ManiaPlanet Server 2011-10-12

[_bb_mort]

Are there any plans on releasing an update for the dedicated server any time soon..? It's been over 3 months since the games launch, and nearly 3 months since any acknowledgement from a Nadeo employee.. and we still have a barely-functional resource hogging-mess..

It was forgivable enough in TMF, but it's hard convincing players to purchase the new game when I can't really tell them that TM2 is anywhere close to being entirely 'ready'. Not to mention, I'm purchasing a new hosting plan in the next few days, and really need to know whether I'm going to have to spend all of my money on RAM, lacking any fix to the linux dedicated server software.

I'm about to buy a new server with 64GB RAM. I'm hoping I'll be able to run *three* TM2 servers then but maybe that still won't be enough

It's not just Linux, the Windows server is just as resource hungry. Still can only run 2 TM2 servers where I used to run 6 or 7 TMUF servers.

[nocturne]

That's not quite my experience.. A basic server with 5 tracks in TA mode on RHEL racks up almost 300mb right off the bat, while the identical server setup on windows takes up less than 70mb of RAM. Also, the problem is limited it seems to only TA mode, so it seems obvious that somebody (*cough*xbx*cough*) just made an incredible screw-up somewhere.

And Jojo.. I don't exactly waste money on TM hosting.. I've ran my own TM community for years, which prior to the freezone fiasco had over a quarter million unique players and over 10,000 regulars; all offered only for the sake of free and fun racing to all players, free of any advertising, sponsorships, and any naggings about 'premium' services -- all paid for right out of my own pockets (along with literally thousands of hours of my personal time).

I wouldn't be so incredibly angry if at least one Nadeo employee (but not you SM guys, sorry) acknowledged any of these problems and made the slightest attempt to convince us that they are being addressed. Instead, we've been outright ignored for over 3 months with no assurance other than a few mods and SM devs saying "well, I'm sure they'll get around to it sooner or later..".

I feel like it should be obvious, but the Nadeo devs should certainly be informed that it's entirely possible that prospective buyers may be checking out these forums prior to buying this game, in which they can clearly see a resounding message blatantly stating 'STAY AWAY FROM THIS GAME!'.

[Mikey]

[BUG REPORT]
I strongly suspect the callback 'ReleaseSpectatorSlot' does not release the slot at all
Some testing indicated the slot is released as a player leaves the server, but not when he goes spectator and the callback is made.
PLEASE CONFRIM and fix

[The_Big_Boo]

I've just tested it again (through ManiaLive) and it works. What are your tests which makes you think it doesn't work ?

[Mikey]

Using Xaseco, and creating a plugin. With simple testing using maximum players set to 2. and having some additional spectators.
If the plugin ForceToSpectator,PlayerA,1 then ReleaseSpectatorSlot,PlayerA
The Spectator PlayerC still cannot join play when ForceToSpectator,PlayerC,2 is used, even though there is only one player (PlayerB), with A and C both spectating
But the same callback issued to PlayerA does work ForceToSpectator,PlayerA,2
It is as if the PlayerA slot is not released, so PlayerC cannot be forced to play as the server is at maximum players as the slot is not released, but PlayerA can be forced back.

[The_Big_Boo]

Note that calling SpectatorReleasePlayerSlot too soon after ForceSpectator may fail because the player isn't actually a spectator yet. After calling ForceSpectator, you should wait the PlayerInfoChanged callback from the server to call SpectatorReleasePlayerSlot.

[Mikey]

Thanks for the info ! I will investiage further and report back only if problems continue.

[undef.de]

While some tests i figure out that in the ManiaPlanet Server the Callback TrackMania.StatusChanged has been changed:

TMF
TrackMania.StatusChanged(5, Running - Finish)

TM2
TrackMania.StatusChanged(6, Running - Exit)

The state "5, Running - Finish" was never called in my tests. Is this a bug or new behavior?

[The_Big_Boo]

New behaviour, status 5 doesn't exist anymore (as far as I know).

[om23]

maybe one day xbx learn to make a notice in his old thread after he opened a new one with new dedicated verison,
but untill he lerned it, there must come some one and do it for him/us/you.

here is the new dedicated