Re: [hlds_linux] Team Fortress 2 Update

2010-02-04 Thread Midnight
Yes.  Someone just reported his server is insecure.  However it was CS 1.6

Probably unrelated.

David Banham wrote:
 Is anyone else having issues getting a VAC connection after the update?

 Adding master server 68.142.72.250:27011
 Adding master server 72.165.61.151:27011
 L 02/05/2010 - 14:57:03: server_cvar: sm_nextmap cp_granary
 Could not establish connection to Steam servers.


 On 5 February 2010 10:24, Jason Ruymen jas...@valvesoftware.com wrote:

   
 Team Fortress 2 has been updated.  This is an optional update, as it just
 contains client fixes.  The specific changes are:

 - Fixed a client crash during the localization of item names
 - Fixed a client crash caused by selecting Random in the class menu and
 then clicking on Edit Loadout

 Jason

 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

 
 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Left 4 Dead 2 Update Coming

2009-11-21 Thread Midnight
Nice.

Jason Ruymen wrote:
 A required update for Left 4 Dead 2 is on the way.  It should be live in 
 about 20 minutes.

 Jason

 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] cs s parameter

2009-11-10 Thread Midnight
so many ways to screw up a command line lol

Guy Watkins wrote:
 This works for me:
 ./steam -command update -game Counter-Strike Source -dir .

 } -Original Message-
 } From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux-
 } boun...@list.valvesoftware.com] On Behalf Of tom nikitopoulos
 } Sent: Tuesday, November 10, 2009 9:26 PM
 } To: Half-Life dedicated Linux server mailing list
 } Subject: Re: [hlds_linux] cs s parameter
 } 
 } thanks,but i have tried that. i have even tried to use parenthesis but it
 } has not worked either
 } 
 } command line off the top of head
 } ./steam -command update -game Counter Strike-Source -dir
 } 
 } maybe something is messed up on my end.
 } 
 } On Tue, Nov 10, 2009 at 1:54 PM, Nicholas Hastings
 } nshasti...@gmail.comwrote:
 } 
 }  -game Counter-Strike Source
 } 
 }  (you can see all of these by using list for the game)
 } 
 }  tom nikitopoulos wrote:
 }   can anyone tell me the parameters for counter strike source.
 }   i have tried to download the css files using the parameter -game
 } Counter
 }   Strike - Source and it is not working
 }   thanks
 }   ___
 }   To unsubscribe, edit your list preferences, or view the list archives,
 }  please visit:
 }   http://list.valvesoftware.com/mailman/listinfo/hlds_linux
 }  
 }  
 } 
 } 
 }  ___
 }  To unsubscribe, edit your list preferences, or view the list archives,
 }  please visit:
 }  http://list.valvesoftware.com/mailman/listinfo/hlds_linux
 } 
 } ___
 } To unsubscribe, edit your list preferences, or view the list archives,
 } please visit:
 } http://list.valvesoftware.com/mailman/listinfo/hlds_linux


 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Modern Warfare 2

2009-11-08 Thread Midnight
Dedicated servers are needed and serve a purpose, otherwise everyone 
would just host from home and not pay for servers.

One main factor that hasn't been mentioned is the fact that GSP's 
provide 5-15 major backbone connections directly to their server which 
means each player has a good chance of hopping networks only 1 time.  
Whereas a home PC has 1 backbone which is their ISP network, you may 
have to change across several networks to get to the ISP network of the 
P2P host.  Many ISP networks, especially Comcast provide cheapest path 
routing and do not provide lowest latency routing like the GSP's do.

Maybe even a bigger problem is you can't just hop on your favorite local 
/ clan server and have your friends / teammates show up whenever.  You 
would need to plan a game beforehand and get them to connect with you at 
the same time, what a drag.  How would you even organize competitive 
match play in a game like MW2?  Talk about a nightmare!



Steven Hartland wrote:
 I would say its not not everywhere but actually 90% of which don't
 have home connections capable of this, that said with the BW you stated
 it still not gonna cut it for servers that most people want to run e.g.
 large high rate servers.

 Regards
 Steve

 - Original Message - 
 From: Arg! chillic...@gmail.com


 Im certainly not saying you cant do it, im simply saying not everywhere has
 internet available capable of doing it, no matter how much you fiddle with
 the rates.


 
 This e.mail is private and confidential between Multiplay (UK) Ltd. and the 
 person or entity to whom it is addressed. In the event of misdirection, the 
 recipient is prohibited from using, copying, printing or otherwise 
 disseminating it or any information contained in it. 

 In the event of misdirection, illegible or incomplete transmission please 
 telephone +44 845 868 1337
 or return the E.mail to postmas...@multiplay.co.uk.


 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Shameless IRC Chat Advertisement

2009-10-30 Thread Midnight
Nice.  Now I have a live support area to send people to for all the L4D 
sourcemod crash bugs.


Crazy Canucks wrote:
 For shame Neph!  You, of all people!  Oh the unadulterated spam!  Woe! 
 Woe!  ;)

 Happy Halloween everyone!

 Drek

 Nephyrin Zey wrote:
   
 Just to remind everyone, a lot of us run a pretty popular srcds 
 administration chat channel on gamesurge (linux and windows, everyone 
 welcome)

 irc.gamesurge.net #hlserveradmins

 Gamesurge is also home to:
 #sourcemod
 #sourceop (the tf2items.com guy)
 #sourcebans
 #hlcoders

 and more. So join up!

   - Neph

 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
 


 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] A2S_INFO response from L4D2

2009-10-28 Thread Midnight
4ø�...@

There it is in text, so probably nothing you can decipher.


David A. Parker wrote:
 Just wanted to mention that I ran a packet capture, and there are 8 
 bytes in the A2S_INFO packet between the port number and tag data which 
 aren't there in the response from L4D, TF2, etc.  I can't make heads or 
 tails of what the bytes are supposed to be, but they are always the same 
 from my server:

 0x03 0x34 0xF8 0xCC 0x9A 0x01 0x40 0x01

  - Dave

 David A. Parker wrote:
   
 Hello,

 The issue with the weird data in the A2S_INFO response is still present 
 in the last update (l4d2 demo linux dedicated server version 2).  It 
 really looks like some sort of data has been added before the tags field 
 in the response packet, but that's just a guess.

  - Dave

 ics wrote:
 
 Well naturally i expect no less than 1-2 updates to dedicated servers 
 before the demo actually goes live and many more updates during the 
 demo, just like last time with l4d However, having lack of features now 
 either is a sign on heavy optimization to the game engine and thus 
 ripping off everything that is not needed or just lack of implementation 
 at the dedicated servers part.I guess we will find out during the demo.

 (And this is pointed to Valve) Last time there were heads up for us, 
 with new features and such explained here on the list. Is there anything 
 we need to know about running the game on our ded servers? Changes to 
 the lobby and how they affect to our settings? Milton said that if game 
 settings are modified, then we're out of luck but what else?

 -ics

 Richard Eid kirjoitti:
   
 We're talking about a demo.  Some features are going to be limited while
 others just plain won't work.  Not to mention that the early-access demo 
 has
 been delayed so Valve can take care of some bugs.

 Until the full game is released, I'd expect that a big set of features 
 might
 be unavailable.  Plus, being that there is a delay, be ready to run
 hldsupdatetool again before the demo goes live.

 -Richard Eid


 On Wed, Oct 28, 2009 at 11:54 AM, ics i...@ics-base.net wrote:

   
 
 I also have an script running that provides server status to the irc,
 players, ip, name, ping, map and game name. Currently it fails with L4D2
 servers so something definitely has changed. No reply.

 -ics

 David A. Parker kirjoitti:
 
   
 Hello,

 Was something new added to the A2S_INFO response in L4D2?  I have a Perl
 script which fetches server info, and it works correctly for other games
 but I get some random data in with the server tags on my L4D2 servers,
 almost like there's a new field in the packet before the tag data.

 You can see it here:


   
 
 http://www.utica.edu/academic/iits/computingservices/networking/resnet/gaming/status.cfm
 
   
  Thanks,
  Dave


   
 
 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

 
   
 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux
   
 
 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Updating servers RANT

2009-10-24 Thread Midnight
Yes it's annoying that this happens during preloads and stuff, but do 
you really expect them to put out enough servers to handle the 1% case 
when things get swamped?

Maybe what is needed is a smarter system.  Set the downloads for 
preloads to only use 1/2 of the servers, or restrict the bandwidth.  
Preloads do not need to be done ASAP, they can take a few days, so why 
have them choke the whole system?  Plus when the system is swamped the 
preloads probably go slow anyway so there would be little difference for 
those downloads.



