----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/16586/#review31767 -----------------------------------------------------------
Ship it! src/master/detector.cpp <https://reviews.apache.org/r/16586/#comment60492> Perhaps a comment here that the detection loop is being stopped at this point, so we enter an error state in order to return Failures for subsequent detect() calls. This might help make it clear why we're setting error here but not in the per-promise fetched() failure case below. src/zookeeper/detector.cpp <https://reviews.apache.org/r/16586/#comment60495> Perhaps spell out here that we're breaking the watch() cycle when a failure is encountered. - Ben Mahler On Jan. 3, 2014, 11:35 p.m., Jiang Yan Xu wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/16586/ > ----------------------------------------------------------- > > (Updated Jan. 3, 2014, 11:35 p.m.) > > > Review request for mesos, Benjamin Hindman, Ben Mahler, and Vinod Kone. > > > Bugs: MESOS-892 > https://issues.apache.org/jira/browse/MESOS-892 > > > Repository: mesos-git > > > Description > ------- > > - Previously a pending future is returned to the client who calls > Detector::detect() after the detector enters a non-operational state due to a > non-retryable error in the last detect() call. > > > Diffs > ----- > > src/master/detector.cpp 7b437ac2ddf92ece0d5ec7f9dae612e622bfb39e > src/tests/master_contender_detector_tests.cpp > d532c17c64c4c797053616cb486a907855bb723b > src/zookeeper/detector.cpp 6723d036bbd2ea814c81d777d81b2fc0b261a0fd > > Diff: https://reviews.apache.org/r/16586/diff/ > > > Testing > ------- > > Added tests. > make check. > > > Thanks, > > Jiang Yan Xu > >
