----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/42861/#review124576 -----------------------------------------------------------
src/slave/slave.cpp (line 430) <https://reviews.apache.org/r/42861/#comment187209> I'm confused by the logic here. Suppose I have two mount volumes at `/mnt/foo` and `/mnt/bar`. It seems this would be rejected -- why? Similarly, suppose I have three volumes: `/foo`, `/mnt/foo`, and `/mnt/foo/baz`. This probably _should_ be rejected (because `/mnt/foo` and `/mnt/foo/baz` overlap), but AFAICS it wouldn't be rejected by this patch. Checking whether _all_ MOUNT volumes have the same parent / overlap doesn't seem right; don't we want to check whether any two volumes overlap? src/slave/slave.cpp (line 432) <https://reviews.apache.org/r/42861/#comment187192> "directory" is a better term than "folder". - Neil Conway On Jan. 28, 2016, 2:12 a.m., haosdent huang wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/42861/ > ----------------------------------------------------------- > > (Updated Jan. 28, 2016, 2:12 a.m.) > > > Review request for mesos, Jie Yu and Joris Van Remoortere. > > > Bugs: MESOS-4521 > https://issues.apache.org/jira/browse/MESOS-4521 > > > Repository: mesos > > > Description > ------- > > Ensure two Mount Disk resources do not have the same root path. > > > Diffs > ----- > > src/slave/slave.cpp 1f4c8368feb0ce19963577582ce745acfb21aa9f > > Diff: https://reviews.apache.org/r/42861/diff/ > > > Testing > ------- > > > Thanks, > > haosdent huang > >