Saul Rennison wrote:
 It's absolutely pathetic, I know. It makes me feel sad that VALVe have  
 a fuck load of money (even Gabe said this in a press conf. a few weeks  
 ago), but can't be added adding servers.

 Thanks,
 - Saul.

 On 24 Oct 2009, at 17:50, Saint K. sai...@specialattack.net wrote:

   
 Yes, tried all the regular stuff.

 Servers are hosted in a datacenter, client doesn't get FW'ed.  
 Servers are connected 100Mbit almost directly on the Amsterdam  
 Internet Exchange (couldn't get much better than that).

 Perhaps this is an Europe issue or whatever, but its surely driving  
 us insane.

 Saint K.

 -Original Message-
 From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- 
 boun...@list.valvesoftware.com] On Behalf Of ics
 Sent: Saturday, October 24, 2009 6:52 PM
 To: Half-Life dedicated Linux server mailing list
 Subject: Re: [hlds_linux] Updating servers RANT

 Sometimes content servers go out of order so to speak and the  
 updatetool
 just tries and tries. You did deleted the ClientRegistry.blob from
 /home/user/.steam folder? You sure you got all the necessary ports
 in/out open and not firewalled shut?

 -ics

 Saint K. kirjoitti:
 
 Seriously, this is taking the piss.

 I am getting extremely pissed off and annoyed by that fucking  
 steamupdatetool clients which DOESN'T UPDATE FOR SHIT

 Takes me several god damn hours before the fucking piece of piss  
 finally decided to start downloading 1 bloody file for half an hour.

 We asked, we requested, WE BLOODY BAGGED for a proper update  
 content server to have our servers updated against. But fuck hey,  
 why should you bloody reply?

 /end of rant.

 ___
 To unsubscribe, edit your list preferences, or view the list  
 archives, please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
 ___
 To unsubscribe, edit your list preferences, or view the list  
 archives, please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

 No virus found in this incoming message.
 Checked by AVG - www.avg.com
 Version: 8.5.423 / Virus Database: 270.14.26/2451 - Release Date:  
 10/24/09 14:31:00

 ___
 To unsubscribe, edit your list preferences, or view the list  
 archives, please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux
 

 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] L4D2 server requirements

2009-10-22 Thread Midnight
I'm really surprised by these numbers.  I have some pretty finicky 
players who play competitively.  I don't know for sure if it is in their 
head or not, but they claim they get lag with 1 game active on a whole 
quad 9550 box which I find hard to believe so I tend to take it with a 
grain of salt.

I would never dream of running 32 games per quad CPU, that is insane 
density imo.  No offense but I find it amazing that you are getting any 
decent performance with the CPU's nearly maxed out, maybe you have some 
secret sauce in your system :)  I get complaints after 25% CPU load on 
the box so we run all of ours under that at all times, which means 3-4 
L4D's per box.  They are using 6-9% cpu per game = up to about 40% of a 
core on newer hardware.  This could be from tweaking done to the game 
settings.



Nick Turner wrote:
 We're running 8 servers per core on 5450s.  64 forks per 2 socket server.  
 16GB is more than enough on Linux.

 We've had an entirely server almost full, the CPUs are nearly maxed out but 
 we didn't get any performance complaints.


 -Original Message-
 From: hlds_linux-boun...@list.valvesoftware.com 
 [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Milton Ngan
 Sent: Thursday, October 22, 2009 10:39 AM
 To: Half-Life dedicated Linux server mailing list
 Subject: Re: [hlds_linux] L4D2 server requirements

 CPU shouldn't be much of an issue. Each game takes on average 10-15% of a 
 core. There are obviously spikes that go higher. 8 player games don't 
 actually take much more CPU than a 1 player game because the server is doing 
 less AI work.

 On these servers, I have between 200-300 people playing at a time. This is 
 very similar to the player numbers we see when running TF servers, except we 
 run far fewer TF instances on the same hardware.

 Memory is still an important factor, because if you run out of RAM, then you 
 will swap and your games will hitch due to I/O latency.

 M.
 
 From: hlds_linux-boun...@list.valvesoftware.com 
 [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Midnight 
 [mido...@gmail.com]
 Sent: Wednesday, October 21, 2009 6:26 PM
 To: Half-Life dedicated Linux server mailing list
 Subject: Re: [hlds_linux] L4D2 server requirements

 What I'm asking is not how many can you spawn in a given amount of
 memory, but how many can be actively running with players in them
 without lagging?  I find that I run out of CPU way before memory since
 the game is a CPU hog.

 Surely you can't run 50 servers with players in them all at the same
 time right?  So how many can you run at the same time?  How much cpu
 does a full 8 player server use?

 Thanks for your input.


 Milton Ngan wrote:
   
 I can support 50 instances of L4D1 on a dual 2.5GHz Quad Core system. So if 
 you only have one, then half that. That being said, I need at least 10GB of 
 RAM on Linux to achieve this.

 Under Windows, there is no forked mode, so each instance will take up more 
 memory resources. So I would estimate around 16GB would be required to do 
 the same thing. On a 4GB Windows 2003 system we were supporting around 13 
 instances quite happily.

 So depending on how much memory you have, it could be that your limit is 
 memory and not CPU.

 M.

 -Original Message-
 From: hlds_linux-boun...@list.valvesoftware.com 
 [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Midnight
 Sent: Wednesday, October 21, 2009 5:19 PM
 To: Half-Life dedicated Linux server mailing list
 Subject: Re: [hlds_linux] L4D2 server requirements

 How many ACTIVE L4D servers can you guys run on a 2.5 Ghz Quad Xeon
 1333MHz?  Seems that I can only run 3-4 per box without people
 complaining about lag, that is less than 1 per core.  I see each server
 using around 25-35% CPU of a core and 6-9 of the whole box.  I know this
 is the linux list but I'm running Windows 2008, but I figure it should
 be basically the same.


 Craig H wrote:

 
 It'll probably take a little more processing power per fork, I don't see it
 as being that substantial of an increase. I'd be amazed if it is really
 worth noting.
 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux



   
 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux


 
 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

Re: [hlds_linux] L4D2 server requirements

2009-10-22 Thread Midnight
I'm running Intel Core 2 Quad /Q9550/ 2.83GHz 12MB L2 Cache 1333MHz 
Front side Bus speed, which should be pretty close to a E5450, though 
maybe not quite.  Yet I can't run anything close to 32 active servers on 
this. 4-5 tops.  Not sure why, just saying.


Eric Riemers wrote:
 True, but 320/5/4 is still 16 servers.. and i stop at 5, so basicly i can
 have 4 in vs mode (32 players) and thats my max.

 Just talking out loud, but i presume this is normal when they are active
 which is almost all the time. (i have people on it at the strangest times)
 but i have to say i have a lot of custom content (vpk files) on there,
 which attracts a lot of people too.

 Dont know if those maps have impact on cpu, death aboard is i think around
 550mb..

 O well, maybe time for a upgrade too :)

 On Thu, 22 Oct 2009 17:33:55 +0800, Ben Jensz fh-l...@3fl.net wrote:
   
 E5450 != 3Ghz classic Xeons.

 A single core (of which there are 4) on an E5450 is more powerful than a
 

   
 3Ghz Xeon CPU.



 Eric Riemers wrote:
 
 I have a dual xeon 3ghz, i had 6 forks running on it, but with 5x 8
 people
 in versus mode people
 started to complain and i indeed could see that the cpu's where maxed
 out.

 Running on linux debian, but if i hear these story's then it should be
 able to do more no?

 On Thu, 22 Oct 2009 14:54:53 +0800, Ben Jensz fh-l...@3fl.net wrote:
   
   
 The night that Nick is referring to was when the Survival pack was 
 released.  We were running 320 instances of Left 4 Dead across 5 IBM 
 HS21XM blades (Dual E5450s - 3Ghz each core) and the majority of the 
 servers had players on them.  All of the cores were running at 90+%
 
 CPU 
   
 usage for a few hours.  Each physical server was using around 11-12Gb
 
 of
   
 
 
   
   
 RAM for that.

 We weren't doing anything special configuration wise, Left 4 Dead was 
 running in a default manner.  Nothing special done OS wise, just a
 
 stock
   
 
 
   
   
 install of CentOS.  I played on one of our servers that night and I 
 didn't notice anything in-game that indicated performance issues.


