> On Dec. 12, 2018, 7:17 p.m., Benjamin Mahler wrote: > > Can you clarify when the transition from "previously" to "now" occurred? Is > > it due to your other pending patches, or was this something we changed in a > > previous release? > > > > If it's part of your pending patches, we might have to keep the empty == > > unset behavior, since I'm not sure which frameworks were relying on that.
I completely changed the patch as we were previously supporting set but empty framework IDs, and should continue to support it even with the changes from https://reviews.apache.org/r/69398/. I filed https://issues.apache.org/jira/browse/MESOS-9481 for the outstanding API cleanup. - Benjamin ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/69557/#review211249 ----------------------------------------------------------- On Dec. 17, 2018, 9:43 a.m., Benjamin Bannier wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/69557/ > ----------------------------------------------------------- > > (Updated Dec. 17, 2018, 9:43 a.m.) > > > Review request for mesos, Benjamin Mahler and Meng Zhu. > > > Bugs: MESOS-9469 > https://issues.apache.org/jira/browse/MESOS-9469 > > > Repository: mesos > > > Description > ------- > > This patch add a test for the case where a framework subscribes with a > set, but empty `FrameworkID`. While this has never been officially > supported, the master code currently contains extensive handling for > this case. > > This test can be removed if MESOS-9481 gets implemented. > > > Diffs > ----- > > src/tests/master_tests.cpp 50b905b5ab508378eddf070eabfe744b68fd4cea > > > Diff: https://reviews.apache.org/r/69557/diff/2/ > > > Testing > ------- > > `make check` > > > Thanks, > > Benjamin Bannier > >
