----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/61183/#review190177 -----------------------------------------------------------
src/slave/slave.cpp Lines 1280-1287 (patched) <https://reviews.apache.org/r/61183/#comment267449> It would be nice if we only did this when necessary - i.e., if there have, in fact, been additional RP registrations. The agent could keep an optional list of RPs which register/update during that period; I think it would be pretty easy to insert them in the handleResourceProviderMessage handler when the agent isn't registered. Here, and in reregistered(). Do you think it's worth it? src/tests/slave_tests.cpp Lines 8529-8531 (patched) <https://reviews.apache.org/r/61183/#comment267450> "of for the interaction with the usual oversubscription protocol" - I'm not sure what you're saying here, could you re-word? src/tests/slave_tests.cpp Lines 8536 (patched) <https://reviews.apache.org/r/61183/#comment267451> s/an a/and a/ src/tests/slave_tests.cpp Lines 8543-8546 (patched) <https://reviews.apache.org/r/61183/#comment267452> Could you comment on the reason for setting `authenticate_http_readwrite = false` as well? - Greg Mann On Nov. 6, 2017, 12:38 p.m., Benjamin Bannier wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/61183/ > ----------------------------------------------------------- > > (Updated Nov. 6, 2017, 12:38 p.m.) > > > Review request for mesos, Jie Yu and Jan Schlicht. > > > Repository: mesos > > > Description > ------- > > The agent's resource provider manager sends a > 'ResourceProviderMessage' when its managed resources change. This > commit adds handling in the agent so that an 'UpdateSlaveMessage' is > sent to the master to update the total resource available on the > agent. > > In order to provide push-like handling of the resource provider > manager's message queue, we chain recursive calls to the handler for > continuous processing. Initially, processing is kicked off from > 'Slave::initialize'. In this simple implementation we e.g., provide no > direct way to stop processing of messages, yet, but it can be achieved > by e.g., replacing the manager with a new instance (this would also > require updating routes). > > Since the agent can only send an 'UpdateSlaveMessage' when it is > registered with a master, a simple back-off of 5 s is implemented which > will defer processing of a ready message should the agent not yet have > registered. > > To facilitate logging we add a stringification function for > 'ResourceProviderMessage's. > > This patch also adjusts a number of tests to no expect two > 'UpdateSlaveMessage's. > > > Diffs > ----- > > src/resource_provider/message.hpp 3c7c3f2baeb726e04edd6ffbb9784699d7afe521 > src/slave/slave.hpp df1b0205124555dcb6a0efa5c237f5e77fa2bdf7 > src/slave/slave.cpp 79ee163b41ade93cae1054985379d61faf6a081a > src/tests/slave_tests.cpp 9c2db7adf45af4209ecc475d938ce4b77c5a3ab4 > > > Diff: https://reviews.apache.org/r/61183/diff/21/ > > > Testing > ------- > > Tested on a number of platforms on internal CI. > > > Thanks, > > Benjamin Bannier > >