TMP4

Does your db running? You can check it with the ps command in FreeBSD. If so maybe the db is not yet fully started before start.sh starts auth. (Happens in low tier vps) Edit start.sh and give more time for the db to start if that's the case. (default is 2 sec) If db not running then we have to find out why. You can debug db.core or check syslog too where does it killed. (db != database but the db binary next to your game file what connects to the database ? )

When you run the sql, and if it fails, it gives you some outup, regardless what mysql client you use, because it is coming from the server. What was that? If you don't know, re-run at least the mob_proto table (the sql file is a text file, open it with notepad++ and copy the mob_proto part and re-run)

Check db's and auth's syserr, I guess db is not running (you can check it with the ps command) because it could not connect to the databse.

Your problem is that you are using an x64 system and you installed the packages what contains the x64 version of the libs, but the sources are not upgraded to the x64 platform (and will not be). Why I know it? Because that exact lib only exists there if you installed the package, probably it's a dependency of DevIL. Even if you uploaded the libs from Libs.zip, your x64 libs in /usr/local/lib will be loaded first. Solution: Remove the packages via pkg delete packagename and only use the libs from Libs.zip. (Or migrate the source to the x64 platfrom, then you can install & use x64 packages with their libs, but I cannot help with that.)

Yes it does. MySQL80 is not supported. Please read this if you want to get some clue about upgrading: If you recompiled the game then you can not get the same error since it will never step into that else if (that's why I said dirty fix). So please recompile it. (In a x32 system, or in the provided vm then upload the /usr/metin2/server/share/bin/game file to the vps) gameclean or cd /usr/metin2/src/game/src && gmake clean gamecompile or cd /usr/metin2/src/game/src && gmake -j9 And since you are using some kind if virtualization in your vps then set BIND_IP & PROXY_IP with your private & public IP address.

10.xx.xx.xx? Then your vps have a private IP address. At game/src/config.cpp you can find this else if (!strncmp(netip, "10.", 3)) Edit the number 10 to any other number but keep the size (or edit the 3) like: else if (!strncmp(netip, "00.", 3) It is some dirty fix for Ymir's "great" logic. You have to recompile the game file. You can only compile in a x32 system, you can use the provided vm, then upload the game file to your vps. Also you probably interested in the PROXY_IP since you'll get kicked out after charselect since that 10.x ip is a private ip, you have to proxy the players to the vps's public ip. In every CONFIG file you'll find #BIND_IP: x and #PROXY_IP: x. Uncomment those (remove the #) and set the BIND_IP value to your private ip, and PROXY_IP value to your public ip.

Sorry but none of your error appears to be presents in my files, therefore you did something wrong with them.

PUBLIC_IP and INTERNAL_IP token is not exists, not valid, you don't need those there. To your previous post, Port 15000 is DB, check if that's running. You can use the ps command.

Enable hidden files in the settings. But that db: could not connect have nothing to do with editing an alias command.

The code looks fine. If you change the question mark image to another one like this: d:/ymir work/ui/public/close_button_01.sub does it show up?

You can include your uitarget.py but I suppose you could insert these lines correctly. I still suspect the images, since it does not have uiscript parts, if it not exists, it simply not showing without a syserr just like your situation.

Does your client have the images itself? d:/ymir work/ui/pattern/q_mark_01.tga d:/ymir work/ui/pattern/q_mark_02.tga [Hidden Content]

I see, for your hosting to make the process shorter. Well, you can create a changelang.sh and do the file renames with it, you can also do an sql query in it to change the database common/locale. Not impossible to do but probably takes some time. Good luck ?

You can switch language, either check Language.txt or the video in the first post. Edit: or do you mean a language changer system?

Ok so there are plenty of error here. 1. Your eix-epk does not have any map folder name, the MapProperty.txt and every file are in the base root directory of your eix-epk. Of course it is not finding the metin2_map_ostrovy\MapProperty.txt if the metin2_map_ostrovy is not exists. Repack your eix-epk, make the folder look like this in Eternexus's folder: map_test1\metin2_map_ostrovy\MapProperty.txt Where map_test1 is the name of eix-epk name, and metin2_map_ostrovy is the actualy map folder. 2. Your Index file is a mess tho At least remove the 2. (*) and 3. (emtpy) lines. PACK * map_test1 pack/ 3. And lastly, metin2_map_ostrovy is totally missing from your atlasinfo.txt.

When I click it, it opens the same page in an iframe just as I included a screenshot. Maybe it is disabled outside of your country.

When I want to download it, it just opens the same page in an iframe ? Please use google drive, mega.nz, mediafire, zippyshare or anything more common what's working.

That tutorial is great. If you want you can send me every file to check where did you mistake clientside eix-epk files index file the 1 or 2 atlasinfo.txt depending if you use the one in the locale folder too serverside map folder

That's fine, you really have a problem with the packing then.

The backslash needed for the folder name, not the pack name. metin2_map_a1/ outdoora1 But it is totally correct if you just add it as a patch: * outdoora1 Please paste the MapProperty.txt content to check if it's correct.

That is your error: Probably the whole folder missing. Make sure you added the new pack to the Index file.

I don't have this error. Maybe you are injecting files to the client via World Editor's D:\Ymir Work folder? Because the client does load the ymir work folder in your D:\ drive and that can cause a lot of problem if the files are different. I don't think it's related to the last update, 20082 npc have model name "GM" in the npclist.txt but there is no model with that name. But 20082 is an unused npc, isn't it? You only get that error if you do /m 20082. But maybe that "GM" npc can be found in some older client because I remember some red-black clothing woman npc with a big asian hat called 'gm npc' in like 2010 ? Later today I'll try to find it. If you had to use chmod/chown -R 777 then you didn't extract the serverfile in a normal way with tar -xf command. Lots of people extract it via winrar in windows and upload the thousands file via ftp but that's a really bad approach. It is not a problem, this just says you don't have postfix in your tables so it uses the default configaration (no postfix). A postfix is like tablename --> tablename_something (the something is the postfix) This is again not an error, rather a notice. You get this when you stops a channel. The error can be if you your channels are crashing but you would probably said it to me. Anyway check if you channels are running via ps command. Just write ps in the freebsd console. If your channels are crashing then you have to figure out why they are crashing. You can use gdb. If all channels all cores are up then it is a networking error that you get kicked after charselect. Maybe you used a PROXY_IP in your local machine and forgot to remove it in your vps. Or your vps actually have private IP and actually needs PROXY_IP set (Google Cloud, AWS for example because those are not traditional vps but some kind of virtualisation like you do with virtualbox)

Those are normal. Hit me in DM with a Teamviewer or Anydesk because I can't help just with texting. (In the next 1 hour)

If nothing changed then you didn't fix your edits regarding the 127.0.0.1 since it can't still says "socket_connect: HOST: (My public ipv4 address):15000, could not connect"" Since you just edited back to 127.0.0.1 I suggest to re-extract the serverfile and change nothing but BIND_IP: your_private_ip PROXY_IP: your_public_ip in the CONFIGs. (Again: don't forget to remove the #)