Crash in ZR-CSAPI: processMessage()


#1

Some testers have seen a crash in ZR-CSAPI when connecting/disconnecting via SSH. the crash file in ~/Library/Logs/DiagnosticReports

indicates this stack trace for the thread that crashes. We think we fixed the problem: it will be in the next beta build.

Thread 3 Crashed:
0   libstdc++.6.dylib                   0x9b890a11 std::string::compare(std::string const&) const + 21
1   us.zoom.ZAPPI.ioWrapper             0x00750e74 bool std::operator==<char, std::char_traits<char>, std::allocator<char> >(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) + 36
2   us.zoom.ZAPPI.ioWrapper             0x0075066e ZMSshServerSession::processMessage(ssh_session_struct*) + 5262
3   us.zoom.ZAPPI.ioWrapper             0x0074ebce ZMSshServerSession::runTask() + 558
4   us.zoom.ZAPPI.ioWrapper             0x00753451 ZMThread::threadProc() + 65
5   us.zoom.ZAPPI.ioWrapper             0x00753368 ZMThread::s_ThreadProc(void*) + 56
6   us.zoom.ZAPPI.ioWrapper             0x0075391f void boost::_bi::list1<boost::_bi::value<ZMThread*> >::operator()<void* (*)(void*), boost::_bi::list0>(boost::_bi::type<void>, void* (*&)(void*), boost::_bi::list0&, int) + 95
7   us.zoom.ZAPPI.ioWrapper             0x007538b6 boost::_bi::bind_t<void, void* (*)(void*), boost::_bi::list1<boost::_bi::value<ZMThread*> > >::operator()() + 70
8   us.zoom.ZAPPI.ioWrapper             0x0075383c boost::detail::thread_data<boost::_bi::bind_t<void, void* (*)(void*), boost::_bi::list1<boost::_bi::value<ZMThread*> > > >::run() + 28
9   libboost_thread.dylib               0x008b38d0 boost::(anonymous namespace)::thread_proxy(void*) + 160
10  libsystem_pthread.dylib             0x9cbe9780 _pthread_body + 138
11  libsystem_pthread.dylib             0x9cbe96f6 _pthread_start + 155
12  libsystem_pthread.dylib             0x9cbe6f7a thread_start + 34

 


#2

We checked in a fix: The bug number is ZOOM-34273. The fix will be in the next beta build.


#3

Thats great!  Any insight into a suspect problem at the code level?  I was just curious what you thought was the suspect code that the thread took a dive to its death.  The problem occurs when we reconnected and restarted the server.