-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/63022/#review192864
-----------------------------------------------------------




src/resource_provider/storage/provider.cpp
Lines 1203 (patched)
<https://reviews.apache.org/r/63022/#comment271190>

    some plugins might not implement create/delete, but DO implement list. such 
a plugin might only support "block"- (or only support "mount"-) type volumes. 
in such cases, it could be useful to report those resources as BLOCK (or MOUNT) 
volumes instead of RAW. I don't *think* it's a blocker for MVP, but we've had a 
couple of conversations recently about exposing low-level primitives to 
specialized storage frameworks, and that it might be useful to report block 
devices as BLOCK disk resources, by default.


- James DeFelice


On Dec. 5, 2017, 5:57 a.m., Chun-Hung Hsiao wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/63022/
> -----------------------------------------------------------
> 
> (Updated Dec. 5, 2017, 5:57 a.m.)
> 
> 
> Review request for mesos, James DeFelice, Jie Yu, Joseph Wu, and Jan Schlicht.
> 
> 
> Bugs: MESOS-8101
>     https://issues.apache.org/jira/browse/MESOS-8101
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> The following lists the steps to reconcile resources:
> 
> 1. Import resources from the CSI plugin:
>   a. Get preprovisioned volumes:
>        ListVolumes
>        ValidateVolumeCapabilities for each volume
>   b. GetCapacity for each profile to get RAW resources
> 2. For each resource in the checkpointed resources:
>   a. Strip away metadata (reservation, persistence, etc)
>   b. Additional resources after subtracting the stripped resources
>      from the imported resources are new resources.
> 3. Report the reconciled resources through UPDATE_STATE
> 
> 
> Diffs
> -----
> 
>   src/resource_provider/storage/provider.cpp 
> d35b0d02992e3730ca47906b34c21e1ba9c653e7 
> 
> 
> Diff: https://reviews.apache.org/r/63022/diff/16/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Chun-Hung Hsiao
> 
>

Reply via email to