 Midnight wrote:
 
 
 I'm really surprised by these numbers.  I have some pretty finicky 
 players who play competitively.  I don't know for sure if it is in
   
   
 their 
   
   
 head or not, but they claim they get lag with 1 game active on a
   
 whole 
   
 quad 9550 box which I find hard to believe so I tend to take it with
   
 a 
   
 grain of salt.

 I would never dream of running 32 games per quad CPU, that is insane 
 density imo.  No offense but I find it amazing that you are getting
   
 any
   
   
   
   
   
 decent performance with the CPU's nearly maxed out, maybe you have
   
 some
   
   
   
   
   
 secret sauce in your system :)  I get complaints after 25% CPU load
   
 on 
   
 the box so we run all of ours under that at all times, which means
   
 3-4 
   
 L4D's per box.  They are using 6-9% cpu per game = up to about 40% of
   
 a
   
   
   
   
   
 core on newer hardware.  This could be from tweaking done to the game
   

   
 settings.



 Nick Turner wrote:
   
   
   
 We're running 8 servers per core on 5450s.  64 forks per 2 socket
 server.  16GB is more than enough on Linux.

 We've had an entirely server almost full, the CPUs are nearly maxed
 
 
 out
   
   
 but we didn't get any performance complaints.



 
 
 
 
 
 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux
 

 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] L4D2 server requirements

2009-10-21 Thread Midnight
How many ACTIVE L4D servers can you guys run on a 2.5 Ghz Quad Xeon 
1333MHz?  Seems that I can only run 3-4 per box without people 
complaining about lag, that is less than 1 per core.  I see each server 
using around 25-35% CPU of a core and 6-9 of the whole box.  I know this 
is the linux list but I'm running Windows 2008, but I figure it should 
be basically the same.


Craig H wrote:
 It'll probably take a little more processing power per fork, I don't see it
 as being that substantial of an increase. I'd be amazed if it is really
 worth noting.
 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] L4D2 server requirements

2009-10-21 Thread Midnight
What I'm asking is not how many can you spawn in a given amount of 
memory, but how many can be actively running with players in them 
without lagging?  I find that I run out of CPU way before memory since 
the game is a CPU hog.

Surely you can't run 50 servers with players in them all at the same 
time right?  So how many can you run at the same time?  How much cpu 
does a full 8 player server use?

Thanks for your input.


Milton Ngan wrote:
 I can support 50 instances of L4D1 on a dual 2.5GHz Quad Core system. So if 
 you only have one, then half that. That being said, I need at least 10GB of 
 RAM on Linux to achieve this. 

 Under Windows, there is no forked mode, so each instance will take up more 
 memory resources. So I would estimate around 16GB would be required to do the 
 same thing. On a 4GB Windows 2003 system we were supporting around 13 
 instances quite happily. 

 So depending on how much memory you have, it could be that your limit is 
 memory and not CPU. 

 M.

 -Original Message-
 From: hlds_linux-boun...@list.valvesoftware.com 
 [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Midnight
 Sent: Wednesday, October 21, 2009 5:19 PM
 To: Half-Life dedicated Linux server mailing list
 Subject: Re: [hlds_linux] L4D2 server requirements

 How many ACTIVE L4D servers can you guys run on a 2.5 Ghz Quad Xeon 
 1333MHz?  Seems that I can only run 3-4 per box without people 
 complaining about lag, that is less than 1 per core.  I see each server 
 using around 25-35% CPU of a core and 6-9 of the whole box.  I know this 
 is the linux list but I'm running Windows 2008, but I figure it should 
 be basically the same.


 Craig H wrote:
   
 It'll probably take a little more processing power per fork, I don't see it
 as being that substantial of an increase. I'd be amazed if it is really
 worth noting.
 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
 
 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] L4D - At my wits end with these crashes

2009-10-15 Thread Midnight


Russell Jones wrote:
  It 
 will crash 4 or 5 times in a row and then not crash. It's frustrating.


   
Welcome to L4D party.  I've spent countless hours troubleshooting 
retarded issues like this.  The game has a history of not behaving well 
at all with sourcemod, and especially with the competitive plugins 
like Readyup mod.  Everytime an update comes out something is broke 
again.  Hope you are masochistic.

Next thing you know the players will start blaming your crappy server 
hardware and swear that nobody else has these problems.

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Very unstable FPS in hlds

2009-09-09 Thread Midnight
I've never ran a server on anything that low end, quite possibly you 
need a faster server.  Just because the CPU is not maxed out doesn't 
mean the server isn't struggling to keep up.  Even on the latest 
hardware you won't get stable fps under load.



Nevermore wrote:
 I talked with the datacenter.
 It is a Pentium 4 with hiperthreading.
 Sory about the confusion

 Now, anybody have any clue about my fps problem?

 El mié, 09-09-2009 a las 19:43 -0400, Nicholas Hastings escribió:
   
 Yes, they were called Pentium D unless you're just talking about 
 hyperthreading

 Saul Rennison wrote:
 
 I'm sure dual-core P4's exist...

 Thanks,
 - Saul.

 On 9 Sep 2009, at 22:11, 1nsane 1nsane...@gmail.com wrote:

   
   
 Pentium D you mean?

 On Wed, Sep 9, 2009 at 3:19 PM, Rick Payton r...@mai-hawaii.com  
 wrote:

 
 
 It might actually be a newer dual core P4, I got a 3.0GHz dual core  
 model
 here at work ...

 --mauirixxx

 -Original Message-
 From: hlds_linux-boun...@list.valvesoftware.com [mailto:
 hlds_linux-boun...@list.valvesoftware.com] On Behalf Of ics
 Sent: Wednesday, September 09, 2009 8:41 AM
 To: Half-Life dedicated Linux server mailing list
 Subject: Re: [hlds_linux] Very unstable FPS in hlds

 That looks like P4 with Hyperthreading on it, not actual dual core  
 cpu.
 What does cat /proc/cpuinfo say?

 -ics

 Dave Williams kirjoitti:
   
   
 Now, i would suggest that your chip *could* be being identified
 
 
 incorrectly. If this is the case it would be the kernel that is  
 most likely
 at fault. Do you know if you have symmetric multi processing  
 support enabled
 in your kernel? Also, what is the OS that you are using?
   
   
 -Original Message-
 From: hlds_linux-boun...@list.valvesoftware.com [mailto:
 
 
 hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Nevermore
   
   
 Sent: 09 September 2009 17:08
 To: Half-Life dedicated Linux server mailing list
 Subject: Re: [hlds_linux] Very unstable FPS in hlds

 It's a dual core intel.

 dmesg says:
 Intel(R) Pentium(R) 4 CPU 3.00GHz stepping 03

 I am not sure about the model.

 Thank you

 El mié, 09-09-2009 a las 11:42 -0400, Eric Greer escribió:

 
 
 What CPU do you have in it?

 Eric



 On Wed, Sep 9, 2009 at 11:32 AM, Nevermoreneverm...@pheek.net  
 wrote:

   
   
 The topic about 1000fps servers made me take a look at my  
 server's fps.
 I am seriously worried, because i have very unstable fps.

 Running HLDS with sys_ticrate 350, fps_max 0 and -pinbgoost 2:

 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 0.50  0.00  0.00   6 0  103.98   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 1.00  0.00  0.00   6 0  214.50   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 1.20  0.00  0.00   6 0  136.95   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 1.17  0.00  0.00   6 0  123.05   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 2.00  0.00  0.00   6 0  141.34   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 2.00  0.00  0.00   6 0  112.69   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 2.00  0.00  0.00   6 0  302.02   0

 Same configuration in a windows box:

 CPU   InOut   Uptime  Users   FPSPlayers
 1.56  0.00  0.002651   978  250.84   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 1.56  0.00  0.002651   978  233.92   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 3.13  0.00  0.002651   978  252.14   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 3.13  0.00  0.002651   978  231.70   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 3.13  0.00  0.002651   978  238.05   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 3.13  0.00  0.002651   978  236.71   0
 stats
 CPU   InOut   Uptime  Users   FPSPlayers
 3.13  0.00  0.002651   978  240.79   0

 CPU usage of the box is at 23%, so this is not the problem.
 I am using debian etch kernel 2.6.18-amd64
 I tried recompiling for 1000hz and nothing changed.

 I dont want to change kernel version, because if i break the  
 system i
 have not phisical access to the box to fix it.

