mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Mann <g...@mesosphere.io>
Subject Re: Review Request 70936: Adjusted task status updates during draining.
Date Tue, 25 Jun 2019 21:48:54 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70936/#review216143
-----------------------------------------------------------




src/slave/slave.cpp
Lines 5701 (patched)
<https://reviews.apache.org/r/70936/#comment303161>

    s/KILLEd/KILLED/



src/slave/slave.cpp
Lines 5702 (patched)
<https://reviews.apache.org/r/70936/#comment303164>

    s/drainInfo/drainConfig/
    
    Here and elsewhere.



src/slave/slave.cpp
Lines 5721 (patched)
<https://reviews.apache.org/r/70936/#comment303165>

    s/a/the/



src/tests/slave_tests.cpp
Lines 12109 (patched)
<https://reviews.apache.org/r/70936/#comment303166>

    Maybe we should unconditionally set the reason to REASON_AGENT_DRAINING? It doesn't make
sense to me that the reason received by the scheduler will depend on what stage the task launch
was in when the drain message was received?



src/tests/slave_tests.cpp
Line 12251 (original), 12263 (patched)
<https://reviews.apache.org/r/70936/#comment303167>

    Is the framework ID necessary? Would a `Future<Nothing>` suffice here?


- Greg Mann


On June 24, 2019, 2:03 p.m., Benjamin Bannier wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70936/
> -----------------------------------------------------------
> 
> (Updated June 24, 2019, 2:03 p.m.)
> 
> 
> Review request for mesos, Greg Mann and Joseph Wu.
> 
> 
> Bugs: MESOS-9823
>     https://issues.apache.org/jira/browse/MESOS-9823
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> When a task is reported as killed to the agent during active agent
> draining we now decorate the reported status with
> `REASON_AGENT_DRAINING` if no other status was previously present. If
> the draining marks the agent as gone via the `mark_gone` draining flag
> we additionally report `TASK_GONE_BY_OPERATOR` instead of the original
> state.
> 
> This patch leaves some ambiguity in what triggered the kill since the
> agent-executor protocol does not transport reasons; instead
> the reason is here only inferred after the killed task has
> been observed. This should usually be fine since due to the inherit race
> between e.g., any user- and drain-triggered kill a user cannot
> distinguish racy reasons.
> 
> 
> Diffs
> -----
> 
>   src/slave/slave.cpp 30039b0857a4d85b4b96fa95d7f8724d57cdec6e 
>   src/tests/slave_tests.cpp c2035976713abb31b3646c0d23771fa40df93271 
> 
> 
> Diff: https://reviews.apache.org/r/70936/diff/1/
> 
> 
> Testing
> -------
> 
> `make check`
> 
> 
> Thanks,
> 
> Benjamin Bannier
> 
>


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