----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59746/#review176821 -----------------------------------------------------------
src/slave/slave.cpp Lines 5152 (patched) <https://reviews.apache.org/r/59746/#comment250300> We might want to have a metric called `aborted_container_launches` or something like that. If a framework is expected not to stop until no more tasks are running, operators could monitor this metric and be alerted if there's a bug. - Gastón Kleiman On June 2, 2017, 1:10 p.m., Alexander Rukletsov wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/59746/ > ----------------------------------------------------------- > > (Updated June 2, 2017, 1:10 p.m.) > > > Review request for mesos, Ian Downes, Jie Yu, Joseph Wu, and Jan Schlicht. > > > Bugs: MESOS-7601 > https://issues.apache.org/jira/browse/MESOS-7601 > > > Repository: mesos > > > Description > ------- > > Discarded future returned from the containerizer->launch() does not > necessarily mean that the container launch has failed. For example, > a framework may stop while its task are being started. > > > Diffs > ----- > > include/mesos/mesos.proto 5f80170fcd3c05add8b6e9e3107cff062818c1dc > include/mesos/v1/mesos.proto 4b528751006f709f841e44f48c9f5c2dc035b402 > src/slave/slave.cpp 0c7e5f4ef905b3897d341c3147a208fc7a8a12e0 > > > Diff: https://reviews.apache.org/r/59746/diff/1/ > > > Testing > ------- > > make check on several Linux distros. > > > Thanks, > > Alexander Rukletsov > >
