mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chun-Hung Hsiao <chhs...@apache.org>
Subject Re: Review Request 70216: Cleanup the recovery logic for refactoring SLRP.
Date Thu, 28 Mar 2019 06:24:02 GMT


> On March 25, 2019, 11:46 a.m., Benjamin Bannier wrote:
> > src/resource_provider/storage/provider.cpp
> > Line 618 (original)
> > <https://reviews.apache.org/r/70216/diff/4/?file=2133677#file2133677line707>
> >
> >     WE are we not checkpointing this anymore since it was always a no-op anyway,
right?

Right. Checkpointing such states are not very useful, as they are meant to keep track of external
interactions (i.e., making calls to the plugin). If another crash happens, the same recovery
logic will put it back to `NODE_READY` again.


> On March 25, 2019, 11:46 a.m., Benjamin Bannier wrote:
> > src/resource_provider/storage/provider.cpp
> > Line 1736 (original), 1553 (patched)
> > <https://reviews.apache.org/r/70216/diff/4/?file=2133677#file2133677line1884>
> >
> >     Should we add a message, here and below?

This will be removed in later patches. Eventually `VolumeManager` would return a failure in
all public-facing functions if `volumeId` is unknown.


> On March 25, 2019, 11:46 a.m., Benjamin Bannier wrote:
> > src/resource_provider/storage/provider.cpp
> > Line 1736 (original), 1553 (patched)
> > <https://reviews.apache.org/r/70216/diff/4/?file=2133677#file2133677line1884>
> >
> >     Should we add a message, here and below?

This will be removed in the next patch.


- Chun-Hung


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70216/#review213977
-----------------------------------------------------------


On March 27, 2019, 5:57 a.m., Chun-Hung Hsiao wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70216/
> -----------------------------------------------------------
> 
> (Updated March 27, 2019, 5:57 a.m.)
> 
> 
> Review request for mesos, Benjamin Bannier, Jie Yu, and Jan Schlicht.
> 
> 
> Bugs: MESOS-9622
>     https://issues.apache.org/jira/browse/MESOS-9622
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> In addition to perform volume state recovery, the `recoverVolumes`
> function also recovers service manager and preparing services now. The
> whole logic will be moved out from SLRP to v0 `VolumeManager` later.
> 
> During volume state recovery, we no longer recover all volumes to steady
> states, since transient states are properly handled in SLRP. To simplify
> the recovery logic, a `publishVolume` method that conforms to the volume
> manager's `publishVolume` is introduced.
> 
> 
> Diffs
> -----
> 
>   src/resource_provider/storage/provider.cpp 2711503cdb58cb9b34af8c9fad0908c5f788a2af

>   src/resource_provider/storage/provider_process.hpp a5536b3d735e01eb1c4dc52d0602d973155f3c93

> 
> 
> Diff: https://reviews.apache.org/r/70216/diff/5/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Chun-Hung Hsiao
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message