Well,xbx wrote:Here is the dedicated for Trackmania United Forever and Trackmania Nations Forever.
You can download it from
http://files2.trackmaniaforever.com/Tra ... -05-19.zip
http://files2.trackmaniaforever.com/Tra ... -05-19.tgz
(both contain the same files, and windows and linux executable.)
- AutoSaveReplays is no longer ignored when using SetServerOptions.
- StartServerInternet() doesn't work like in TMU/TMN -> return an error instead of pretending everything is fine...
- A few settings that influence the server ressource usage (SetMaxPlayers, SetMaxSpectators, EnableP2PUpload, EnableP2PDownload, AutoSaveReplays, AutoSaveValidationReplays) are now reserved to the SuperAdmin.
- Avoid to waste server P2P and cache by downloading the skins, avatar and lightmaps that everybody already has. (the ones given with the game)
- other little bug fixes (like the server taking hours to launch if there are many autosaved replays.)
- crash when a tcp connection is broken under certain rare conditions.
- server restart no longer wipes the guestlist.txt and blacklist.txt
- ForceSpectatorTarget() now works with cameratype = 2 and no target
- Cup Mode: the game no longers loops when there's a winner and everybody else is disconected or in spectator mode.
- use fallback timer for FreeBSD
- detailed playerinfos work for the server login
- new functions SetBuddyNotification() GetBuddyNotification() to be able to disable the chat message. (either per player or for all the players on the server)
A certain amount of people have been plagued by crashes of the dedicated server on their linux OS since the release of tmu-f.
It's a crash which is triggered somehow. I'll share my experience and we can pit this against all the other experiences and formulate a decision.
Allthought I "think" it is triggered by some events, this topic is needed because the crash is very evasive and isn't a direct result of server operation by the user but might be caused by records script events. Let's hope that there's a singular action you can take to make the server crash. But we also need to face the posibility that the crash happens in several phases.
There might be 2 or more criteria needed to let the server crash.
So let's put us all in the right direction at least.
If you participate in this topic, please make sure you are using 2008-05-19. As some fixes were done to evade other crashes that are not relevant to this issue. If you are using an older version, you might disrupt this poll cause the crashes may not be related to this particular issue.
so only test with 2008-05-19 please, if you want to participate.
You might use these versions and have used your "old" configuration files which is missing some data. It can be significant for this test or it may not.. We'll have to see the result.
So the things you might consider posting feedback on is:
Crashes regularly or no crash ?
Windows or linux
Do you run with old config files (2008-04-15 etc)
Do you start the server with old matchsetting files ?
Is your server having 60+ users regularly ?
Tell a bit about your config, do you autosave replays ?
Do you use different than standard upload/download rates etc..
Do you or don't you run server scripts ? In case you do:
Do you run Xaseco
Do you run Onyx 2
Do you run Fast 3
Do you run Aseco 2
Do you run aseco 1.2 ?
It's a vast examination I know, and even i'm not sure this is the right direction, other angles may be discussed in this topic too ofcourse.
I'm no programming guru, nor do I have intimate knowledge of the inner workings of the server like xbx does. Still i'd like to also bring the discussion forward on incomplete array data (for callbacks or configuration servers) whether they might or might not cause the server to behave badly and crash. That's why I'm asking you if you're using older scripts that were designed for TMU.
And that's why I'm also calling out to the script coders to tell if they have followed the new methods standards, and well.. I'd also like to ask if users of "new" scripts still have troubles.
If it's really internel, network tripping or other internal stuff of the dedicated beyond our immediate control, then this topic may be quite useless and it will be just a case for Nadeo to find this. But at least we've tried.
So here we go, I hope we can have a little of that beta test spirit out in the public
It seems that lots of the crashes are already resolved by the latest server, but you never know what you may find so this topic still has it's value.