GAE Pathfinder in MG ( all problems solved? )

[tomreyn]

I (Linux) played with Ultifd (Windows) yesterday, using RoutePlanner. Ultifd was hosting, and we had debug logging active for the path finder, and went out of synch. I hope he still has his logs available and will be able to report (please see my email, Ultifd).

Unrelated to this, I just setup a local game where I was observing and had 7 mega bots compete each other. After some minutes, the game printed a one-liner message on console, then finally ran into a segfault:

(click to show/hide)

glest.bin: /home/user1/SCM/megaglest-trunk/source/glest_game/ai/annotated_map.cpp:105: void Glest::Game::AnnotatedMap::updateMapMetrics(const Shared::Graphics::Vec2i&, int): Assertion `cellMap->isInside(pos.x + size - 1, pos.y + size - 1)' failed.
v3.4.1-dev-GNUC: 40405 [64bit]-Feb 15 2011 09:11:14, SVN: [Rev: 1752M], [STREFLOP]

[thread info removed]
Core was generated by `./glest.bin'.
Program terminated with signal 6, Aborted.
#0  0x00007ff135c11ba5 in raise () from /lib/libc.so.6
#0  0x00007ff135c11ba5 in raise () from /lib/libc.so.6
#1  0x00007ff135c156b0 in abort () from /lib/libc.so.6
#2  0x00007ff135c0aa71 in __assert_fail () from /lib/libc.so.6
#3  0x00000000004bb76f in Glest::Game::AnnotatedMap::updateMapMetrics (this=0x7ff116115800, pos=..., size=<value optimised out>) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/ai/annotated_map.cpp:105
#4  0x00000000006877e3 in updateMapMetrics (this=0x2744a60, unit=0x7ff10dbb57c0) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/ai/cartographer.h:139
#5  Glest::Game::UnitUpdater::updateBuild (this=0x2744a60, unit=0x7ff10dbb57c0) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/world/unit_updater.cpp:628
#6  0x00000000006805a6 in Glest::Game::UnitUpdater::updateUnitCommand (this=0x2744a60, unit=0x7ff10dbb57c0) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/world/unit_updater.cpp:247
#7  0x00000000006825c0 in Glest::Game::UnitUpdater::updateUnit (this=0x2744a60, unit=0x7ff10dbb57c0) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/world/unit_updater.cpp:142
#8  0x00000000006a06b0 in Glest::Game::World::updateAllFactionUnits (this=0x2744390) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/world/world.cpp:284
#9  0x00000000006a5307 in Glest::Game::World::update (this=0x2744390) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/world/world.cpp:367
#10 0x00000000004cb8c7 in Glest::Game::Game::update (this=0x2744370) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/game/game.cpp:776
#11 0x000000000053f119 in Glest::Game::Program::loopWorker (this=0x11dcdf0) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/main/program.cpp:283
#12 0x000000000053b865 in Glest::Game::glestMain (argc=<value optimised out>, argv=<value optimised out>) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/main/main.cpp:1767
#13 0x000000000053c4f6 in glestMainWrapper (argc=1, argv=0x7fff791a9248) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/main/main.cpp:1820
#14 main (argc=1, argv=0x7fff791a9248) at /home/tomreyn/SCM/megaglest-trunk/source/glest_game/main/main.cpp:1832

pathfinder.log and all other possibly relevant information is (temporarily) available, too.

[ultifd]

Well I didn't get an email, but I guess I can do a quick report now. Nevermind, found it.

So we test a 2 vs 2 magic vs tech game... Once I clicked on the upgrade button (to battlemage) on 2 initiates (also those were the first ones that I was going to upgrade), I think it went out of synch.  This was windows to linux.
Here is my pathfinder log: http://www.mediafire.com/?3pqyqevi5p5xa2d
Thanks and good luck, I hope it'll help a little bit/somehow!

[tomreyn]

So we test a 2 vs 2 magic vs tech game... Once I clicked on the upgrade button (to battlemage) on 2 initiates (also those were the first ones that I was going to upgrade), I think it went out of synch.  This was windows to linux.

I've uploaded my (client) logs for this game to http://freeit.org/glest/logs/pathfinder.7z where they will be available for a a limited time.

[titi]

I think these out of syncs can be ignored!
We currently have general out of sync problems in svn! We have to test the routeplanner again if this is fixed.