-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/49543/#review141093
-----------------------------------------------------------


Fix it, then Ship it!




You also need to disable the tests in this patch now, yes?


3rdparty/libprocess/src/tests/test_linkee.cpp (line 38)
<https://reviews.apache.org/r/49543/#comment206481>

    Ditto here, can you use a pointer?



3rdparty/libprocess/src/tests/test_linkee.cpp (lines 40 - 41)
<https://reviews.apache.org/r/49543/#comment206493>

    "Leak" seems a bit misleading, perhaps just say 'sockets' and comment that 
we never close these in order to leave them open?



3rdparty/libprocess/src/tests/test_linkee.cpp (line 41)
<https://reviews.apache.org/r/49543/#comment206480>

    No static non-PODs, can you use a pointer and 'new' it?



3rdparty/libprocess/src/tests/test_linkee.cpp (line 103)
<https://reviews.apache.org/r/49543/#comment206482>

    s/:/: /
    s/construct/create/ here and below for outgoing



3rdparty/libprocess/src/tests/test_linkee.cpp (line 151)
<https://reviews.apache.org/r/49543/#comment206484>

    The future may be discarded here as well, often we just use a tenary to 
print "discarded" for the discarded case.



3rdparty/libprocess/src/tests/test_linkee.cpp (line 160)
<https://reviews.apache.org/r/49543/#comment206485>

    You shouldn't need to do this since Try allows non-const access: 
`outgoing->send`



3rdparty/libprocess/src/tests/test_linkee.cpp (line 166)
<https://reviews.apache.org/r/49543/#comment206487>

    How about os::sleep here to be a bit more readable:
    
    ```
    os::sleep(Seconds(1))
    ```


- Benjamin Mahler


On July 1, 2016, 11:53 p.m., Joseph Wu wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/49543/
> -----------------------------------------------------------
> 
> (Updated July 1, 2016, 11:53 p.m.)
> 
> 
> Review request for mesos, Benjamin Mahler and Artem Harutyunyan.
> 
> 
> Bugs: MESOS-5759
>     https://issues.apache.org/jira/browse/MESOS-5759
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> The ProcessRemoteLinkTests `RemoteUseStaleLink` and 
> `RemoteStaleLinkRelink` were failing because the `test-linkee` was
> closing sockets shortly after the test `::shutdown` the link.  This 
> is the expected behavior of libprocess.
> 
> Rewrites the `test-linkee` to keep all sockets open, regardless of the
> state of the connection.
> 
> 
> Diffs
> -----
> 
>   3rdparty/libprocess/src/tests/test_linkee.cpp 
> 7ba0f4833a5163d2323e1477509fd2fbacc23690 
> 
> Diff: https://reviews.apache.org/r/49543/diff/
> 
> 
> Testing
> -------
> 
> make check (CentOS 7)
> 
> 3rdparty/libprocess/libprocess-tests --gtest_filter="ProcessRemoteLinkTest*" 
> --gtest_break_on_failure --gtest_repeat=10000
> 
> 
> Thanks,
> 
> Joseph Wu
> 
>

Reply via email to