-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70377/#review214392
-----------------------------------------------------------




src/tests/fault_tolerance_tests.cpp
Lines 2035-2036 (patched)
<https://reviews.apache.org/r/70377/#comment300598>

    Do we really need these lines?
    
    `DEFAULT_FRAMEWORK_INFO` already has a principal set, and the failover 
timeout doesn't seem to be necessary, because the first driver never 
disconnects.



src/tests/scheduler_http_api_tests.cpp
Lines 581-619 (patched)
<https://reviews.apache.org/r/70377/#comment300600>

    We now have some handy helpers to create/subscribe a framework and get the 
ID: 
https://github.com/apache/mesos/blob/003e47d852f6eb9a1cd7c86750d91b37269aec26/src/tests/operation_reconciliation_tests.cpp#L724-L747
    
    The `MockHTTPScheduler` also has a `error` callback that can be used to 
check whether the master rejects the re-registration attempt.


- Gastón Kleiman


On April 3, 2019, 8:42 a.m., Andrei Sekretenko wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70377/
> -----------------------------------------------------------
> 
> (Updated April 3, 2019, 8:42 a.m.)
> 
> 
> Review request for mesos and Gastón Kleiman.
> 
> 
> Bugs: MESOS-2842
>     https://issues.apache.org/jira/browse/MESOS-2842
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Added two tests to check that the framework cannot change its principal: one 
> for HTTP API, another for process API.
> 
> Actually, the process API test reproduces the master crash described in 
> MESOS-2842.
> 
> 
> Diffs
> -----
> 
>   src/tests/fault_tolerance_tests.cpp 
> 91cc9a2bcb98562cfdb60df229a88a0dabb34792 
>   src/tests/scheduler_http_api_tests.cpp 
> d5b5eec0df0c01c4706c9bb4438c9daa305bd376 
> 
> 
> Diff: https://reviews.apache.org/r/70377/diff/1/
> 
> 
> Testing
> -------
> 
> Ran make check - these two tests fail, others don't.
> 
> 
> Thanks,
> 
> Andrei Sekretenko
> 
>

Reply via email to