> On June 18, 2014, 4:49 p.m., Ben Mahler wrote: > > Rather than exposing Process API functions for event count, can we just > > count the events in the master as a shorter term solution? > > > > It's not clear to me what the right strategy is as far as the Process API > > is concerned.
We could. I never felt comfortable exposing the event queue raw as it was which required child classes to lock/unlock explicitly. One option I considered was exposing through a method that would lock, copy, unlock, return the queue to the caller. I'd like to get this right for this review as I think it's a good opportunity to put the event queue back in private where it should be. - Dominic ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/22756/#review46152 ----------------------------------------------------------- On June 18, 2014, 2:01 p.m., Dominic Hamon wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/22756/ > ----------------------------------------------------------- > > (Updated June 18, 2014, 2:01 p.m.) > > > Review request for mesos and Ben Mahler. > > > Bugs: MESOS-1510 > https://issues.apache.org/jira/browse/MESOS-1510 > > > Repository: mesos-git > > > Description > ------- > > see summary > > > Diffs > ----- > > 3rdparty/libprocess/include/process/process.hpp > c29cd21d0f1d2c9d29140f0e98929e981d56060f > > Diff: https://reviews.apache.org/r/22756/diff/ > > > Testing > ------- > > make check > > > Thanks, > > Dominic Hamon > >
