Daniel Heemann a écrit :
> - We're using v0.8beta4 in network mode.
> - Server is a WindowsXP machine which also sometimes crashes but there's no
> development stuff installed, so there's currently no option to get some
> debug information from this side, perhaps we can change this within the
> next time.
The preferred way to use wormux as server is under linux. Of course,
bugs that would occur only when the server is under windows are
interesting beasts we'd like to squash.:)
> - Client is up2date Gentoo and Wormux compiled from sources with
> only --enable-debug specified as configure switch (beside --prefix) - all
> gdb outputs are generated on this machine.
I guess that if you're able to play with those client/server, version is
0.8beta4 also for gentoo, contrary to what I first thought ('up2date'
gentoo = svn).
> 1) After a round is finished the stats are displayed but after klicking the
> green 'OK'/'Continue' button at the bottom the game crashes w/o any error
> message but it seems it just exits
> (wormux-dbg-log.crash_after_round_completed).
And no stderr... Well, that's strange. If it had crashed, it would have
generated a trace. Here it seems to just exit gracefully.
Anyway, that's the first report of such problem. Very weird, and the
backtrace, being empty, is quite unusable.
How often does the bug occur? Is there a pattern to its occurrence ?
If you are willing to help us debug, you can try the following:
- meet us on irc.freenode.org, channel #wormux :)
- try to set a breakpoint inside of MessageEndOfGame in game/game.cpp
from your description, the bug occurs at least during or after
menu.Run() line
> 2) A segfault is caused really regularly every perhaps 4th or 5th round - I
> couldn't assign it to any special action/input or similar so far
> (wormux-dbg-log.sigsegv_Weapon::GetType).
Code for bonus box has changed a bit since then. I'm not sure how this
affects the code. I think I'll request some help on that one.
> 3) Only once so far the game crashed like shown in the last log
> (wormux-dbg-log.missed_assertion_Network::GetInstance-sync_lock)
Strange one. For some reason, the client was already waiting for
synchronisation to finish, as if server stuttered such requested. Do you
know what happened on the server's side?
> Hope this helps you.
In any case, thanks for spending that much time preparing such bug
reports. I'm personally not completely available for work on wormux, so
I'm not sure how far and fast those problems will be processed.
Best regards,
--
Kurosu
_______________________________________________
Wormux-dev mailing list
Wormux-dev@gna.org
https://mail.gna.org/listinfo/wormux-dev