 Thank you


 ___
 To unsubscribe, edit your list preferences, or view the list  
 archives,
 
 
 please visit:
   
   
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux


 
 
 ___
 To unsubscribe, edit your list preferences, or view the list  
 archives,
 
 
 please visit:
   
   
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux
 

Re: [hlds_linux] FPS Drops

2009-08-30 Thread Midnight
You are likely running too many slots for that old CPU. I think it can't 
handle that many players.



Ulrich Block wrote:
 Sounds clear. But i also tried only one singel Server on the machine. 32 
 slots tick 66 with standart fps_max 300. I guess a AMD X2 5800+ with 4 
 GB of RAM should be enough for one server. I watched the same issue. 
 Your Answer also do not explain why i can have a full 12 slot server 
 with tick 100 and high fps and getting no drops even the high slot 
 sevrer is full and getting the issue

 Adam Nowacki schrieb:
   
 Ulrich Block wrote:
   
 
 I am running multiple dod:s servers on one machine at the same machine. 
 When they are full there is still free cpu time and power left. But i am 
 watching FPS drops on a srcds server when it reaches 20 player slots or 
 more. The other running services are not affected at the same time. A 
 server with 12 slots at a tickrate with 100 and stable 950 fps for 
 example is not affected. Even if the other one has its fps drops.
 I am running debian lenny and tried bigmem, amd and custom kernels with 
 rt patch. With all kernels the same problem.
 Anyone an idea what the cause can be? I already tried running dods 
 vanilla. Same problem :(
 
   
 Source server is using cpu time in short bursts simulating game tick and 
 then sitting idle waiting for next tick. This idle time lowers the 
 reported average cpu use making you think that you have still a lot of 
 cpu power available. Solution: run less servers or get a better cpu (or 
 two ;).

 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
 


 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] srcds virtualized

2009-08-28 Thread Midnight
Hmm, I'm not sure how you are running stable servers on a platform under 
medium to heavy load from other VMs.  I guess if you are running low end 
servers it would be ok, but most people these days seem to want 1000fps 
stable servers.  Anyway seems you are very happy with your XEN setup for 
what you are using it for so to each his own.


Valtteri Kiviniemi wrote:
 Hi,

 Disk I/O is the most important thing on virtualized environments. You 
 have multiple operating systems using the same storage so its important 
 that your storage is fast. If you have slow storage, then your virtual 
 servers are sticky and slow because of the slow disk speed. You might 
 have fast cpu and a lots of ram, but you have to remember that you have 
 multiple operating systems with different workloads using the same 
 storage and that will need to be fast if you want to run a reasonable 
 ammount of virtualservers. Ofc. im not directly speaking on srcds's 
 hosting, so it might mabe a little offtopic, but im just making out the 
 differences between VMWare and Xen.

 I had almos 200MB/s of raw disk troughput with Xen and only 60MB/s with 
 VMWare measured from inside the VPS. That is a huge difference. That was 
 tested via Areca raid array, but I also tested the disk speed with no 
 raid, only single sata-disk and it was still better on Xen.

 - Valtteri Kiviniemi

 Midnight kirjoitti:
   
 Disk I/O is not the main factor for running game servers anyway, so 
 that's not really a reason to choose one option over the other in this case.



 Valtteri Kiviniemi wrote:
 
 Hi,

 You are correct. But I'm just saying my opinion here, and I think that 
 Xen is better.

 VMWare ESXi is maybe a bit more user friendly than XenServer 5.5, but I 
 don't still understand why ESXi is so much slower. I'am using both of 
 them because my company sell's virtual servers and some customers want 
 VMWare ones.

 I have identical hardware on all machines but im still seeing 30-40% 
 more performance on Xen virtual servers than on VMWare. Dont know why, 
 but disk i/o is way better on Xen than VMWare.

 - Valtteri Kiviniemi

 Eric Greer kirjoitti:
   
   
 If everyone wants to get technical with all of this nonsense... you can run
 srcds just fine on a VPS - as long as there is enough power.
 Xen Quite simply adds another layer hardware layer that data must pass
 through.  However, we're talking nanoseconds here people.  Not like another
 hop on your way to chicago - another *virtual* device on the way to the
 hardware and back.  It's like nothing.  VMWare ESXi adds a few more layers
 as it passes through more virtual devices... but it still does not matter.

 A VM can be provisioned with plenty enough power to do any source server
 just fine. You just have to give it plenty of dedicated resources.

 I feel like people start taking emotions into computing at some point.
  There aren't any - its all benchmarks and numbers.  If the system can CPU
 bench some number has memory available and bandwidth... it can run the
 server - simple as that.

 A VPS is generally considered 'weaker' because it can share resources with
 other VMs - but it doesn't have to.  If for some reason you wanted to give
 root shell access to a game server customer, you could VM them.  Yes, 
 theres
 a good 100Mb of memory overhead for the hypervisor, but it can be worth it.

 Eric


 On Thu, Aug 27, 2009 at 12:05 PM, Valtteri Kiviniemi 
 valtteri.kivini...@dataproof.fi wrote:

 
 
 Hi,

 You should probably read the facts before posting. Ofc. its not exactly
 the same, but if you know nothing about Xen you would know that the
 performance difference between (for example 2.6.18-xen and 2.6.18
 kernels) are so small, that you cant even notice it.

 Maybe with ESXi you have greater performance difference compared to
 bare-metl but not with xen.

 - Valtteri Kiviniemi

 Kveri kirjoitti:
   
   
 believe me, if you have paravirtualized enviroment you don't have
 equal performance than on bare-metal. Paravirtualization adds another
 layer, so does overhead. Maybe performance in CSS, but I doubt about it.

 I'm using full VT on 4x quad core xeons with 16gb ram and providing
 1000fps 1.6 servers (yes, stable 1000fps, kernel self-pached with RT
 and some HZ tweaks), CSS servers with 100 ticrate and and some tf2
 servers without any problems.

 Kveri

 On 25.8.2009, at 20:52, Valtteri Kiviniemi wrote:

 
 
 Hi,

 We are running multiple TF2 servers with Xen 3.4.1 paravirtualized.
 Performance is exactly the same as bare-metal, maybe even better. Only
 downside is that you need xen-patched kernel so to get most stable and
 working environment you have to use the default 2.6.18.8-xen kernel.
 Ofc. you can compile a 1000hz domU kernel like we have.

 There is also pv_ops kernels which are included in the xen-unstable
 tree. They are the normal kernel.org kernel with patches that make it
 suitable for Xen hypervisor.

 In my opinion Xen

Re: [hlds_linux] srcds virtualized

2009-08-27 Thread Midnight
Disk I/O is not the main factor for running game servers anyway, so 
that's not really a reason to choose one option over the other in this case.



Valtteri Kiviniemi wrote:
 Hi,

 You are correct. But I'm just saying my opinion here, and I think that 
 Xen is better.

 VMWare ESXi is maybe a bit more user friendly than XenServer 5.5, but I 
 don't still understand why ESXi is so much slower. I'am using both of 
 them because my company sell's virtual servers and some customers want 
 VMWare ones.

 I have identical hardware on all machines but im still seeing 30-40% 
 more performance on Xen virtual servers than on VMWare. Dont know why, 
 but disk i/o is way better on Xen than VMWare.

 - Valtteri Kiviniemi

 Eric Greer kirjoitti:
   
 If everyone wants to get technical with all of this nonsense... you can run
 srcds just fine on a VPS - as long as there is enough power.
 Xen Quite simply adds another layer hardware layer that data must pass
 through.  However, we're talking nanoseconds here people.  Not like another
 hop on your way to chicago - another *virtual* device on the way to the
 hardware and back.  It's like nothing.  VMWare ESXi adds a few more layers
 as it passes through more virtual devices... but it still does not matter.

 A VM can be provisioned with plenty enough power to do any source server
 just fine. You just have to give it plenty of dedicated resources.

 I feel like people start taking emotions into computing at some point.
  There aren't any - its all benchmarks and numbers.  If the system can CPU
 bench some number has memory available and bandwidth... it can run the
 server - simple as that.

 A VPS is generally considered 'weaker' because it can share resources with
 other VMs - but it doesn't have to.  If for some reason you wanted to give
 root shell access to a game server customer, you could VM them.  Yes, theres
 a good 100Mb of memory overhead for the hypervisor, but it can be worth it.

 Eric


 On Thu, Aug 27, 2009 at 12:05 PM, Valtteri Kiviniemi 
 valtteri.kivini...@dataproof.fi wrote:

 
 Hi,

 You should probably read the facts before posting. Ofc. its not exactly
 the same, but if you know nothing about Xen you would know that the
 performance difference between (for example 2.6.18-xen and 2.6.18
 kernels) are so small, that you cant even notice it.

