----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/60500/#review180728 -----------------------------------------------------------
src/messages/flags.proto Lines 56 (patched) <https://reviews.apache.org/r/60500/#comment255947> I see Jie's point. From a framework's standpoint a name is a single domain of IP connectivity, so ideally if we have a network name that is both a CNI and CNM network (DC/OS overlay or Calico comes to mind here), it implies that it should be a single network and hence having a common DNS configuration across CNI/CNM is not incorrect. That said instead of using Mesos and Docker as the fields if we use CNI and CNM as the field names would that make sense? We would then be pegging against the standard rather than the `Containerizer`? The reason I am proposing to keep two different fields here is primarily to take into consideration any incosistencies that might pop up in these standards at some point (right now there don't seem to be any for DNS). - Avinash sridharan On July 5, 2017, 7:03 a.m., Qian Zhang wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/60500/ > ----------------------------------------------------------- > > (Updated July 5, 2017, 7:03 a.m.) > > > Review request for mesos, Avinash sridharan and Jie Yu. > > > Bugs: MESOS-7709 > https://issues.apache.org/jira/browse/MESOS-7709 > > > Repository: mesos > > > Description > ------- > > Introduced `--default_container_dns` agent flag. > > > Diffs > ----- > > docs/configuration.md 007d0f556375f331148c9ad09be8b301ab31a01c > src/messages/flags.hpp 70ad58c33067a0b058deafdcc4ffc5f554b6ca72 > src/messages/flags.proto e87075f19714739b93d9f4aa33ea8686a1bb5613 > src/slave/flags.hpp c6803eb54e09a5497755e1e5fef2872193eacba6 > src/slave/flags.cpp 398768656b5fa3b7c85474de2b4b008bf7b85cb3 > > > Diff: https://reviews.apache.org/r/60500/diff/2/ > > > Testing > ------- > > > Thanks, > > Qian Zhang > >
