----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/70200/#review213691 -----------------------------------------------------------
src/master/master.cpp Line 9557 (original), 9557-9564 (patched) <https://reviews.apache.org/r/70200/#comment299688> Should this lambda leverage move semantics? It could take an rvalue and do an assignment instead of a copy: ``` update.mutable_update_operation_status() ->mutable_status() = status; ``` It would also be possible to create a single `scheduler::Event` message instead of one per lambda invocation. src/tests/operation_reconciliation_tests.cpp Line 200 (original), 204 (patched) <https://reviews.apache.org/r/70200/#comment299689> `s/ASSERT/EXPECT/` here and on the other tests. The updated test doesn't call `result->response()`, so there is no risk of it crashing. src/tests/storage_local_resource_provider_tests.cpp Lines 4658-4660 (original), 4662-4664 (patched) <https://reviews.apache.org/r/70200/#comment299693> `%s/ASSERT/EXPECT/g` - Gastón Kleiman On March 13, 2019, 11:09 a.m., Greg Mann wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/70200/ > ----------------------------------------------------------- > > (Updated March 13, 2019, 11:09 a.m.) > > > Review request for mesos, Benjamin Bannier, Benjamin Mahler, Chun-Hung Hsiao, > Gastón Kleiman, James DeFelice, Megha Sharma, and Jiang Yan Xu. > > > Bugs: MESOS-9648 > https://issues.apache.org/jira/browse/MESOS-9648 > > > Repository: mesos > > > Description > ------- > > This patch updates the v1 scheduler API's RECONCILE_OPERATIONS > call to provide a 202 Accepted response with an empty body, > rather than a response containing all reconciliation results. > In this new scheme, reconciliation requests are satisfied with > operation status updates on the scheduler's event stream. > Related tests are also updated. > > NOTE that this is a breaking change for schedulers consuming > the experimental operation reconciliation API. > > > Diffs > ----- > > src/master/http.cpp d6d47405f871f88235cc203ef5cf9f1460754e0c > src/master/master.hpp 953cc5b8ab6a8e1920a3ad63fb2dd6382e3603ec > src/master/master.cpp dc68fc324de7242737123015fbac19a2129778ce > src/tests/operation_reconciliation_tests.cpp > 6a815ad694e2a608ce324715c920833f825793a0 > src/tests/storage_local_resource_provider_tests.cpp > 7945384867f26fa15dc734a235ae509d5d6d350f > > > Diff: https://reviews.apache.org/r/70200/diff/1/ > > > Testing > ------- > > `make check` > > > Thanks, > > Greg Mann > >