leak

Core: 6888

SD2: 792

UDB: 364

patches: arena, anticheat

#0  Unit::SetVisibility (this=0xc44c4aedc278aa18, x=VISIBILITY_GROUP_STEALTH) at ../../../src/game/Unit.cpp:8749

http://paste2.org/p/116444

GitExtensionsSetup07.msi kills my explorer on every rightclick into a folder. (32bit winxp)

Is there a precompiled version of Cheetah?

Are those crashes really from the arena patches changes or general mangos ones?

Core: 6859

SD2: 762

UDB: 363

patches: arena, anticheat, mtmaps

* 1 process 28185  Unit::SetVisibility (this=0xc44cf52ec280edc0, 
   x=VISIBILITY_GROUP_STEALTH) at ../../../src/game/Unit.cpp:8751

http://paste2.org/p/111844

No clue if they are fake or not, but that guy has a point. An official letter shouldn't have that much spelling/grammar issues...

And please.. not that stupid DBC file discussion again. Everybody who has read the updated EULA of wow should know exactly what is allowed or not and basicly nothing around emulating the game is allowed.

(http://www.worldofwarcraft.com/legal/eula.html Point 2 A, if people are interested)

But once again, please no stupid discussion about that again.

From a servers forum too, claming that those DMCA letters are fake

http://img360.imageshack.us/img360/160/wowoneef3.jpg

http://img360.imageshack.us/my.php?image=twowowkz1.jpg

Yup, seems to work fine.

Core: 6859

SD2: 762

UDB: 363

patches: arena, anticheat, mtmaps

#5  0x0000000000655063 in std::list<SpellModifier*, std::allocator<SpellModifier*> >::remove (this=0x7f46bfed1690, __value=@0x4437abb8)

http://paste2.org/p/108910

@wza123

Welcome to get the MapInstanced victim club

Should/will multithreaded maps increase performance on a single core cpu?

I doubt it, this patch was mainly ment to get the load off the main thread to have more room till 1 cpu core is used with 100% by that thread.

mangos: 6817

sd2: 736

patches: arena, anticheat, mtmaps

* 1 process 13200  WorldSession::HandleAddIgnoreOpcodeCallBack (result=0x7f8b34114e00, accountId=<value optimized out>)

http://paste2.org/p/106450

Atm it's really going backwards in terms of stability

Still getting riddled with the MapInstanced::GetInstance crash.

No need to get grumpy.

But from a plain objective perspective from outside of mangos you see lots of bigger (good?) patches which never made it in the core and nearly everyone of it got abandoned from the original author (for what reason ever time, interest, etc.), which is a little sad i think(subjective part). Netcode made it (suddenly?) into the core and it turned out it sucks for windows, so somehow the relation between "not getting into the core because not ready" and "got into it but was not ready" is a bit weird.

I picked my words carefully and don't want to step on anyones toes. This is just a bit of feedback.

Well if mangos differs between stdout and stderr you might try

2> instead of >

Yea we might need another epic drama at mangos (forking, devs splitting up,etc.) before arena patch or anything related to it will make it into the core. But hey at least there is now a dev branch.. *sigh*

As good as gentoo is for performance orientated setups (using it myself for ages) the chances are also little higher for breaking stuff.

Dunno but maybe linux newbies are better off with binary packge based distributions.

If you're think your knowledge (and especially rtfm reading capabilities) are advanced enough i think gentoo ranks around #1 for gameservers.

mangos: 6817

sd2: 736

patches: arena, anticheat, mtmaps

2 Crashes:

1 process 9736  Unit::SetVisibility (this=0xc44c7235c297f3bf, x=VISIBILITY_GROUP_STEALTH) at ../../../src/game/Unit.cpp:8733

http://paste2.org/p/101801

Thanks Derex for bringing this guide back to the forums.

6 Before you start debugging mangos you need to have ensured that it is compiled with debug information. To build mangos with debug info just add --with-debug-info to your options when running ../configure. Then compile and install mangos as usual.

--with-debug-info enables -g and sets MANGOS_DEBUG which ain't helping you that much unless your loglevel is "debug".

I'd rather go with "-g -ggdb", if you wanna go if you want no unused code to be compiled in (it's mostly a few sLog.outDebug but hey, aren't we all hunting for performance?)

Speaking of performance:

I switched to a slightly different debugging method.

ulimit -c unlimited

Enables a coredump to be created on signals like SEGVAULT

You'll find a "core.PID" in your mangos bin dir,

gdb mangos-worldd --core core.`cat worldd.pid` --batch --quiet -ex "info threads" -ex "thread apply all bt full" -ex "quit"

Creates the debug output from that coredump file.

This method should be less performance draining than running mangos with an attached debugger and according to gentoo guys (http://www.gentoo.org/proj/en/qa/backtraces.xml) a little less error-prone.

I couldn't find any disadvantages of that method so far, maybe the debug output quality is less detailed, i don't know. If someone has more information about that - feel free to share it.

Ah yeah, before i forget: Depending on the size of your server (playercount-> ram usage) you might wanna remove that coredump after you got the debug log out of it. Those files can get pretty big and will rapidly fill your harddrive.

Vmaps enabled?

/your/location/

git clone git://github.com/mangos/mangos.git mangos

would create you

/your/location/mangos/

p.s. Wrong forum

- enable core dump (less slowdown than launching with gdb I think) :

ulimit -c unlimited

Interesting method. Anybody knows more about that possible slowdown issue?

So am i.

There are plenty of threads/guides about that on this very forum already. Search function might help.

There should be guides about that on forums about debugging.

Google will answer you that question.

Running the server with GDB by default.

Two machines, worldd runs alone on X2 3ghz with 6GB.

Vmaps disabled only for not-instanced world (0,1, 530 mapids), vmaps are still bugged like hell and create massive spell lag if globally enabled (tested recently again), not to mention the crazy memory consumption.

Recent revisions have some lag issues again, before i had no lag at all. Gonna check what latest revision is like any soon.

You are running windows, right? Atm it seems like that mangos + ACE + win -> fail

**edit

Ah before i forget

Without mtmaps the cpu consumption for mangos threads at 1k players was like

12% net

12% net

80% main

(i assume)

now it is like

12% net

12% net

40% main

20% map

20% map

(might be even less on map threads, i have no real numbers in mind atm)

The load is slightly higher but there is at least no more thread that gets fucked because a single core can't deliver the required performance.

I run x64 gentoo with mysql 5.0.60.

About 14 custom patches (arena, mtmaps, anticheat are the bigger ones)

80% of my crashes are from MapInstanced::GetInstance or Spell: DoAllEffectOnTarget as reported serveral times.

But either nobody cares about crash reports anymore or nobody is able to fix those two things.

I managed to get 100h+ uptime with 1k players once but then another warlock killed the server with the famous Spell: DoAllEffectOnTarget bug.

The correct branch for Multithreaded Maps is

git://github.com/scamp/mangos.git

The currently linked one contains other stuff and not the pure mtmaps patch