----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/32123/#review76757 -----------------------------------------------------------
src/slave/containerizer/mesos/containerizer.hpp <https://reviews.apache.org/r/32123/#comment124426> IIUC killed=true here if it has been killed externally by the slave calling destroy() rather than being killed internally as a result of a limitation. But later we set killed=true in the Termination to indicate it was killed by a limitation...? Can you document what killed means and perhaps rename it, if my understanding above is correct? src/slave/containerizer/mesos/containerizer.hpp <https://reviews.apache.org/r/32123/#comment124418> call preparation(s) to be consistent with the similiar Future<<>> isolation, below? src/slave/containerizer/mesos/containerizer.cpp <https://reviews.apache.org/r/32123/#comment124427> It wasn't a race before was it? I thought the previous behavior was the cleanup would get skipped if the container was preparing and the relevant Container struct would get dropped, leaking resources. src/slave/containerizer/mesos/containerizer.cpp <https://reviews.apache.org/r/32123/#comment124432> so killed means both instructed to be killed and killed because of a limitation? src/tests/containerizer_tests.cpp <https://reviews.apache.org/r/32123/#comment124417> s/unitl/until src/tests/containerizer_tests.cpp <https://reviews.apache.org/r/32123/#comment124422> kill extra newline src/tests/containerizer_tests.cpp <https://reviews.apache.org/r/32123/#comment124433> What do we expect the Termination struct to contain? According to above we should have killed=true and no status present? - Ian Downes On March 16, 2015, 12:30 p.m., Timothy Chen wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/32123/ > ----------------------------------------------------------- > > (Updated March 16, 2015, 12:30 p.m.) > > > Review request for mesos, Ian Downes and Jie Yu. > > > Repository: mesos > > > Description > ------- > > Fix destroying containerizer during isolator prepare. > This ensures the prepare phase is completed before destroying the container. > > > Diffs > ----- > > src/slave/containerizer/mesos/containerizer.hpp > ae61a0fcd19f2ba808624312401f020121baf5d4 > src/slave/containerizer/mesos/containerizer.cpp > fbd1c0a0e5f4f227adb022f0baaa6d2c7e3ad748 > src/tests/containerizer_tests.cpp 565903b754bbf5891891b183a283aee038eb1b8f > > Diff: https://reviews.apache.org/r/32123/diff/ > > > Testing > ------- > > make check > > > Thanks, > > Timothy Chen > >
