> On Aug. 28, 2015, 3:42 p.m., Jie Yu wrote: > > src/slave/slave.cpp, lines 4240-4248 > > <https://reviews.apache.org/r/37908/diff/1/?file=1058395#file1058395line4240> > > > > Hum, why zero is a special case here? What if the RE initially returns > > 10cpus, and due to some reaons, it returns nothing later. Do we still need > > to get notified about this change? > > > > I would suggest changing LOG(INFO) to VLOG(1) without other changes.
Alrighty - will do; don't have any strong opinions other than getting it out of non-oversubscribed cluster logs :) - Niklas ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/37908/#review96940 ----------------------------------------------------------- On Aug. 28, 2015, 3:19 p.m., Niklas Nielsen wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/37908/ > ----------------------------------------------------------- > > (Updated Aug. 28, 2015, 3:19 p.m.) > > > Review request for mesos and Jie Yu. > > > Repository: mesos > > > Description > ------- > > Resource estimates are continuously written to the logs even though they are > empty (as with the majority of Mesos clusters at the moment). To reduce the > noise, this patch moves log lines for empty resources to VLOG(2). > > > Diffs > ----- > > src/slave/slave.cpp 8cfad7b86675b10c98524174975828bbcf47aed2 > > Diff: https://reviews.apache.org/r/37908/diff/ > > > Testing > ------- > > make check - with and without the patch, verifying the log statements. > > > Thanks, > > Niklas Nielsen > >
