gaoesa

That sounds like a server setting on your test server. Make sure you have the correct g_XPSave and age settings in your test server config and try again.

The aliases database is lost. Just remove the useradb.db from the database directory and the mod will create a new one. I played on your server for a couple maps and I didn't notice anyone getting their XP reset. Including myself.

Have you confirmed that new guids are losing their XP? I don't know of such thing or bug which would cause it. What did the log say after the dbcleanup? You will probably just have to start the aliases database over. I.e. move the useradb.db outside of the database directory.

Aliases database is probably corrupted. The XP is something completely unrelated. However, you can try "/rcon dbcleanup" command to see if that fixes the latter issue.

The wires are thick so that players will see them. The mod includes no method for changing the size. Using the mines effectively requires practicing, but that way it is more rewarding for all.

The config files are loaded at startup and not during the game. The player counts are updated on every change in teams, so the information of the highest concurrent player amount is available for the next map. It's an admin preference if they want to let the settings change during the map or if they prefer having a consistent setup for the duration of the map.

There is no method for executing commands in the client. Only option for such a thing would be a custom menu, but that would not get executed unless the user opens the menu and clicks some button.

It sounds like the issue is in 2.55 client. I doubt there are things that can be done to prevent it. Luckily, it happens only when players download the packs so on subsequent times, they don't have problems. One thing you can do is to instruct players to update their clients.

I don't know of any parsers. I'm using my own PHP script on my server which has the following selections: if(strpos($line,'say:')===0) { $color=2; } else if(strpos($line,'sayteam:')===0) { $color=5; } else if(strpos($line,'saybuddy:')===0) { $color=3; } else if(strpos($line,'privmsg:')===0) { $color=3; } else if(strpos($line,'tprivmsg:')===0) { $color=3; } else if(strpos($line,'mprivmsg:')===0) { $color=3; } else if(strpos($line,'adminchat:')===0) { $color=8; } else if(strpos($line,'voice:')===0) { $color=2; } else if(strpos($line,'^7 called a vote. Voting for:')) { $color=0; } I'm posting this as a sample of the keywords if anyone wants to do a script for the purpose.

To diagnose the reason for the crash we would need to know what is different from the default on your server. Based on this and your previous report there is something at the engine level. Do you use proxy binaries between the engine and the mod?

I can't reproduce this. Do you use custom command to start the panzerwar, if so, please post it.

Thank you for the info. There is one differing crash happening in the engine, that the mod is catching. The others are consistently a crash in the mod that I thought we had already fixed. Even though it was not certain as there never was a way to reproduce it. Now it seems something on your server is triggering it a lot and it would be great if we could get more information about it.

Just to make it clear on this case. The reason why the silent_crash.txt file is empty is because the mod file has been wiped out already. The exception handler is in that binary. I.e. the execution was not in silent mod binary when it happened. Based on the original thread this was posted, the game does not go all the way to the desktop, which indicates the engine encounters an issue it can't resolve and it has to close the game. However, even still this could be happening because of programmatic errors, less likely though, because some of the ET default utility functions have their own ways to handle bad input data. Console logs should provide enough information to recognize such cases.

Do you mean all the way tot he Windows desktop or to the ET main menu? If all the way, please aske as many players as possible to post their silent_crash.txt file, found from the silent directory of their installation, to here. Or please forward those files if that works better for you.

Unfortunately there is nothing inbuilt to deal with this. The team command itself requires 1 second interval or so, but nothing more.

http://mygamingtalk.com/wiki/index.php/Silent_Shrubbot#.21subnetban !subnetban 127.0.0.1 Will ban 127.0.0.* !subnetban 127 Will ban 127.*.*.*

If you make a default install on your server, without even any custom maps, does it work then?

It could cerainly be related to the OpenGL. Yes, I meant AntiVirus. Also, I meant if it would affect the client but I didn't want to suggest you disabling it before connecting to an unknown server. However, if the 2.60 works fine in the same computer, it doesn't seem very likely that drivers would be the cause for you. The r_softwareGL 0 just ensures that you can't emulate OpenGL but you must have a GPU capable for it and the drivers. However, very often in Windows, I assume Windows because of the 2.55 version you are using, low level failures such as drivers and hardware generate BSOD and not such a clean exit.

Looks like the issue is missing db files that were lost in server reinstallation. The corrupted database file is the aliases file, which has no operational effect for the server, just missing info about player names.

Have you tested with your home server setup that possible AV programs don't cause issues?

I don't think cg_damageKick adds anything to the game or the experience. True, hit sounds act from time to time like cheats. The rare occations are acceptable in my opinion as they also add to the experience.

I think these are unrelated so I split the topic. I assume there is no crash info based on the previous report. Do you have a lot of custom sounds and maps on your server. There are some limitations that are more severe for 2.55 clients than 2.60 clients.

I got the database files through private message and I will take a look at them.

Thank you. The documentation was indeed wrong and the version when the flag was removed was 0.5.1. If it is required to go past the intermission, it can be done with "vstr nextmap" command.

I don't find this from the documentation. It is true that originally that was there, but that was a before 0.4.0 version. The g_includedMaps makes sure that played maps are not getting ignored based on the last played rule or by g_excludedMaps. But it does not guarantee that they appear in the vote list. In this case it seems you have so many maps, that they rarely pop up in the list. The g_includedMaps was added in 0.6.2.