mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vinod Kone <vinodk...@gmail.com>
Subject Re: Review Request 50253: Added native implementation for the V0 Mesos Adapter.
Date Tue, 26 Jul 2016 23:44:55 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/50253/#review143635
-----------------------------------------------------------




src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 64)
<https://reviews.apache.org/r/50253/#comment209464>

    make it explicit that this is v0::Scheduler and v1::MesosBase.



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 76)
<https://reviews.apache.org/r/50253/#comment209465>

    // v0 scheduler interface methods.



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 119)
<https://reviews.apache.org/r/50253/#comment209466>

    // v1 MesosBase methods.



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 172)
<https://reviews.apache.org/r/50253/#comment209470>

    move these out of the class below public method definitions.



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 175)
<https://reviews.apache.org/r/50253/#comment209471>

    s/executor/scheduler/



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 430)
<https://reviews.apache.org/r/50253/#comment209473>

    add a comment to see send() for more details.



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 463)
<https://reviews.apache.org/r/50253/#comment209474>

    set `subscribeCall` to false?
    
    more importantly we need to drain the queue, with a big comment on why we do it and why
it is safe.



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 600)
<https://reviews.apache.org/r/50253/#comment209479>

    instead of CHECK here and everywhere else in this functio, do a validation and drop instead?



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 613)
<https://reviews.apache.org/r/50253/#comment209477>

    i think you can always call this?



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 623)
<https://reviews.apache.org/r/50253/#comment209481>

    log a warning/error?



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 632)
<https://reviews.apache.org/r/50253/#comment209480>

    again, is this necessary?



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 643)
<https://reviews.apache.org/r/50253/#comment209482>

    log warning/error?



src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp (line 660)
<https://reviews.apache.org/r/50253/#comment209483>

    log a warning/error.


- Vinod Kone


On July 20, 2016, 7:02 p.m., Anand Mazumdar wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/50253/
> -----------------------------------------------------------
> 
> (Updated July 20, 2016, 7:02 p.m.)
> 
> 
> Review request for mesos, Joris Van Remoortere and Vinod Kone.
> 
> 
> Bugs: MESOS-5788
>     https://issues.apache.org/jira/browse/MESOS-5788
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> This change adds the C++ implementation for the JAVA V0 to V1 Mesos
> implementation adapter (driver + scheduler).
> 
> 
> Diffs
> -----
> 
>   src/Makefile.am cc83525a4455bbb0e654d346921d66ed2436411d 
>   src/java/jni/org_apache_mesos_v1_scheduler_V0Mesos.cpp PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/50253/diff/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Anand Mazumdar
> 
>


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