 Maybe with ESXi you have greater performance difference compared to
 bare-metl but not with xen.

 - Valtteri Kiviniemi

 Kveri kirjoitti:
   
 believe me, if you have paravirtualized enviroment you don't have
 equal performance than on bare-metal. Paravirtualization adds another
 layer, so does overhead. Maybe performance in CSS, but I doubt about it.

 I'm using full VT on 4x quad core xeons with 16gb ram and providing
 1000fps 1.6 servers (yes, stable 1000fps, kernel self-pached with RT
 and some HZ tweaks), CSS servers with 100 ticrate and and some tf2
 servers without any problems.

 Kveri

 On 25.8.2009, at 20:52, Valtteri Kiviniemi wrote:

 
 Hi,

 We are running multiple TF2 servers with Xen 3.4.1 paravirtualized.
 Performance is exactly the same as bare-metal, maybe even better. Only
 downside is that you need xen-patched kernel so to get most stable and
 working environment you have to use the default 2.6.18.8-xen kernel.
 Ofc. you can compile a 1000hz domU kernel like we have.

 There is also pv_ops kernels which are included in the xen-unstable
 tree. They are the normal kernel.org kernel with patches that make it
 suitable for Xen hypervisor.

 In my opinion Xen is the best solution for gameserver virtualization
 because it is the fastest. ESXi virtuals are not paravirtualized so
 they
 have slower disk i/o and network performance. They also use more
 resources.

 If you want same performance as bare-metal you need paravirtualized
 guest operating systems and Xen is the best solution for that.

 We have a physical 2 x 2.5GHz Quad-core Xeon machine with 16 GB ram
 and
 a ARECA ARC-1220 raid controller with RAID10 array.

 We are also running many other virtuals on the same machine without
 them
 affecting the gameserver virtual performance.

 With Xen you can for example assign 4 physical cores to the gameserver
 virtual and use the other 4 for other virtuals.

 - Valtteri Kiviniemi

 Daniel Worley kirjoitti:
   
 I don't have exact numbers, but I've run srcds both natively and
 under ESXi
 on a PowerEdge server.  Under both I was able to run multiple
 instances, no
 issues.  I saw no difference in performance playing on the servers,
 but once
 again I don't have numbers to back it up.

 On Tue, Aug 25, 2009 at 11:07 AM, Claudio Beretta 
 
 beretta.clau...@gmail.com
   
 wrote:
 HiI'd like to know your experiences with running srcds in a
 virtualized
 environment. Searching mail-archive for past discussions about
 this subject
 didn't provide a reliable conclusion to this topic.
 From what i understand, only hypervisors such as ESXi, XEN (and
 maybe
 Hyper-V) are 

Re: [hlds_linux] HLTV never disconnects

2009-08-19 Thread Midnight
In server.cfg add tv_enable 0

Daniel Duarte wrote:
 I'm trying to help a friend with his servers but i'm out of ideas. He is
 using HLTV to record his matches but the HLTV stays connect forever, even
 after he quits hltv.exe. Right now the HLTV is on for 316h 15m!

 In the server we can't see the HLTV connected but in the server browser he
 is there taking a slot.

 ideas?


 tia,
 Daniel
 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] HLTV not starting

2009-06-16 Thread Midnight
make sure your install is up to date with -verify_all


Marcel wrote:
 No idea? :/

 Marcel wrote:
   
 Hi,

 My HLTV isn't starting and I have no idea why.
 The [S_API FAIL] line should not be a problem. I think it's only 
 because I don't have the hlds updatetool in this directory.

 But why Master module failed to initialize. (init failed) ??
 That doesn't sound good ;)

 Startup Line:
 export LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH; ./hltv -ip $someip -port 
 $someport +maxclients 10


 It shows
 ---
 Console initialized.
 FileSystem initialized.
 Network uses $someip as host IP.
 Network initialized.
 [S_API FAIL] SteamAPI_Init() failed; unable to update local steamclient. 
 Continuing with current version anyway.
 Master module failed to initialize. (init failed)
 ERROR! System::AddModule: couldn't initialize module master.
 * FATAL ERROR *
 Proxy::Init: add master module.
 *** STOPPING SYSTEM ***
 ERROR! System::AddModule: couldn't initialize module (null).
 Type 'help' for a list of commands.



 When I do a exec hltv.cfg:
 exec hltv.cfg
 Executing file hltv.cfg.
 Segmentation fault

 My hltv.cfg:

 name111
 hostnamesome name
 delay   90
 serverpassword  foo
 connect ip:27015
 adminpassword   somepw
 chatmode1
 logfile 1
 updaterate  20
 rate1
 proxypassword   
 spectatorpassword   




 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux


 

 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Timeout after client connected

2009-06-01 Thread Midnight
Is this on any particular game?  L4D maybe?  It seems to still have the 
disconnect on map change issue.  Quite annoying really.


Arg! wrote:
 if you run sourcemod, update to at least 1.2.1

 On Mon, Jun 1, 2009 at 10:33 PM, Nightbox alexandrualexa...@gmail.comwrote:

   
 but no, everytime someone connect to server, it crashs

 2009/6/1 Nightbox alexandrualexa...@gmail.com

 
 hmm, semms working now

 2009/6/1 AnAkIn . anakin...@gmail.com

 Disable your plugins, use a default cfg and see.
   
 2009/6/1 Nightbox alexandrualexa...@gmail.com

 
 Guys, i've discovered that, when someone connect to my server, he will
 receive the timeout error. (countdown)

 Do someone know how to fix this ? It's imposible to play.
 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

 
   
 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

 
 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] 32 Slot Linux Server

2009-04-09 Thread Midnight
Have you tried turning off -pingboost 2?

Try running at 33 tick / 100fps, I bet you won't have a problem.


nos...@gentiniphotography.com wrote:
 Hello everyone!

 New user here... I have a few questions.  Is there anyone out there
 successfully running a lag-free 32 slot Linux TF2 Server?  I have recently
 purchased a dedicated server with the following specs:

 10mbps uplink
 2 GB Ram
 Quadcore 3.2ghz Xeon

 (I can get more information if you guys need it).

 Anyways, I'm running a 66 tickrate server with the server process being
 confined to one processor, and when the server gets 26+ people in it the CPU
 is at 95+%.  We tried running 32 slot for a short while but the server lag
 was unbearable.  We have it down to 28 now and it still can get bad at times
 once we cross the 26 person threshold.

 The only notable server start up options are as follows:

 -pingboost 2 -fork -tickrate 66

 I must admit I'm a newbie in terms of running TF2 servers but I'm pretty
 competent when it comes to Linux.

 The server is CentOS (uname info):
 Linux xx.host 2.6.26.8-rt16 #2 SMP PREEMPT RT Sun Mar 29 10:18:58 CDT
 2009 i686 i686 i386 GNU/Linux

 I have also recompiled the kernel using the Fragaholic wiki
 http://wiki.fragaholics.de/index.php/EN:Linux_Kernel_Optimization#Setting_your_servers_to_run_with_realtime_scheduling

 Did not see much improvement over the default kernel.

 Any help is much appreciated

 -Brian
 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] 32 Slot Linux Server

2009-04-09 Thread Midnight
Seems that you are trying to run competition performance on a 32 slot 
pub.  That's asking a lot from the hardware.  You can probably squeak by 
with 66 tick if you keep the fps low enough and turn off pingboost.



David A. Parker wrote:
 Yes, or you can set it with +fps_max in the startup command line.

 nos...@gentiniphotography.com wrote:
   
 We'll try that tonight and I'll post up the results...

 We're trying to run 66 tick though...

 The FPS is set through the server.cfg file with fps_max, correct?

 -Brian

 On Thu, Apr 9, 2009 at 4:13 PM, Midnight mido...@gmail.com wrote:

 
 Have you tried turning off -pingboost 2?

 Try running at 33 tick / 100fps, I bet you won't have a problem.


 nos...@gentiniphotography.com wrote:
   
 Hello everyone!

 New user here... I have a few questions.  Is there anyone out there
 successfully running a lag-free 32 slot Linux TF2 Server?  I have
 
 recently
   
 purchased a dedicated server with the following specs:

 10mbps uplink
 2 GB Ram
 Quadcore 3.2ghz Xeon

 (I can get more information if you guys need it).

