e1ea9 
> 
> 
> Diff: https://reviews.apache.org/r/57469/diff/1/
> 
> 
> Testing
> -------
> 
> make check
> 
> The added test uses the new v1 helpers and ideally should not be committed till we do
a sweep of this file to update all the tests. That being said the file is already inconsistent
though.
> 
> 
> Thanks,
> 
> Anand Mazumdar
> 
>


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