-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/47511/#review135526
-----------------------------------------------------------




docs/docker-volume-isolator.md (lines 8 - 11)
<https://reviews.apache.org/r/47511/#comment200524>

    I would rephase it a little bit.
    
    The docker volume isolator is introduced in Mesos 0.29.0. This isolator 
adds external volume support for mesos containers via docker volume driver API.



docs/docker-volume-isolator.md (line 13)
<https://reviews.apache.org/r/47511/#comment200531>

    the movivation of docker volume support,



docs/docker-volume-isolator.md (lines 30 - 33)
<https://reviews.apache.org/r/47511/#comment200534>

    I would prefer expend what is attractive for this feature:
    
    Currently, the persistent volume in mesos provides stateful services to 
store their data locally in mesos. However, the local disk is size limited. The 
external storage is needed for applications (e.g., database) to achive a more 
reliable/ambitious goal. In addition, data migration between agents/containers 
is possible by plugable external volume support.



docs/docker-volume-isolator.md (lines 35 - 38)
<https://reviews.apache.org/r/47511/#comment200535>

    could we simplify this to a shorter sentence and merge to above paragraph?



docs/docker-volume-isolator.md (lines 40 - 43)
<https://reviews.apache.org/r/47511/#comment200536>

    The docker/volume isolator manipulate docker volume driver API by dvdcli, 
to intergrate with external volumes.



docs/docker-volume-isolator.md (lines 45 - 55)
<https://reviews.apache.org/r/47511/#comment200537>

    Or it seems to me that you should merge this to the 'without' paragraph.



docs/docker-volume-isolator.md (lines 57 - 59)
<https://reviews.apache.org/r/47511/#comment200538>

    Simplify a little bit:
    
    Thus, the docker volume support provides Mesos...



docs/docker-volume-isolator.md (lines 77 - 79)
<https://reviews.apache.org/r/47511/#comment200539>

    This is not 100% true, right?



docs/docker-volume-isolator.md (lines 77 - 86)
<https://reviews.apache.org/r/47511/#comment200540>

    These three may not necessary to be TaskInfo. They can also be the case for 
custom executor.



docs/docker-volume-isolator.md (lines 114 - 116)
<https://reviews.apache.org/r/47511/#comment200541>

    Could you rephase this sentence? Thanks.



docs/docker-volume-isolator.md (lines 130 - 132)
<https://reviews.apache.org/r/47511/#comment200544>

    The `docker/volume` isolator must be specified in the `--isolation` at agent
    startup, and the `docker/volume` isolator has a dependency on 
`filesystem/linux` isolator.



docs/docker-volume-isolator.md (line 181)
<https://reviews.apache.org/r/47511/#comment200548>

    s/one/operators/g



docs/docker-volume-isolator.md (line 184)
<https://reviews.apache.org/r/47511/#comment200549>

    where the volume comes from. Operator need to ...



docs/docker-volume-isolator.md (line 189)
<https://reviews.apache.org/r/47511/#comment200547>

    How to specify `container_path`:


- Gilbert Song


On May 29, 2016, 4:22 p.m., Guangya Liu wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/47511/
> -----------------------------------------------------------
> 
> (Updated May 29, 2016, 4:22 p.m.)
> 
> 
> Review request for mesos, Gilbert Song and Jie Yu.
> 
> 
> Bugs: MESOS-5216
>     https://issues.apache.org/jira/browse/MESOS-5216
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Added documentation for `docker/volume` isolator.
> 
> 
> Diffs
> -----
> 
>   docs/docker-volume-isolator.md PRE-CREATION 
>   docs/images/docker-volume-isolator.png PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/47511/diff/
> 
> 
> Testing
> -------
> 
> You can review the document here: 
> https://github.com/jay-lau/mesos/blob/master/docs/docker-volume-isolator.md
> 
> 
> Thanks,
> 
> Guangya Liu
> 
>

Reply via email to