 Anyways, I'm running a 66 tickrate server with the server process being
 confined to one processor, and when the server gets 26+ people in it the
 
 CPU
   
 is at 95+%.  We tried running 32 slot for a short while but the server
 
 lag
   
 was unbearable.  We have it down to 28 now and it still can get bad at
 
 times
   
 once we cross the 26 person threshold.

 The only notable server start up options are as follows:

 -pingboost 2 -fork -tickrate 66

 I must admit I'm a newbie in terms of running TF2 servers but I'm pretty
 competent when it comes to Linux.

 The server is CentOS (uname info):
 Linux xx.host 2.6.26.8-rt16 #2 SMP PREEMPT RT Sun Mar 29 10:18:58 CDT
 2009 i686 i686 i386 GNU/Linux

 I have also recompiled the kernel using the Fragaholic wiki

 
 http://wiki.fragaholics.de/index.php/EN:Linux_Kernel_Optimization#Setting_your_servers_to_run_with_realtime_scheduling
   
 Did not see much improvement over the default kernel.

 Any help is much appreciated

 -Brian
 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 
 please visit:
   
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux


 
 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

 

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] [TF2] fps drops

2009-03-22 Thread Midnight
If you weren't looking at the stats you probably wouldn't even notice 
there is any problem.


DontWannaName! wrote:
 Im pretty sure fps drops are normal and cant be 'fixed'

 2009/3/22 Ben B brutalgoerge...@gmail.com

   
 I've been trying to stop my fps drops... playing with kernel settings, that
 idler thing, chrt, renice...  gah.  it's making me insane.

 On Sun, Mar 22, 2009 at 11:05 AM, Zuko zuc...@gmail.com wrote:

 
 how to get rid of this fps drops? http://zuko.isports.pl/syf/graph.png

 18:00:44 stats
 18:00:45 CPU   InOut   Uptime  Users   FPSPlayers
 53.80 66526.64 311163.72  81 2  989.12  24
 18:00:45 stats
 18:00:46 CPU   InOut   Uptime  Users   FPSPlayers
 54.00 68730.95 317612.44  81 2  991.08  24
 18:00:46 stats
 18:00:46 CPU   InOut   Uptime  Users   FPSPlayers
 54.00 67757.55 314314.38  81 2  991.08  24
 18:00:47 stats
 *18:00:47 CPU   InOut   Uptime  Users   FPSPlayers
 54.00 67545.59 308442.91  81 2   20.77  24*
 18:00:47 stats
 18:00:47 CPU   InOut   Uptime  Users   FPSPlayers
 54.00 67919.29 294907.53  82 2  991.08  24
 18:00:48 stats
 18:00:48 CPU   InOut   Uptime  Users   FPSPlayers
 54.00 68191.39 286455.69  82 2  990.10  24

 redhat, Intel(R) Xeon(R) CPU 5150  @ 2.66GHz

 --
 Żuko
 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Left 4 Dead Update Available

2009-03-11 Thread Midnight
We had 1 person report their server is not showing up on Windows.  Most 
of our other servers are showing up.  Haven't sorted out why.  Just 
going to reinstall and hope that fixes whatever was broke in his config.


Reaper wrote:
 Yes on Linux we're experiencing the same issue.  Windows servers are visible
 but have a different problem (showing bots in player count).

   

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] SourceTV Relay Proxy?

2009-03-09 Thread Midnight
The wording is kinda hard to understand maybe.  But it is saying that 
the game server itself cannot be used as a proxy.  That means it can't 
host viewers.  But you can connect a proxy to it.

Are you having problems getting a standalone proxy to connect to your 
server?

-Midnight

Magnus wrote:
 Hi.
 We run our srcds with -tvmasteronly becouse we think it's stupid to 
 enable 255 slots on the gameserver.
 But as i understand the wiki 
 (http://developer.valvesoftware.com/wiki/Source_TV) it's not possible to 
 connect a Dedicated sourcetv to the one on the gameserver.

 -tvmasteronly
 SourceTV can only serve one client and can't be used as relay proxy. 

 So, how should we do, if we don't whant our customers to be able to set 
 the slots by them self, on their gameservers sourcetv?
 But they should be able to connect Sourcetv to the one on their gameserver.


 /M


 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] fps fluctuates every time

2009-02-24 Thread Midnight
Yes.  Turn off anything that will lower the CPU speed.



Andy Giesen wrote:
 I would turn off SpeedStep and probably VT.

 On Tue, Feb 24, 2009 at 2:37 PM, kERPLUNK k...@datafull.com wrote:
   
 im right now looking my bios options..

 i have ACPI State S1 or S3 (s3 setted)
 Intel SpeedStep (on)
 Intel VT (on)
 Intel XD (on)

 may i have to change some of this options?

 - Original Message -
 From: kERPLUNK k...@datafull.com
 To: Half-Life dedicated Linux server mailing list
 hlds_linux@list.valvesoftware.com; Gary Stanley
 g...@velocity-servers.net
 Sent: Tuesday, February 24, 2009 8:26 PM
 Subject: Re: [hlds_linux] fps fluctuates every time


 in kernel do not have anything about cpu speed throttling.. nothing
 in bios, i have intel motherboard, and doesnt have nothing too about apm..

 - Original Message -
 From: Gary Stanley g...@velocity-servers.net
 To: Half-Life dedicated Linux server mailing list
 hlds_linux@list.valvesoftware.com; kERPLUNK k...@datafull.com;
 Half-Life dedicated Linux server mailing list
 hlds_linux@list.valvesoftware.com
 Sent: Tuesday, February 24, 2009 8:23 PM
 Subject: Re: [hlds_linux] fps fluctuates every time


 At 05:19 PM 2/24/2009, Matthias Bleile wrote:
 
 which clocksource do you use?
   
 It's not the clocksource. It's power management stuff enabled in the
 BIOS, or maybe some type
 of chipset errata. Or it's something enabled in the kernel, like CPU speed.





 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux


 ___
 To unsubscribe, edit your list preferences, or view the list archives, 
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

 

 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] server.cfg questions

2009-02-22 Thread Midnight
Though your hardware is pretty strong, 1000fps x 32 players seems like a 
lot for the CPU to have to do every 1ms.

I don't usually run that many players on 1000fps to prevent exactly this 
kind of problem, but I can't say for sure it is just the load causing 
it.  Perhaps others can chime in on the 1000x32 scenario if they are 
able to get that stable.

Maybe try 500fps or 333fps and see if things improve.  If so then you 
have your answer.


Christopher Szabo wrote:
 Hi!

 I’m running a Counter-strike 1.6 publicserver (32 slots) and I have a problem 
 with my settings.
 Everybody has a high choke, 30-100. I have tried many rate settings but none 
 is working well. 
 The connection isn’t the problem because it’s a datacenter and other 
 publicservers in the same
 datacenter don’t have the high choke. Anyway, I looked into my server.cfg and 
 found some
 things that I don’t really know what they do.

 max_queries_window -1
 max_queries_sec_global -1
 max_queries_sec -1

 And if that’s not the problem, what could be? I’m using amxmodx, the latest 
 and a few plugins
 like ptb, hpk, admin_listen and admin_esp. That’s all. Using sys_ticrate 
 1000, running kernel
 in 1000 HZ and using pingboost 2. Should I lower the ticrate?

 Dell Power Edge R300
 Dell Intel X5000 Chipset
 Quadcore Intel Xeon X5470 3.33Ghz 1333Mhz FSB
 Hynix 2x2GB Dual Rank-RAM
 2st Seagate 73GB SAS 15K rpm (Raid-1)

   

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] OFF TOIPC - Question to GSP's regarding Ventrilo

2009-01-30 Thread Midnight
Same here.  At least I wasn't the only one they snubbed repeatedly.  It 
would really serve them right if we all switched to Mumble and stop 
paying them our money since they don't want our business.


Steffen Tronstad wrote:
 Yes, forgot to mention that - I tried, my post was deleted with no answer.


 -Opprinnelig melding-
 Fra: hlds_linux-boun...@list.valvesoftware.com 
 [mailto:hlds_linux-boun...@list.valvesoftware.com] På vegne av Andrew 
 Armstrong
 Sendt: 30. januar 2009 13:26
 Til: 'Half-Life dedicated Linux server mailing list'
 Emne: Re: [hlds_linux] OFF TOIPC - Question to GSP's regarding Ventrilo

 I'd try their forums, pretty sure they keep an eye on those.

