> On March 20, 2014, 9:38 p.m., Ben Mahler wrote: > > What is the reasoning behind this one? > > > > We've typically found that operating with 2GB kept for the OS and > > mesos-slave is more than enough, and we'll probably refine the 1GB default > > as we have more measurements of slave/OS overhead from actual clusters.
Actually I just saw that it was a todo to have a better scheme for disk . And its nice if we have actual data for analysis.Refining the 1GB default would be great as I was also thinking that why the default is set to 1GB and not something from actual data. could you just view the diff. It is for disk, not memory. - ASHUTOSH ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/19427/#review37981 ----------------------------------------------------------- On March 19, 2014, 11:10 p.m., ASHUTOSH JAIN wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/19427/ > ----------------------------------------------------------- > > (Updated March 19, 2014, 11:10 p.m.) > > > Review request for mesos and Ben Mahler. > > > Bugs: MESOS-1106 > https://issues.apache.org/jira/browse/MESOS-1106 > > > Repository: mesos-git > > > Description > ------- > > using 50% as criteria for disk less than 10GB else leave 5GB for OS. > > > Diffs > ----- > > src/slave/containerizer/containerizer.cpp d0a1023 > > Diff: https://reviews.apache.org/r/19427/diff/ > > > Testing > ------- > > make check > > > Thanks, > > ASHUTOSH JAIN > >
