----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/41314/#review110094 -----------------------------------------------------------
Can you try to explain why we need a sequence per receiver instead of a single, global one? 3rdparty/libprocess/src/process.cpp (lines 1136 - 1138) <https://reviews.apache.org/r/41314/#comment169878> Have a hard time understanding the comment. What do you think about this variant? ``` CHECK(authentications.find(receiver) != authentications.end()); // Remove the receiver's authentication sequence once it has been // processed completely. if (authentications[receiver].last == future) { authentications.erase(receiver); } ``` In the end, I am not even sure we need this comment - leaving it up to you. - Till Toenshoff On Dec. 12, 2015, 4:12 p.m., Alexander Rojas wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/41314/ > ----------------------------------------------------------- > > (Updated Dec. 12, 2015, 4:12 p.m.) > > > Review request for mesos, Bernd Mathiske, Ben Mahler, Michael Park, and Till > Toenshoff. > > > Repository: mesos > > > Description > ------- > > The committed version of the HTTP Authentication remove a cleanup function > which > ended causing mesos to hang. This patch reintroduces the discarded cleanup > function and per receiven process message queuing. > > > Diffs > ----- > > 3rdparty/libprocess/src/process.cpp > 43c83e30d5c59bd9845f5126ed79985d7c70b0c2 > > Diff: https://reviews.apache.org/r/41314/diff/ > > > Testing > ------- > > At least three times > `make -j8 check` > > Additionaly > `./3rdparty/libprocess/libprocess-tests --gtest_filter="HTTPTest.*" > --gtest_repeat=10000 --verbose` > `./3rdparty/libprocess/libprocess-tests --gtest_filter="HTTPTest.Endpoints" > --gtest_repeat=10000 --verbose` > > > Thanks, > > Alexander Rojas > >