-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/68307/#review207328
-----------------------------------------------------------


Ship it!




I would prefer another wording of the description but the code itself looks 
good to me, how about:

> This test ensures that when a master receives a new authentication 
> request for an scheduler which already has initiated authentication,
> the master discards the the old one and proceeds with the new one.
>
> Since the master does not distinguish between agents or schedulers
> during authentication, this tests is sufficient to cover both cases.

- Alexander Rojas


On Aug. 14, 2018, 12:24 a.m., Benjamin Mahler wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/68307/
> -----------------------------------------------------------
> 
> (Updated Aug. 14, 2018, 12:24 a.m.)
> 
> 
> Review request for mesos, Alexander Rojas, Gastón Kleiman, Meng Zhu, Till 
> Toenshoff, and Vinod Kone.
> 
> 
> Bugs: MESOS-9144
>     https://issues.apache.org/jira/browse/MESOS-9144
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> This test ensures that when the master sees a new authentication
> request for a particular agent or scheduler (we just test the
> scheduler case is tested here since the master does not distinguish),
> the master will discard the old one and proceed with the new one.
> 
> 
> Diffs
> -----
> 
>   src/internal/devolve.hpp 8007f420b3a912d7eff1fa5faa7d8502eb3e9115 
>   src/internal/evolve.hpp e792ff591eff537e2a4661afe08795f00eb35843 
>   src/master/master.hpp 209b998db8d2bad7a3812df44f0939458f48eb11 
>   src/master/metrics.hpp df28a486ead68421970723060850de3ac32e68a7 
>   src/tests/authentication_tests.cpp c9a8f85951a50e278ae509f4efa7105755015ce9 
> 
> 
> Diff: https://reviews.apache.org/r/68307/diff/2/
> 
> 
> Testing
> -------
> 
> Ran in repetition.
> 
> 
> Thanks,
> 
> Benjamin Mahler
> 
>

Reply via email to