> On Aug. 4, 2017, 12:29 a.m., Jie Yu wrote:
> > why we need this flag? I think we need a way for operator to prevent top 
> > level containers on a host to share with agent pid namespace.
> 
> Gilbert Song wrote:
>     @Qian, I guess you are trying to simplify the agent flag name, but I 
> think we may want to emphasize that:
>     1. this is for top level container only
>     2. it is operator specific
>     
>     Do you have anyother concern for the flag name mentioned in the JIRA?

@Jie and @Gilbert, I know the original proposed flag name is 
`--disallow_top_level_pid_ns_sharing`, however, I went through all the agent 
flags, it seems there is no one named with `--disallow` or `--allow` as the 
prefix, that's why I named it `--share_pid_namespace`. However, I agree the 
original name is more explicit and clear, so I will change it back later.


- Qian


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/61406/#review182120
-----------------------------------------------------------


On Aug. 3, 2017, 11:20 p.m., Qian Zhang wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/61406/
> -----------------------------------------------------------
> 
> (Updated Aug. 3, 2017, 11:20 p.m.)
> 
> 
> Review request for mesos, Gastón Kleiman, Gilbert Song, Jie Yu, Kevin Klues, 
> and Vinod Kone.
> 
> 
> Bugs: MESOS-7853
>     https://issues.apache.org/jira/browse/MESOS-7853
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Introduced `--share_pid_namespace` agent flag.
> 
> 
> Diffs
> -----
> 
>   docs/configuration.md 041c3dfb9c0c1718770f74dfb33a9f5d6fbe9b61 
>   src/slave/flags.hpp 032880dfa68cd29420e559d34e592e57827cfc07 
>   src/slave/flags.cpp 4171604090ffebe79fa35579458cabff7270c0de 
> 
> 
> Diff: https://reviews.apache.org/r/61406/diff/1/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Qian Zhang
> 
>

Reply via email to