 -Original Message-
 From: hlds_linux-boun...@list.valvesoftware.com
 [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Steffen
 Tronstad
 Sent: Friday, 30 January 2009 11:16 PM
 To: hlds_linux@list.valvesoftware.com
 Subject: [hlds_linux] OFF TOIPC - Question to GSP's regarding Ventrilo

 Dear all GSPS on this list

 I've been trying to get a hold of flagship/ventrilo.com officials by
 email since mid november last year, have any of you been in contact with
 them recently? Do you know of any mail adresses they are actually
 monitoring/answering?

 I'm getting worried that I'm paying my monthly fee for no use - will
 they even notice if I stop paying them?

 I need to get in touch with them to change our legal name and logo for
 the current license. 


 Regards
 Steffen

 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux


 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] OFF TOIPC - Question to GSP's regarding Ventrilo

2009-01-30 Thread Midnight
I think I'm going to start offering free Mumble hosting for anyone that 
will use it.  Flagship needs to learn a lesson not to ignore their 
business partners.



RTL-Servers wrote:
 Hello,

 For a company reeping the funds in regardless, it wouldn't exactly be 
 difficult to send out a No your not welcome to a license email now 
 would it ?

 It just seems getting a vent licence is like tripping over the fountain 
 of youth to then find the holy grail.

 My two cents.

 -Lee

 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] OFF TOIPC - Question to GSP's regarding Ventrilo

2009-01-30 Thread Midnight
And so the Mumble revolution begins...

Chris wrote:
 I, like most people on this list, have used all three servers/clients.
  While I grew up on vent, their business strategy is not to my
 liking so I started setting up mumble servers.  With about 15 minutes
 of reading and configuring, I had better sounding and better
 performing mumble servers. I won't look back, vent is dead to me.
   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] timeouts at map change

2008-12-22 Thread Midnight


Steven Sumichrast wrote:
 Glad to hear others are having trouble.

   

Haha :/

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] [hlds] Left 4 Dead Demo Dedicated Server filesavailable

2008-11-10 Thread Midnight
You can script Windows also for that matter, and setup cron jobs.  
Sounds like you just didn't put the time into it on Windows like you did 
on Linux.

You might have doubled the update rate, doesn't mean you are getting 
double the data.  Game data is game data, it's not going to be any 
different on a different OS, the game still sends data just the same.  
Windows doesn't add extra headers to the UDP packet either, a UDP packet 
is a UDP packet, so whatever you change you are seeing is probably just 
in your head.

Crazy Canucks wrote:
 Can't help you with your specific problem, and this is a bit ot, but 
 stick with Linux, you'll be happy you did.  I switched from Windows to 
 Linux, and yes, it was a steep learning curve, but once I started 
 figuring things out, I was extremely happy I did.

 Just an example of the benefits:

 I host game servers for my clan on the weekend, and I host our website 
 on the the same box as the game servers.  When I start up the game 
 server, I take down the full website, and put up a minimal one in it's 
 place.  Anyway, on Windows, I used to have a check list, with about 20 
 steps on it for starting up and shutting down the game servers.

 I now have scripts that have replaced those twenty steps with one.  I 
 have a script that I run to start the servers, and I enter the name of 
 the game I want to play, the script handles everything else, and a cron 
 script shuts down the game server and puts back up the full website in 
 the morning after the game nights.

 My servers are bandwidth challenged.  When I switched to Linux I was 
 able to essentially double the update rate for my servers.  I can only 
 guess that Windows attaches some kind of bulky header to the packets 
 which Linux doesn't.  Switching to Linux was the smartest thing I've 
 ever done for my clan servers.

 Drek

 Arthur wrote:
   
 Any hints as to which packages? New to Linux its proving a steep learning
 curve - windows is so much easier :(




 -Original Message-
 From: [EMAIL PROTECTED]
 [mailto:[EMAIL PROTECTED] On Behalf Of Jordy van
 Wolferen
 Sent: Monday, 10 November 2008 7:22 PM
 To: Half-Life dedicated Linux server mailing list
 Subject: Re: [hlds_linux] [hlds] Left 4 Dead Demo Dedicated Server
 filesavailable


 This is only for an AMD64 box.
 Maybe you can update the necessary packages for 32 bit, without
 upgrading to lenny.

 On Mon, Nov 10, 2008 at 7:33 AM,  [EMAIL PROTECTED] wrote:
   
 
 Its a 32bit intel box,


 -Original Message-
 From: [EMAIL PROTECTED]
 [mailto:[EMAIL PROTECTED] On Behalf Of Jordy van
 Wolferen
 Sent: Monday, 10 November 2008 5:17 PM
 To: Half-Life dedicated Linux server mailing list
 Subject: Re: [hlds_linux] [hlds] Left 4 Dead Demo Dedicated Server
 filesavailable


 Did you keep the ia32-libs installed? You still need them

 On Mon, Nov 10, 2008 at 6:40 AM,  [EMAIL PROTECTED] wrote:
 
   
 Never mind :)


 Relised i forgot to extract the .deb file, however still doesnt work
   
 
 running
 
   
 http://ftp.de.debian.org/debian/pool/main/g/glibc/libc6_2.7-15_i386.deb

 Just segmentation faults on start up,

 Anyideas?



 -Original Message-
 From: [EMAIL PROTECTED]
 [mailto:[EMAIL PROTECTED] On Behalf Of
 [EMAIL PROTECTED]
 Sent: Monday, 10 November 2008 4:20 PM
 To: 'Half-Life dedicated Linux server mailing list'
 Subject: Re: [hlds_linux] [hlds] Left 4 Dead Demo Dedicated Server
 filesavailable


 Few questions for you,

 a) how do you remove the LD_PRELOAD paramater?

 b) output of my console now


 Could not locate steam binary:./steam, ignoring.
 ERROR: ld.so: object '/usr/games/l4d/temp/libc6_2.7-15_i386.deb' from
 LD_PRELOAD cannot be preloaded: ignored.
 Failed to open bin/dedicated_i486.so (/lib/tls/libc.so.6: version
 `GLIBC_2.4' not found (required by bin/dedicated_i486.so))
 Add -debug to the ./srcds_run command line to generate a debug.log to
   
 
 help
 
   
 with solving this problem
 ERROR: ld.so: object '/usr/games/l4d/temp/libc6_2.7-15_i386.deb' from
 LD_PRELOAD cannot be preloaded: ignored.
 Mon Nov 10 16:18:51 EST 2008: Server restart in 10 seconds
 ERROR: ld.so: object '/usr/games/l4d/temp/libc6_2.7-15_i386.deb' from
 LD_PRELOAD cannot be preloaded: ignored.

 Dual Core2 Quad Box running debian


 Thanks in advance

 -Original Message-
 From: [EMAIL PROTECTED]
 [mailto:[EMAIL PROTECTED] On Behalf Of Jordy van
 Wolferen
 Sent: Monday, 10 November 2008 12:14 PM
 To: Half-Life dedicated Linux server mailing list
 Subject: Re: [hlds_linux] [hlds] Left 4 Dead Demo Dedicated Server
 filesavailable


 I fixed it on my Debian Etch AMD64 server.
 I used this package:

 http://packages.debian.org/lenny/libc6-i386

 wget

   
 
 http://ftp.nl.debian.org/debian/pool/main/g/glibc/libc6-i386_2.7-15_amd64.de
   
 
 b
 ar x libc6-i386_2.7-15_amd64.deb
 tar xvzf data.tar.gz

 Then LD_PRELOAD the lib like this:

 export 

Re: [hlds_linux] sys_ticrate

2008-10-01 Thread Midnight
Actually that part at the bottom is entirely wrong.

 The key reason to run higher FPS is the render time. At 1000FPS, the 
 server is rendering one frame every 1 millisecond (ms). This means 
 that the worst-case adder to the player ping is only 1ms, IE: the 
 player gets more accurate data and can get it more often.

 At 300FPS it's only 3ms which is perfectly acceptable, but at 100FPS 
 it's 10ms, which is a significant percentage of a 100 ping (10%). A 
 player with a 100 ping would actually be getting 110ms response time 
 from the server.
That is incorrect.  tickrate actually controls how much delay is added 
to ping time.  At 100 tickrate there will always be 10ms of ping added 
to the real DOS ping, regardless of how high the sys_tickrate/max_fps is 
so long as it is over 100.  This is becaues packets are going out 100 
times/second regardless of the server FPS.

