> On March 17, 2017, 10:44 a.m., Vinod Kone wrote: > > src/launcher/default_executor.cpp > > Line 910 (original), 921 (patched) > > <https://reviews.apache.org/r/56215/diff/6/?file=1666160#file1666160line923> > > > > is this safe because before `taskHealthUpdated` is called we would've > > already sent a TASK_RUNNING update?
Yes, we always send a TASK_RUNNING before even starting a healh check. - Alexander ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/56215/#review169271 ----------------------------------------------------------- On March 16, 2017, 4:45 p.m., Alexander Rukletsov wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/56215/ > ----------------------------------------------------------- > > (Updated March 16, 2017, 4:45 p.m.) > > > Review request for mesos, Gastón Kleiman and Vinod Kone. > > > Bugs: MESOS-7249 > https://issues.apache.org/jira/browse/MESOS-7249 > > > Repository: mesos > > > Description > ------- > > Sometimes when a new task status update is generated in the executor, > we have to make sure specific data is duplicated from the previous > task status to, e.g., avoid shadowing of these data during > reconciliation. For instance, consider a check status being sent; > in this status update we must include the latest known health > information. > > This patch also refactors `update()` routine into two separate calls: > `createTaskStatus()` which is responsible for creating a task status > from scratch and `forward()`, which is responsible for forwarding task > status updates to the agent. > > > Diffs > ----- > > src/launcher/default_executor.cpp cbd4f7ecd042e7fa603bd69774d95472df2c896d > > > Diff: https://reviews.apache.org/r/56215/diff/6/ > > > Testing > ------- > > See https://reviews.apache.org/r/56218/ > > > Thanks, > > Alexander Rukletsov > >
