> On Feb. 21, 2017, 1:50 p.m., Vinod Kone wrote:
> > include/mesos/mesos.proto, line 1581
> > <https://reviews.apache.org/r/56830/diff/1/?file=1640428#file1640428line1581>
> >
> >     also for unknown status (e.g., OOM)

Since Mesos never defined tasks as processes only, It's whatever the executor 
decides, I think it's fine to not define it directly here but rather refer to 
`TASK_FAILED`. i.e., however TASK_FAILED is defined.

Would this be clear:

```
// The task group would be only killed when the task failed,
// i.e., it will be transitioned to TASK_FAILED state.
```


- Jiang Yan


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/56830/#review166233
-----------------------------------------------------------


On Feb. 19, 2017, 9 p.m., Anand Mazumdar wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/56830/
> -----------------------------------------------------------
> 
> (Updated Feb. 19, 2017, 9 p.m.)
> 
> 
> Review request for mesos, Vinod Kone and Jiang Yan Xu.
> 
> 
> Bugs: MESOS-7067
>     https://issues.apache.org/jira/browse/MESOS-7067
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Describes the on termination policy associated with a task and is
> applied upon a task termination.
> 
> 
> Diffs
> -----
> 
>   include/mesos/mesos.proto 030e79c003f6560e9c0627db12fb1baba411151d 
>   include/mesos/v1/mesos.proto 7f6f858a7d9387d202510730d400e490298e6574 
> 
> Diff: https://reviews.apache.org/r/56830/diff/
> 
> 
> Testing
> -------
> 
> make check (tests later in the chain)
> 
> 
> Thanks,
> 
> Anand Mazumdar
> 
>

Reply via email to