trunk segfaults when timing out from a server

Bug #336397 reported by BabyBug on 2009-03-01
Affects Status Importance Assigned to Milestone
Armagetron Advanced
Status tracked in Trunk
Manuel Moos
Manuel Moos

Bug Description

summary says it all, rev 874. Attached a recording and the backtrace:

#0 0x00007f8e0b6c5fb5 in raise () from /lib/
#1 0x00007f8e0b6c7bc3 in abort () from /lib/
#2 0x00007f8e0b705228 in ?? () from /lib/
#3 0x00007f8e0b70acb8 in ?? () from /lib/
#4 0x00007f8e0b70d276 in free () from /lib/
#5 0x00000000005da76d in GrowingArrayBase::ResizeBase ()
#6 0x00000000005e8a1c in tHeapBase::Insert ()
#7 0x000000000057c7dd in planned_send::planned_send ()
#8 0x000000000057c823 in nMessage_planned_send::nMessage_planned_send ()
#9 0x000000000057e2eb in nMessageBase::Send ()
#10 0x000000000058407d in sn_ConsoleOutRaw ()
#11 0x0000000000584318 in sn_ConsoleOutString ()
#12 0x000000000058448a in sn_ConsoleOut ()
#13 0x000000000051c188 in ePlayerNetID::Greet ()
#14 0x000000000044ec49 in init_game_objects ()
#15 0x0000000000452b74 in gGame::StateUpdate ()
#16 0x00000000005dad43 in tCallback::Exec ()
#17 0x00000000005cbad7 in rPerFrameTask::DoPerFrameTasks ()
#18 0x00000000005cf05d in rSysDep::SwapGL ()
#19 0x00000000005ba430 in uMenu::Message ()
#20 0x0000000000455778 in sg_ClientFullscreenMessage ()
#21 0x0000000000455956 in sg_TodoClientFullscreenMessage ()
#22 0x00000000005f30b2 in st_DoToDo ()
#23 0x00000000005b929d in uMenu::OnEnter ()
#24 0x000000000045387b in net_game ()
#25 0x00000000005b6d82 in uMenu::HandleEvent ()
#26 0x00000000005b939e in uMenu::OnEnter ()
#27 0x00000000005b6d82 in uMenu::HandleEvent ()
#28 0x00000000005b939e in uMenu::OnEnter ()
#29 0x0000000000454496 in MainMenu ()
#30 0x00000000004296c4 in main ()

BabyBug (babybug) wrote :
Yann Kaiser (epsy) wrote :

ah, experienced that

also had my config file just .. z0rked after that

Changed in armagetronad:
importance: Undecided → Critical
milestone: none → 0.3.2
status: New → Confirmed
Manuel Moos (z-man) wrote :

I cant seem to get the recording to play back properly, and my timeout attempts (CC to a nonexistent IP, join a server and KILL it) failed to reproduce it. Could we get steps to reproduce?

The callstack is odd, though, and may be enough for a fix already. The client is attempting to take over the game.

Manuel Moos (z-man) wrote :

The callstack actually could also happen on 0.2.8, when the connection drops just after the greeting fullscreen message is received. The client probably also is trying to send data to an out of range peer, that's what's causing the crash. The bug, of course, is that it's trying to apply game logic and to send stuff.

Changed in armagetronad:
assignee: nobody → z-man
status: Confirmed → Fix Committed
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers

Bug attachments