> On Aug. 15, 2018, 8:54 p.m., Chun-Hung Hsiao wrote: > > src/slave/http.cpp > > Lines 3356 (patched) > > <https://reviews.apache.org/r/68147/diff/3/?file=2073008#file2073008line3356> > > > > We should return an `InternalServerError` on failed/discarded. > > Benjamin Bannier wrote: > I don't think this is required as a failed `Future` is automatically > mapped onto an `InternalServerError`.
You are right about this. Which one is more appropriate in your opinion: `InternalServerError` or `ServiceUnavailable`? Feel free to drop this if you prefer the former. > On Aug. 15, 2018, 8:54 p.m., Chun-Hung Hsiao wrote: > > src/slave/slave.cpp > > Lines 8107-8108 (patched) > > <https://reviews.apache.org/r/68147/diff/3/?file=2073010#file2073010line8107> > > > > `"Resource provider manager is not initialized"` > > Benjamin Bannier wrote: > Wouldn't that be leaking internal information? I tried to map this onto > an operator-relevant issue here. How about `"Agent has not registered yet"`? > On Aug. 15, 2018, 8:54 p.m., Chun-Hung Hsiao wrote: > > src/tests/api_tests.cpp > > Lines 7171-7179 (patched) > > <https://reviews.apache.org/r/68147/diff/3/?file=2073012#file2073012line7171> > > > > ``` > > Future<http::Response> v1Response = > > post(slave.get()->pid, v1Call, contentType, DEFAULT_CREDENTIAL_2); > > ``` > > Benjamin Bannier wrote: > I would like to explicitly test the return HTTP status code, and using > the `post` helper just maps a `Forbidden` on a `Failure`. Leave as is? Make sense. Dropping this. Maybe we could slightly modify the helper in the future. - Chun-Hung ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/68147/#review207351 ----------------------------------------------------------- On Aug. 16, 2018, 2:31 p.m., Benjamin Bannier wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/68147/ > ----------------------------------------------------------- > > (Updated Aug. 16, 2018, 2:31 p.m.) > > > Review request for mesos, Chun-Hung Hsiao and Jan Schlicht. > > > Bugs: MESOS-8403 > https://issues.apache.org/jira/browse/MESOS-8403 > > > Repository: mesos > > > Description > ------- > > This patch adds support for triggering permanent removal of local > resource providers. We also add authorization and tests as part of this > patch. > > > Diffs > ----- > > include/mesos/agent/agent.proto 74488e873cbf99ca487403b70691912cf3788288 > include/mesos/v1/agent/agent.proto 5d1ab6fd7f6f21593444467c66ed35ee90aea07f > src/slave/http.hpp 78200874cb03a3d5af4ee0443c03a1d94cd494cc > src/slave/http.cpp 1b6d266a1a7821c9de6871cbca43317b3c392a32 > src/slave/slave.hpp 802d4eb9e9eba2f1175dd85e56dcc80a61e32f74 > src/slave/slave.cpp 78e8666f402be58af5b6e20a715da4998af2615c > src/slave/validation.cpp df5e1373dbe497bc859455dcaf4e064e923bd72e > src/tests/api_tests.cpp 9c9fa9105a27ac7bb7f30d7eb67512d1e1d63d15 > > > Diff: https://reviews.apache.org/r/68147/diff/5/ > > > Testing > ------- > > `make check` > > > Thanks, > > Benjamin Bannier > >
