-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/64546/#review193776
-----------------------------------------------------------


Fix it, then Ship it!





src/tests/resource_provider_manager_tests.cpp
Line 1133 (original), 1137 (patched)
<https://reviews.apache.org/r/64546/#comment272425>

    We need to make sure that the info used below has an id,
    
        ASSERT_TRUE(resourceProvider->info.has_id());



src/tests/resource_provider_manager_tests.cpp
Line 1134 (original), 1138-1139 (patched)
<https://reviews.apache.org/r/64546/#comment272426>

    I'd suggest to just overwrite `resourceProviderInfo`
    
        resourceProviderInfo = resourceProvider->info;
    
    We can then use `resourceProviderInfo.id()` in place of 
`resourceProviderId`.


- Benjamin Bannier


On Dec. 12, 2017, 4:15 p.m., Jan Schlicht wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/64546/
> -----------------------------------------------------------
> 
> (Updated Dec. 12, 2017, 4:15 p.m.)
> 
> 
> Review request for mesos, Alexander Rukletsov, Benjamin Bannier, and Jie Yu.
> 
> 
> Bugs: MESOS-8315
>     https://issues.apache.org/jira/browse/MESOS-8315
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Callbacks of already invalidated instances could be called in
> 'ResubscribeResourceProvider' when simulating a resource provider
> restart. This has been resolved by not only restarting a resource
> provider's driver, but creating a new 'MockResourceProvider' instance.
> 
> 
> Diffs
> -----
> 
>   src/tests/resource_provider_manager_tests.cpp 
> e37a53ac6a03e2ea58dd6580fc8a399a1398d950 
> 
> 
> Diff: https://reviews.apache.org/r/64546/diff/1/
> 
> 
> Testing
> -------
> 
> make check
> 
> src/mesos-tests 
> --gtest_filter=*ResourceProviderManagerHttpApiTest.ResubscribeResourceProvider*
>  --gtest_repeat=1000 --gtest_break_on_failure (with 'stress' running in the 
> background)
> 
> 
> Thanks,
> 
> Jan Schlicht
> 
>

Reply via email to