----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/64098/#review191979 -----------------------------------------------------------
Looks good to me overall. Great to see this patch coming. Do we also need to send status updates when we re-register an already registered but disconnected agent (`reconcileKnownSlave()` method)? I was thinking about the scenario described in [MESOS-8185](https://issues.apache.org/jira/browse/MESOS-8185). It seems that we're safe as long as the master persists "kill task" requests in memory until it actually gets a terminal task status, and the agent retries status updates until they are acked. Am I correct? src/master/master.cpp Lines 6789 (patched) <https://reviews.apache.org/r/64098/#comment269978> I think this is not specific to unreachable agents. Can be an agent that was recovered after failover. src/master/master.cpp Lines 6790 (patched) <https://reviews.apache.org/r/64098/#comment269982> `REASON_RECONCILIATION`? src/master/master.cpp Lines 6800-6802 (patched) <https://reviews.apache.org/r/64098/#comment269977> Nit: There's double space between "disconnected" and "framework". Also, I think it will fit into 80 character line. src/tests/persistent_volume_tests.cpp Lines 1512-1515 (patched) <https://reviews.apache.org/r/64098/#comment269985> Should we just ignore this status update here? It doesn't seem related to the thing under test. src/tests/upgrade_tests.cpp Lines 219-221 (patched) <https://reviews.apache.org/r/64098/#comment270033> Ditto, should we just ignore this status update since it's not related to the thing test? - Ilya Pronin On Nov. 27, 2017, 4:55 p.m., Megha Sharma wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/64098/ > ----------------------------------------------------------- > > (Updated Nov. 27, 2017, 4:55 p.m.) > > > Review request for mesos, Ilya Pronin, James Peach, and Jiang Yan Xu. > > > Bugs: MESOS-6406 > https://issues.apache.org/jira/browse/MESOS-6406 > > > Repository: mesos > > > Description > ------- > > Master will send task status updates to frameworks when an agent > re-registers. > > > Diffs > ----- > > src/master/master.cpp 2ddd67ada3731803b00883b6a1f32b20c1bb238f > src/tests/master_allocator_tests.cpp > 3400d70bb0ba564eac43c4639eee0efd4d8059e6 > src/tests/master_tests.cpp 9c450b9f592d9e09a468f537d9b500e97acc636b > src/tests/partition_tests.cpp e49c474167076b4136a161ed29b11db9a13455a7 > src/tests/persistent_volume_tests.cpp > acfeac16884b00581a3523607ff26f44f6dca53a > src/tests/slave_recovery_tests.cpp c864aa92d9ff128a89dbc25653385de25653f56a > src/tests/upgrade_tests.cpp 7f434dbba858f636719eec24e92b306b76430c4c > > > Diff: https://reviews.apache.org/r/64098/diff/3/ > > > Testing > ------- > > with make check > > > Thanks, > > Megha Sharma > >