> On June 10, 2014, 11:38 a.m., Dominic Hamon wrote: > > src/master/master.cpp, line 383 > > <https://reviews.apache.org/r/22427/diff/1/?file=606521#file606521line383> > > > > why not clean these up during finalize? > > Jiang Yan Xu wrote: > Because they are owned?
good point! i'm a little concerned about mixing the two lifetime scopes of a process. We have ctor to dtor and initialize to finalize. Now this lives from initialize to dtor, which is a little strange but probably harmless. as an aside, owned doesn't mean scoped. it is scoped as a side-effect of RAII but it is meant to indicate unique ownership. - Dominic ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/22427/#review45268 ----------------------------------------------------------- On June 11, 2014, 10:39 a.m., Jiang Yan Xu wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/22427/ > ----------------------------------------------------------- > > (Updated June 11, 2014, 10:39 a.m.) > > > Review request for mesos, Benjamin Hindman, Ben Mahler, and Vinod Kone. > > > Bugs: MESOS-1444 > https://issues.apache.org/jira/browse/MESOS-1444 > > > Repository: mesos-git > > > Description > ------- > > To simplify the implementation, RateLimiters are added statically during > initialize() and never removed. > > visit(MessageEvent) and visit(ExitedEvent) are throttled together to keep > their order intact. > > > Diffs > ----- > > src/master/master.hpp 7a121856806ff2d79661fa31e491ddd6e4512c59 > src/master/master.cpp 4a01b1aacfff83c62aa2ec3b46ad64e631488d09 > src/tests/rate_limiting_tests.cpp PRE-CREATION > > Diff: https://reviews.apache.org/r/22427/diff/ > > > Testing > ------- > > Added a test for MESOS-1444. Will add more for MESOS-1445 > > > Thanks, > > Jiang Yan Xu > >
