> On March 16, 2018, 6:47 p.m., Jie Yu wrote:
> > src/slave/containerizer/mesos/isolators/volume/sandbox_path.cpp
> > Lines 404 (patched)
> > <https://reviews.apache.org/r/66104/diff/1/?file=1978194#file1978194line404>
> >
> >     Let's not have this CHECK here. It's possible that cleanup is called 
> > multiple times (caller assumes that this function is idempotent).

Correction, cleanup won't be called multiple times for a containre, but might 
be called for an unknown container.


- Jie


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/66104/#review199336
-----------------------------------------------------------


On March 15, 2018, 10:31 p.m., Jason Lai wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/66104/
> -----------------------------------------------------------
> 
> (Updated March 15, 2018, 10:31 p.m.)
> 
> 
> Review request for mesos, Eric Chung, Gilbert Song, Jie Yu, and Zhitao Li.
> 
> 
> Bugs: MESOS-8651
>     https://issues.apache.org/jira/browse/MESOS-8651
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> The `volume/sandbox_path` isolator inserts a string of the sandbox path
> to its `sandboxes` hashmap instance variable upon the launch of each
> container. However, it never cleans it up properly and can cause
> unbounded growth of the hashmap object, as isolators are global
> singleton objects.
> 
> The patch ensures the sandbox path associated with a given container ID
> gets removed from the `sandboxes` hashmap upon container cleanup.
> 
> 
> Diffs
> -----
> 
>   src/slave/containerizer/mesos/isolators/volume/sandbox_path.hpp 
> 20d5b32fb7ada1a17a40bf1a2438db4d85cf1063 
>   src/slave/containerizer/mesos/isolators/volume/sandbox_path.cpp 
> 5801977e93bcb8f463a2635f73e763098f2aa97d 
> 
> 
> Diff: https://reviews.apache.org/r/66104/diff/1/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Jason Lai
> 
>

Reply via email to