----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/66704/#review201510 -----------------------------------------------------------
Fix it, then Ship it! src/slave/slave.cpp Line 2912 (original), 2903-2904 (patched) <https://reviews.apache.org/r/66704/#comment282775> Can you add the comment below between this two lines? ``` // Master set the `launch_executor` flag and this is not a command task. ``` src/slave/slave.cpp Lines 3014 (patched) <https://reviews.apache.org/r/66704/#comment282776> This comment seems not accurate, because we can also reach here in the legacy case where the master did not set the `launch_executor` flag, i.e., launching a new executor is not requested by master. - Qian Zhang On April 19, 2018, 6:11 a.m., James Peach wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/66704/ > ----------------------------------------------------------- > > (Updated April 19, 2018, 6:11 a.m.) > > > Review request for mesos, Gilbert Song, Greg Mann, and Jie Yu. > > > Bugs: MESOS-8585 > https://issues.apache.org/jira/browse/MESOS-8585 > > > Repository: mesos > > > Description > ------- > > Refactored the executor launch path so to make the conditional > logic clearer. After this there is just one place that adds and > launches a new executor and we will subsequently be able to add > error handling there. > > > Diffs > ----- > > src/slave/slave.cpp e5d6c3fac5054a6b0a0b5b77abd850a35be6ccc5 > > > Diff: https://reviews.apache.org/r/66704/diff/2/ > > > Testing > ------- > > make check (Fedora 27) > > > Thanks, > > James Peach > >
