> On July 27, 2016, 1:31 p.m., Joseph Wu wrote: > > 3rdparty/libprocess/src/libevent_ssl_socket.cpp, lines 106-110 > > <https://reviews.apache.org/r/50477/diff/1/?file=1454471#file1454471line106> > > > > This comment still seems relevant. > > Benjamin Mahler wrote: > Yeah, it's still relevant. However, I removed this comment because it > seemed like a very lengthy way of saying: don't access 'this' after the > 'this' is destructed. When I was reading this code and trying to understand > all of it, this was one comment that didn't help (it seemed very clear to me > from the code). Are you ok with removing it?
Yup, removing it is fine with me. - Joseph ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/50477/#review143784 ----------------------------------------------------------- On July 26, 2016, 7:03 p.m., Benjamin Mahler wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/50477/ > ----------------------------------------------------------- > > (Updated July 26, 2016, 7:03 p.m.) > > > Review request for mesos, Joris Van Remoortere and Joseph Wu. > > > Bugs: MESOS-5913 > https://issues.apache.org/jira/browse/MESOS-5913 > > > Repository: mesos > > > Description > ------- > > The deferred call to SSL_shutdown within ~LibeventSSLSocketImpl > can occur after the socket fd has been closed by Socket::~Impl. > > This can lead to a TLS Alert message being sent on any fd if > it the fd is re-used between the close and the SSL_shutdown! > > Thanks to Jan-Philip Gehrcke for reporting the issue. > > > Diffs > ----- > > 3rdparty/libprocess/include/process/socket.hpp > 881b44b987e5894cac838dae046ab7dbad20b000 > 3rdparty/libprocess/src/libevent_ssl_socket.hpp > 4d376d8b7c1b29105de69bed2e4077f8c94fed0b > 3rdparty/libprocess/src/libevent_ssl_socket.cpp > f4c0b0b97960400b0282837979bf0ed17f56a068 > > Diff: https://reviews.apache.org/r/50477/diff/ > > > Testing > ------- > > make check > > Ran my repro steps (issue HTTP requests while hammering the master with HTTPS > requests). Issue disappears after this fix. > > > Thanks, > > Benjamin Mahler > >
