Some last errors to get rid of...
 

First of all, my server is running and I am not encountering any issues ingame. But... there are some error messages in my 3 server command windows and in the logs I would like to eliminate.

I know its a big bunch, sorry for that. I appreciate any help.

Btw, whats the proper way to shut down the server? Using CTRL-C so far, but I am not sure if theres a better way...

My config:
Titanium client
Latest Server executables (self-compiled)
Latest PEQ Database
Sourced in all SQL updates from 1711 to 1884 (some errors though)
mysql-5.1.56
ActivePerl-5.10.1.1008


eqemu_debug_world_5820.log
eqemu_debug_zone_5932.log
eqemu_error_zone_5932.log
eqemu_quest_zone_5932.log
login_1303241046.log

Most of those logs are normal and there is nothing you could or should do about them. To correct the one about the formation plugins, you should be able to replace the contents of that file with the new one I posted here:

http://www.eqemulator.org/forums/sho...733#post198733

I think the rest of them are all normal to see accept for the soulbinder one which is probably just a minor issue with the soulbinder plugin.

Hmm, what about the missing .path and spell files (.pl) ?

You say the op_code errors are normal? It gives me an uneasy feeling reading a patch file (e.g. 6.2) cannot be applied. Especially if its a file called "titanium_world"... :(

How do you shut down your server instance smoothly?

Missing .path files just means you don't have a .path file for that zone in your maps folder. Most zones don't have them unless you make them yourself manually.

The spell file errors are not really errors. They are a bit misleading and probably should be added to some filter so they don't show up in normal perl errors. Basically, all that means is a spell was cast that does not have a spell script in your quests/spells folder associated with it. This will be true for nearly all spells unless you are making custom spell scripts for them.

The patch errors is just how the server figures out which client version is running on the client that is connecting. It starts checking if it matches the 6.2 opcodes, and if not, then checks Titanium and so on until it finds the matching client version.

The other opcode errors you saw are for opcodes being sent that we currently don't do anything with on the server side and can be ignored. There are literally hundreds of unknown opcodes that the client uses that we don't have setup on the server end. There isn't anything you can do about that unless you want to identify each one and write handling code for them.

I use "#worldshutdown 300 30" in game to shut the server down with a timer and message to my players. Then, I run a stop script since my server runs on Linux, which then basically just kills the processes. On Windows, you should be able to just close your command prompt windows to shut your server down.

Tyvm for the info! :)

Having a clean Titanium install, how come I get an op_code error for "titanium_patch" ? Thats still the part I dont get.

You get it because it's an opcode that has no purpose and we never mapped it to something that has a purpose, it's not needed for running a server because we don't need it (or we haven't found a use for it)

This just means it didn't match the patch it is trying. It checks all of the patches registered to what it is expecting (in this case, it's the enter world opcode) and when it finds it, it stops looping through them. This is normal, and log files aren't always bad things, even if they are filtered under error.

Thanks!
 

Thanks for clearing up those opcode errors. I was seeing a ton of those and was worried there was something wrong with the way I had set up my server!

Its me again ;)

I got some more bugs / strangenesses on my list and I dont want to create new threads for every single one:

1. My SoD client still crashes sometimes when zoning. If that happens theres no loading screen, just the "entering..." message and a busy mouse cursor. I need to kill the process and restart to get to that zone. Buddies of mine got the same error.
Btw, I tried all the CPU affinity fixes. I am playing in Windowed Mode.

2. Sometimes when I enter Highpass Hold, I am warped in some void space in the middle of the map. I can see some of my surroundings, but I cant get out. Theres a message: Large warp detected. It only happens in that zone for me and it happens almost 50% of all tries.

3. After zoning, my char is set always to a fixed startpoint in that zone, even if I left the last zone at its edge. Lets say I am zoning to WC from EC at the northern border, I start in WC on the path again.
Sometimes, even the heading is wrong. Take Highpass Hold for example. After leaving Highpass through the East Karana gorge, you start in East Karana facing Highpass Hold.

4. The command #movechar doesnt work or I am doing something wrong. I thought you can zone other characters directly without to zone yourself and summon them. I get a message "Character has been moved" but nothing happens.

5. Sometimes I get an error message when trying to loot a mobs corpse:
Something like OP_LootRequest failed.

6. Leaving the tutorial through the cave entrance behind the waterfall is bugged. Sometimes its working, sometimes not. I saw an error message about not being experienced to use a portal. Dont remember the exact words.

I appreciate any help on these issues.

The #movechar command only works if the char is completely logged out (exited EQ). If they are in-game or at char select, it will fail to work. It is really only good to move stuck chars after they are logged out.

Those are more of a debug type of error than anything. Really, I think it would be OK for them to be removed (or at least commented out) since I think all of the corpse bug issues have been resolved. Secrets has mentioned that before too, so maybe she will remove them sometime on the SVN :P

Whenever I see this it's because I'm trying to loot an empty corpse that hasn't despawned yet. The most common cause for me is clicking really fast to loot multiple corpses. Using /hidecorpse looted fixes that problem.

Okay, some log entries from my latest zoning crash/hang: (Its really annoying!)


eqemu_debug_zone.log
eqemu_error_zone.log

Are you zoning multiple characters at once or something? On SoD, I notice less zoning crashes/hangs than previous clients, and it is very rare for me. The only time I really see hang while zoning like you describe is if world crashes at some point before I zone. A world crash will always cause you to hang on your next zone attempt no matter what client you are on. I dunno if that is the case with your hang issues, but is it possible your world is crashing and zones remaining up?

Nay, world is running fine. And it happens even if I am the only one zoning.

No idea about the other "bugs" ?

Sounds like maybe your safe point or zone in point for highpass might be for the old version and putting you in some part of the zone that is causing you to fall through the world or something. Check your zone_points and zone table for that stuff. You may have to use the client bitwise fields to make it work properly for all clients. The large warp detected might be from you falling or maybe from the auto-port if you fall under the world.

For zoning and getting moved when going between zones like EC to WC, I haven't really messed with that much. I know you can change the axis to something like 999999 in the zone_points table to do something related to that type of zone line. I dunno if zoning to the same spot in the next zone is fully supported though.