server shutdown failure (segfault)

[Kevin DuBois]

Occasional shutdown segfault seen:

Program received signal SIGSEGV, Segmentation fault.
0xb6e3f3c6 in ?? () from /lib/arm-linux-gnueabihf/libc.so.6
(gdb) bt
#0 0xb6e3f3c6 in ?? () from /lib/arm-linux-gnueabihf/libc.so.6
#1 0xb6b5f94e in __gnu_cxx::new_allocator<std::_Sp_counted_ptr_inplace<mir::frontend::detail::SocketMessenger, std::allocator<mir::frontend::detail::SocketMessenger>, (__gnu_cxx::_Lock_policy)2> >::deallocate (this=<error reading variable: Cannot access memory at address 0x1c>,
    __p=<error reading variable: Cannot access memory at address 0x18>)
    at /usr/arm-linux-gnueabihf/include/c++/4.9.2/ext/new_allocator.h:110
Backtrace stopped: previous frame inner to this frame (corrupt stack?)

Happens more often when a client is actively sending messages (eg, swapping rapidly), and the server gets a SIGTERM.

[Daniel van Vugt (vanvugt)]

I'm seeing one or two possibly related server crashes this week. Annoyingly can't get stack traces for them yet.

[Kevin DuBois (kdub)]

IIRC, this was a dup of https://bugs.launchpad.net/mir/+bug/1502782

[Kevin DuBois (kdub)]

ah, actually, looking at the backtrace again, disregard message #2... looks different

[Alan Griffiths (alan-griffiths)]

Is this still a thing?

[Launchpad Janitor (janitor)]

[Expired for mir (Ubuntu) because there has been no activity for 60 days.]

[Launchpad Janitor (janitor)]

[Expired for Mir because there has been no activity for 60 days.]