> On Dec. 16, 2016, 11:39 p.m., Jie Yu wrote:
> > src/slave/containerizer/mesos/isolators/network/cni/cni.cpp, line 1183
> > <https://reviews.apache.org/r/54716/diff/2/?file=1587919#file1587919line1183>
> >
> >     let's use networkConfigPath consistently

The reason I have `cniConfigPath` here is that we use `networkConfigPath` to 
denote the checkpointed CNI configuration (line 1189 in `attach`). This seemed 
consistent with what we were doing in `detach`.


- Avinash


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/54716/#review159510
-----------------------------------------------------------


On Dec. 17, 2016, 8:16 a.m., Avinash sridharan wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/54716/
> -----------------------------------------------------------
> 
> (Updated Dec. 17, 2016, 8:16 a.m.)
> 
> 
> Review request for mesos, Jie Yu and Qian Zhang.
> 
> 
> Bugs: MESOS-6567
>     https://issues.apache.org/jira/browse/MESOS-6567
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Since the `network/cni` isolator will soon be able to
> add/modify/delete CNI configuration files without the need for agent
> restart, we are changing the initialization logic to not bail out if
> we see errors while reading a CNI configuration file or don't find a
> CNI  plugin for a given CNI configuration. If either of these errors
> occur the `network/cni` isolator would simply skip the specific CNI
> network and complete the initialization.
> 
> 
> Diffs
> -----
> 
>   src/slave/containerizer/mesos/isolators/network/cni/cni.hpp 
> b8fc755a8dd4757d904f7e97a71d3cf7f29d2033 
>   src/slave/containerizer/mesos/isolators/network/cni/cni.cpp 
> ddb4d33e9ecd0ffd118e8e68cb7ec90346b08049 
> 
> Diff: https://reviews.apache.org/r/54716/diff/
> 
> 
> Testing
> -------
> 
> sudo make check --gtest_filter=*CNI*
> 
> 
> Thanks,
> 
> Avinash sridharan
> 
>

Reply via email to