-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/21667/#review45980
-----------------------------------------------------------

Ship it!


Ok looks good, let's add a comment so others know why we're using a UUID here.

I've included a suggestion below with some references to Erlang.


src/sched/sched.cpp
<https://reviews.apache.org/r/21667/#comment81146>

    How about a comment here:
    
    // We use a UUID here to ensure that the master can reliably
    // distinguish scheduler runs. Otherwise the master may receive
    // a delayed ExitedEvent and deactivate the framework incorrectly.
    // TODO: Investigate better ways to solve this problem, do
    // bidirectional links (ala Erlang) provide better link semantics?
    // We could use unique PIDs in libprocess but rely on name
    // registration when we need to identify the process without the PID.
    // This is how Erlang does it: 
http://www.erlang.org/doc/reference_manual/processes.html#id84804 


- Ben Mahler


On June 16, 2014, 10:41 p.m., Dominic Hamon wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/21667/
> -----------------------------------------------------------
> 
> (Updated June 16, 2014, 10:41 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Ben Mahler, and Vinod Kone.
> 
> 
> Bugs: MESOS-1374
>     https://issues.apache.org/jira/browse/MESOS-1374
> 
> 
> Repository: mesos-git
> 
> 
> Description
> -------
> 
> See summary.
> 
> 
> Diffs
> -----
> 
>   src/sched/sched.cpp 6e14f1cf78b9518a04c35f856fa2358a3d72ddcf 
> 
> Diff: https://reviews.apache.org/r/21667/diff/
> 
> 
> Testing
> -------
> 
> make check.
> 
> Added os::setenv("LIBPROCESS_PORT", "4242") to src/tests/main.cpp and make 
> check.
> 
> 
> Thanks,
> 
> Dominic Hamon
> 
>

Reply via email to