> On June 16, 2016, 12:59 a.m., Vinod Kone wrote: > > 3rdparty/libprocess/src/metrics/metrics.cpp, line 262 > > <https://reviews.apache.org/r/48601/diff/4/?file=1420115#file1420115line262> > > > > hmm. you are still copying all the metrics into a hashmap and returning > > it, incuring an extra copy, which is not the same as what we were doing > > before. is there a way to refactor this without incurring extra copies for > > the /metrics/snapshot handler?
actually it's not probably that bad because we are already doing a copy in `snapshot()` - Vinod ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/48601/#review137863 ----------------------------------------------------------- On June 15, 2016, 4:27 p.m., haosdent huang wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/48601/ > ----------------------------------------------------------- > > (Updated June 15, 2016, 4:27 p.m.) > > > Review request for mesos, Abhishek Dasgupta, Anand Mazumdar, zhou xing, Jay > Guo, Michael Park, and Vinod Kone. > > > Bugs: MESOS-5484 > https://issues.apache.org/jira/browse/MESOS-5484 > > > Repository: mesos > > > Description > ------- > > Exposed metrics information via `process::metrics::snapshot`. > > > Diffs > ----- > > 3rdparty/libprocess/include/process/metrics/metrics.hpp > 25e8fcaeb9c71eaa43096165745e79c8198c139a > 3rdparty/libprocess/src/metrics/metrics.cpp > be03406c58439b7ab7c1ba12c19dd9f30aff109e > > Diff: https://reviews.apache.org/r/48601/diff/ > > > Testing > ------- > > > Thanks, > > haosdent huang > >
