mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Benjamin Bannier <bbann...@apache.org>
Subject Re: Review Request 58048: Added 'id' and 'metadata' fields to 'Resource.DiskInfo.Source'.
Date Fri, 15 Sep 2017 13:16:56 GMT


> On Sept. 13, 2017, 2:34 a.m., Jie Yu wrote:
> > src/common/resources.cpp
> > Lines 185 (patched)
> > <https://reviews.apache.org/r/58048/diff/5/?file=1820113#file1820113line185>
> >
> >     hum, i think we also need to check? wondering if we need to fix the rest.
> >     ```
> >     if (left.has_id() != right.has_id()) {
> >       return false;
> >     }
> >     ```

Yes, this was slightly broken. I pushed a fix in https://reviews.apache.org/r/62282/.


> On Sept. 13, 2017, 2:34 a.m., Jie Yu wrote:
> > src/common/resources.cpp
> > Lines 372 (patched)
> > <https://reviews.apache.org/r/58048/diff/5/?file=1820113#file1820113line374>
> >
> >     I think we should treat BLOCK the same as MOUNT.

Yes, per our offline discussion this makes sense to me now.


> On Sept. 13, 2017, 2:34 a.m., Jie Yu wrote:
> > src/v1/resources.cpp
> > Lines 353-358 (original), 386-391 (patched)
> > <https://reviews.apache.org/r/58048/diff/5/?file=1820115#file1820115line386>
> >
> >     Do you still need this?

Removed.


> On Sept. 13, 2017, 2:34 a.m., Jie Yu wrote:
> > src/v1/resources.cpp
> > Lines 491-493 (patched)
> > <https://reviews.apache.org/r/58048/diff/5/?file=1820115#file1820115line491>
> >
> >     I think for RAW and if id are equal, we should return true (same as we did for
MOUNT)

Yes, this was missing. I updated the test to confirm self-containment (`assert(r.contains(r))`).
I also fixed the unintentional fall-through in the handling of `MOUNT` just above.


- Benjamin


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/58048/#review185248
-----------------------------------------------------------


On Sept. 13, 2017, 5:23 p.m., Benjamin Bannier wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/58048/
> -----------------------------------------------------------
> 
> (Updated Sept. 13, 2017, 5:23 p.m.)
> 
> 
> Review request for mesos, Jie Yu and Jan Schlicht.
> 
> 
> Bugs: MESOS-7312
>     https://issues.apache.org/jira/browse/MESOS-7312
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Ids will allow to create distinguishable resources, e.g., of RAW or
> BLOCK type. We also add a metadata field which can be used to expose
> additional disk information.
> 
> 
> Diffs
> -----
> 
>   include/mesos/mesos.proto fa475dcc3aa0b15ee4b2b740a24ab3c0b9a3274c 
>   include/mesos/v1/mesos.proto dfba4181203e23f7eedf67c19379d031e0993fd5 
>   src/common/resources.cpp 14b600ca1577be4910164396c75b866b53439ade 
>   src/tests/resources_tests.cpp 8a86efe60024d9e36294f0acb3f43bdd3f52f5bc 
>   src/v1/resources.cpp a5cc15591b7611274148139c43465adae44c1dbb 
> 
> 
> Diff: https://reviews.apache.org/r/58048/diff/6/
> 
> 
> Testing
> -------
> 
> make check (OS X, Fedora25)
> 
> 
> Thanks,
> 
> Benjamin Bannier
> 
>


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