VladimirMangos

In [10194]. Thank you

In additional to adding enums to enum list you need also checks sources for use enum name for std. gm level that go before or after you custom enum. For example it can be SEC_MODERATOR < 0 check and then if you add your custom enum before SEC_MODERATOR then it will true for it. And maybe this you not expect. So check affected cases in code.

So moving this version as outdated from under review..

as you want. Because in any form this is not for core repo. Just hack (why use sure that next time deleted player will not have corrupted typeid field also...) prevent crash in single show up case. and _hide_ possibility find problem.

Install VS 2010 runtime x86 support or build vmap extractor/assembler by self from provided source. I start think that we must drop provided binaries for vmap extractor/assembler to make ppl build in any cases by self....

You patch will low help. Maybe it will not crash by send corrupted data to client and crash clients. or crash later Because Object:: version retunr typeid then in same way switch(obj->GetTypeId()) will use this value for seelction. Corrupted data just will not detected at this point and will showup own affects later.. Need find why deleted object pointer still listed in update list. As i say it crash or deleted player pointer, and ofc old memory when Player object has been placed can be corrupted at time access. But pointer itself must be not corrupted. maybe if add output pointer value in ~Player to log with name and at crash case will can be possible catch what player has been affected and ask user about what he do at crash time...

but then you will not login by 3.3.5.

Hmm, TYPEID_PLAYER = 4,\\ This only possible if Player object deleted at call moment. In [s0011] I backport single found not backported commit with crash fix. I not sure that this help.

400 bracnh not supported. this meaning: it internal development branch made public for intresting ppl, but not expected working at any specific moment, buildable or do any useful. We accept patches suggested for this branch not not help in it build and use. So branch for really good undestanding how fix code and how server/client work.

used for mangos build gcc is 4.3.2 ? Maybe you have diff gcc versions and linker find wrong gcc version libs.

Possible we time look at silverice. He good understand update packets system work (show in added his patches), also have good code style (i mostly commit his patches without any changes), and attempt always write generic not hack code in general. He also in trinity team as i understand but not very active last time in it and i not see any pro-trinity position in brian like way so that don't must create problem for us. He also sometime present at channel, maybe not so active in discussion but i can say that i often discuss some code problems with him at private channel.

Anyone can reproduce it with fresh sources (after http://github.com/mangos/mangos/commit/09754dd089fad032f9529ef577f6a283bb66d4d9 ) ?

you not need it or you attempt make some horribale hack and then we not support this. that why this functions not exported.

binaries builds as x86 application so you need install x86 runtime (it independent from x64 version)

then this meaning "some custom patches applied". Spaces can't be from nowhere This not meaning that this custom patches affect problem, or maybe affect.

dummy effects/auras named in like way because it not have standard implementation, and each spellid (or group) expected special code, this is not only just call another spell but often something different. In like case any DB table useless and not help.

StartHeroicPlayerLevel, also you sure that you not have some custom patches for GM level set...

you have anti-freeze protection triggering. You can for example increase in config time delay after that it triggered. We can't make reload in other thread, more, we _must_ lock any map threads when per-map threading will used because before full reload complete unsafe use related data. This just note for self for future. But for resolve this problem possible need implement some anti-freeze disable at reload time...

Wrong forum for discuss this

1. wrong config file edited 2. you have duplicate option above changed with 10 3. other

then just build extractor by self from sources, or install VS2010 runtime support

I not crash at reported class/race combination. I check with different DB that YTDB.

reports not related to each other. 2.x and 3.x have totally different way send stacking info to client and recent 3.x aura work changes not added in any form to mangos-0.12

data have some differences in 0.12 client by structure but first look this not must be something impossible. So this planned without any date pre-declaration.

you need vs2010 runtime installed at computer. You can get it from MS site for example Microsoft Visual C++ 2010 Redistributable Package (x86) Microsoft Visual C++ 2010 Redistributable Package (x64)