ATAG

It's time to make a more useful mysql lib, like the "official" one ...

It's not an error, it only means that your block_exception table is empty.

I use it for 1.5 years without any problem... perfect

The most understandable translation from that topic: So as i understand, u have at least 2 cores on the same channel and 3 players...

You have to add to game/src/Makefile You have some errors in your Makefile (becuse of the missing perforce server). Change P4_VERSION to any string...

Sounds very good, but the main exe's VT results are very BAD

I prefer OO calc to edit (txt) protos...

Any syserr/syslog from db core?

Try a debug version: when login begin local mapidx = pc.get_map_index() syschat("Map index: "..mapidx or "error") if 352 == mapidx then ...

Thx a lot, but these files can't be "untouched" because it's not working out of the box (some items are missing from the protos) and ther's also some fun in game_option.quest An (as i remember) the regen files on the "beta" maps (dragon rock) are false... maybe it was my fault, i don't know, after 2 hours i get mobs up there.

If you have more than 1 "when kill begin" like quest, you have to priorize in locale_list. Server runs the first (and only one) possible "event/callback" only. Sorry for my bad english, i hope you understand me

16>.AffectFlagContainer.cpp(68) : error C2440: 'initializing' : cannot convert from 'CAffectFlagContainer::Element' to 'char &' Right click on the BYTE and click Go to Definition and change the char to unsigned char. 6>.PythonGraphic.cpp(294) : error C2664: 'CPythonGraphic::SaveJPEG' : cannot convert parameter 2 from 'char *' to 'LPBYTE' Double click on the error and change the line to this: bool bSaved = SaveJPEG(c_pszFileName, (LPBYTE)pbyBuffer, uWidth, uHeight); Thanks! But now i have another problem, i get this error when i try to open the game with the compiled binary: Im using the 40250 Test Client The mainline_released branch is for the SG client (root). Use novaline for the test client.

Take the 0s dif, and set the last "byte" to 0xF (if it is not a noop-like crap dif)

Why r28249? And what is your name? As terrible as your release is...

As i know, in (also) 34k not needed any diff with europian locale O.o PS:On my little crap server it's working without any dif

Is it really a problem?

#ifdef _IMPROVED_PACKET_ENCRYPTION_ struct TPacketKeyAgreement { static const int MAX_DATA_LEN = 256; BYTE bHeader; WORD wAgreedLength; WORD wDataLength; BYTE data[MAX_DATA_LEN]; }; struct TPacketKeyAgreementCompleted { BYTE bHeader; BYTE data[3]; // dummy (not used) }; #endif // _IMPROVED_PACKET_ENCRYPTION_ Try to disable _IMPROVED_PACKET_ENCRYPTION_, i don't know what is it exactly.

Why don't you use the original flame_dungeon quest?

Build with the -static-libstdc++ CFLAG. It will produce a little bigger file size, but i don't think if it's really a problem.

Don't use pc.name for filenames, if name change is enabled.

Mutch better to make a firewall rule to access the ntp port only by the timeserver.

If you haven't got ntpd running, use the ntpdate command (with any time server) to set the clock: ntpdate time.kfki.hu

oblivion

The longer message is made by the test_server mode. Try to set the item flag "ITEM_FLAG_LOG", maybe you will get more details...

It's very interesting...