> On April 19, 2017, 5:07 p.m., Anand Mazumdar wrote: > > src/tests/protobuf_utils_tests.cpp > > Lines 289 (patched) > > <https://reviews.apache.org/r/58360/diff/3/?file=1694372#file1694372line289> > > > > Can you add a comment around why we added this test? > > Chun-Hung Hsiao wrote: > It seems to me that if we don't want to enforce a hard dependency on > protobuf-3.x, we should not add it.
I don't mind taking out the test, but I definitely need to make sure we are committed to supporting large message (>64MB). The only alternative I can see is to ensure we have proper API design to paginate any potential protobuf message which could go large within the 64MB limit imposed by protobuf2. AFAIK both `GET_STATE` and `OFFERS` messages could exceed this and render the cluster not usable for an important use case. Do you have any concern about a hard dependency to protobuf 3.x? - Zhitao ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/58360/#review172365 ----------------------------------------------------------- On April 19, 2017, 4:23 p.m., Zhitao Li wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/58360/ > ----------------------------------------------------------- > > (Updated April 19, 2017, 4:23 p.m.) > > > Review request for mesos and Anand Mazumdar. > > > Bugs: MESOS-6644 and MESOS-7228 > https://issues.apache.org/jira/browse/MESOS-6644 > https://issues.apache.org/jira/browse/MESOS-7228 > > > Repository: mesos > > > Description > ------- > > Before protobuf 3.2.0, this test would fail because the 64MB limit > imposed by older version of protobuf. > > > Diffs > ----- > > src/tests/protobuf_utils_tests.cpp 5239182812835b93a28e85146b2df2b20ae77328 > > > Diff: https://reviews.apache.org/r/58360/diff/3/ > > > Testing > ------- > > > Thanks, > > Zhitao Li > >
