mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jiang Yan Xu <...@jxu.me>
Subject Re: Review Request 53096: Fix handling in shared count in total resources in the sorter.
Date Thu, 26 Jan 2017 00:58:54 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/53096/#review162890
-----------------------------------------------------------


Fix it, then Ship it!




This review is now closely related to the followup review /r/55359/ which almost makes sense
to combine them. But because this patch fixes a bug (with tests to prove that) I guess we'll
still keep them separate. Just thought I'd mention it for posterity.


src/master/allocator/mesos/hierarchical.hpp (line 479)
<https://reviews.apache.org/r/53096/#comment234272>

    As mentioned in the reply from the previous conversation:
    
    s/updateTotal/updateSlaveTotal/
    
    And the comments:
    
    ```
    // Helper to update the agent's total resources maintained in the allocator and the role
    // and quota sorters (whose total resources match the agent's total resources).
    ```



src/master/allocator/mesos/hierarchical.cpp (line 720)
<https://reviews.apache.org/r/53096/#comment234480>

    s/total/agent's total/



src/master/allocator/mesos/hierarchical.cpp (line 754)
<https://reviews.apache.org/r/53096/#comment234481>

    s/allocated resources/allocation/ 
    
    It's synonymous but `allocation` is more consistent with the comments above (now that
we don't need to say "total and allocated resources").
    
    Same applies to another occurrence below.



src/master/allocator/mesos/hierarchical.cpp (line 1977)
<https://reviews.apache.org/r/53096/#comment234492>

    Can we also add the following sentense before "So, we update them using the resources
in `slaves[slaveId].total`." to explain it more thoroughly:
    
    ```
    (which don't get updated in the allocation runs or during recovery of allocated resources)
    ```



src/tests/persistent_volume_tests.cpp (lines 1443 - 1447)
<https://reviews.apache.org/r/53096/#comment234495>

    "We kill the long-lived task": this duplicates the same "We kill the long-lived task"
below. I guess you mentioned them in two places because they are far apart but can we move

    
    ```
      Future<TaskStatus> status4;
      EXPECT_CALL(sched, statusUpdate(&driver, _))
        .WillOnce(FutureArg<1>(&status4));
    ```
    
    down to right before we sent killTask so we just need to comment on it once?
    
    Plus, the block of code below it right now is all about killing the `nonSharedVolume`
which is not relevant.


- Jiang Yan Xu


On Jan. 9, 2017, 2:37 p.m., Anindya Sinha wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/53096/
> -----------------------------------------------------------
> 
> (Updated Jan. 9, 2017, 2:37 p.m.)
> 
> 
> Review request for mesos and Jiang Yan Xu.
> 
> 
> Bugs: MESOS-6444
>     https://issues.apache.org/jira/browse/MESOS-6444
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> We maintain a single copy of shared resource in the role and quota
> sorter's total resources. So, when we update these resources, we
> remove the previous resources at this agent and add the new resources
> at this agent.
> 
> 
> Diffs
> -----
> 
>   src/master/allocator/mesos/hierarchical.hpp a6424d624864155e1c87a28a63b784512c5c8722

>   src/master/allocator/mesos/hierarchical.cpp 91b1ec43940a788459f045ca4a4b82d4e8373bca

>   src/tests/persistent_volume_tests.cpp 8198b6b5ad323d17835ba067c7ff3d34ef948125 
> 
> Diff: https://reviews.apache.org/r/53096/diff/
> 
> 
> Testing
> -------
> 
> Tests passed.
> 
> 
> Thanks,
> 
> Anindya Sinha
> 
>


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