The higher FPS setting may help accuracy still, but it does not affect 
ping as indicated on that page.  Really though how many pixels do you 
think a player moves if the server is clocking at 3ms frames vs 1 ms 
frames?  Maybe a couple pixels?  The inaccuracy of the guns themselves 
is way larger than any movement of the player in that timespan - I 
haven't calculated out the exact pixel difference but I think we can all 
agree it is pretty small amount in 2ms of time.  That's why going from 
333 to 1000 FPS is essentially pointless.  The only thing it might might 
matter for is sniper/deagle, but even the pro's don't have pixel level 
accuracy.


Gary Stanley wrote:
 At 08:02 PM 10/1/2008, Matthias Bleile wrote:
   
 Of course you get an answer.

 Right here -
 http://supportwiki.steampowered.com/wiki/Optimizing_a_Dedicated_Server
 at the very bottom.

 

 Really?
 http://supportwiki.steampowered.com/w/index.php?title=Optimizing_a_Dedicated_Serveraction=history

 That page is user created, with valve supervising. It looks like some 
 users created the content. I don't see Alfred's name on the history, 
 nor anyone else that regulars the mailing lists that is from valve.



 Gary Stanley - [EMAIL PROTECTED] - [EMAIL PROTECTED]

 http://leaf.dragonflybsd.org/~gary

 Put your hand on a hot stove for a minute, and it seems like an hour.
   Sit with a pretty girl for an hour, and it seems like a minute. 
 THAT'S relativity.  - Albert Einstein



 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] sys_ticrate

2008-10-01 Thread Midnight


Gary Stanley wrote:
 At 08:46 PM 10/1/2008, Midnight wrote:
   
 That is incorrect.  tickrate actually controls how much delay is added
 to ping time.  At 100 tickrate there will always be 10ms of ping added
 to the real DOS ping, regardless of how high the sys_tickrate/max_fps is
 so long as it is over 100.  This is becaues packets are going out 100
 times/second regardless of the server FPS.
 

 That is incorrect. Maybe in game, but not certainly layer3..

   
Well from the game's perspective it's sending out 100 packets a second.  
My point was it is adding on average 10ms not 1ms like that web page 
indicates.  How those are processed over the network is something else 
entirely, but it only further makes the point that 1000FPS servers are 
ridiculous.  All of these factors of instability generally swamp the 
theoretical 1-3ms of variance in the game server's frame rate by an 
order of magnitude.
 The higher FPS setting may help accuracy still, but it does not affect
 ping as indicated on that page.  Really though how many pixels do you
 think a player moves if the server is clocking at 3ms frames vs 1 ms
 frames?  Maybe a couple pixels?  The inaccuracy of the guns themselves
 is way larger than any movement of the player in that timespan - I
 haven't calculated out the exact pixel difference but I think we can all
 agree it is pretty small amount in 2ms of time.  That's why going from
 333 to 1000 FPS is essentially pointless.  The only thing it might might
 matter for is sniper/deagle, but even the pro's don't have pixel level
 accuracy.
 

 gettimeofday() is used to step time in the engine. The problem with 
 gettimeofday is that its an ESTIMATE of wallclock time, and it's very 
 sensitive to changes on the machine
 it's running on. gettimeofday is archaic, nonstandard, imprecise, and 
 valve should use clock_gettime() instead to do timing.

   
Yep I have to agree there.
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Intel Quad-Core Xeon - problem with CPU load

2008-09-20 Thread Midnight
My thought on this, if you are running 10+ servers with 500fps 100 tick 
then ya you are probably hitting the limits of the hardware in terms of 
context switching.  I run a similar box and can't go more than about 30% 
cpu usage on it without problems with lag.


John Morgan wrote:
 Hello ppl,

 I've a huge problem with performance of my servers. I'm hosting arround 30
 gameservers per 1 server, mostly CS 1.6, few CS:S and CS:CZ. Allways when
 ~160-200 players play CPUs load jump from 30% to ~70-80% (suddenly all
 processes are using more power of CPU) and players get lags. No matter what
 kind of gameservers are working. On 1 server I'm put only CS 1.6, on another
 I mixed CS 1.6 with CS:S. Both has these same problem.

 I tried in 2 different Data Centers so it's not a network problem. Maybe my
 system was improperly setup.

 Configuration of servers: 2x Quad E5420, motherboard S5000PAL, 6x2GB FB-DIMM
 667, 2x SAS with RAID 1 (i tried with 1 disk, no positive result).

 System: Slackware 12.1 and Centos 5.2
 Kernel: 2.6.24.3, 2.6.25.4, 2.6.26.2 for Slack and 2.6.18-92.el5PAE for
 Centos (default)
 Kernel's Processor Settings which I tested: BIG SMP/PC Compatible,
 300/1000Hz, low-latency/server, preempt big kernel on/off, Core 2/newer
 Xeon

 How it looks:

 Players: http://img81.imageshack.us/my.php?image=playersrt1.jpg
 CPU Load: http://img221.imageshack.us/my.php?image=cpucr5.jpg

 I don't know what can I do. Old Dual Xeons 3.0Ghz with HT works fine with
 ~100. I think that 2x Quad-Core Xeon should works fine with 300 playing
 players. Don't You think so?

 P.S. Sorry for my english ;-)

 Regards,
 J.Morgan
 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Intel Quad-Core Xeon - problem with CPU load

2008-09-20 Thread Midnight
I've done the CPU Core pinning thing, didn't help anything in my case.  I don't 
think it is swapping cores that is the problem, it is just the ability to get 
CPU time in a stable and consistent rate with so many games swapping in/out of 
the CPU.  I think less games with higher slot counts is the only way to 
maximize a 8-way Xeon setup.

I have a theory that FB RAM also makes the situation worse by adding additional 
latency to the overall system performance.  We see a DPC latency of 3000us vs 
30-50us on a server board vs non server board.  I have setup a new server just 
for running 500-1000FPS CS/CSS games.  It is DDR3 non FB RAM, 1 CPU and 
overclocked to the max, basically an enthusiast board setup vs a server setup.  
So far in testing it seems to be able to run more high speed servers than a 2x 
Xeon server board.  So we are looking to use that server to run more lower end 
game servers that are not cranked to the max on the tickrate.

 not it's not crazy.  Running more than one cpu means the OS and programs 
 have to keep track of which core they are on, maintain cache coherency, 
 maintain the memory mappings..etc etc.  If(when) the thread jumps to 
 another cpu then the whole dataset has to get rarranged.  I would start 
 hard assigning servers to their own core and see if that helps out.

   
___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] [hlds] Upcoming Team Fortress 2 Update

2008-06-11 Thread Midnight
What happened to the /Orangebox directory?  Are you putting that back in 
place?


Jason Ruymen wrote:
 Sometime later today we'll have a new required Team Fortress 2 update.
  
 Jason
  
 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds

   


___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] orangebox folder under linux gone or not?

2008-06-10 Thread Midnight
This same thing has happened on the Windows tf2 install.  Orangebox dir 
is gone, it is installing into /tf2 now.

This is really screwing things up for anyone who has mods installed.  
Thanks Valve!


Miano, Steven M. wrote:
 I just did a fresh install this morning, the /srcds_l/orangebox was where it 
 placed my srcds_run. It sounds odd that you would have been able to 
 install/update tf into the root directory, I didn't see/hear/read anything 
 about that.

 ~Steven

   
 -Original Message-
 From: [EMAIL PROTECTED] [mailto:hlds_linux-
 [EMAIL PROTECTED] On Behalf Of G2G-Support
 Sent: Tuesday, June 10, 2008 5:56 AM
 To: hlds_linux@list.valvesoftware.com
 Subject: [hlds_linux] orangebox folder under linux gone or not?

 Hi,

 a few weeks ago I realized that after a fresh install of TF2 the
 orangebox folder under linux was gone. So I reinstalled every server to
 use this new format. Now, after the newest update by entering ./steam
 -command update -dir . -game tf -verify_all steam is downloading the
 whole orangebox folder again. Am I doing something wrong or did Valve
 make a change again? What is now correct? With the orangebox folder the
 server can't automatic update itself because the steam binary resides in
 the wrong folder.

 kind regards
 C.Schmitz

 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux
 

 CONFIDENTIALITY NOTICE: This e-mail may contain information that is 
 privileged, confidential or otherwise protected from disclosure. If you are 
 not the intended recipient of this e-mail, please notify the sender 
 immediately by return e-mail, purge it and do not disseminate or copy it.

 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

   


___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Quoting Habits

2008-02-15 Thread Midnight
This is a multi-part message in MIME format.
--
Lies.

Ryan Devonshire wrote:
 I never quote :(


--
[ midowns.vcf of type text/x-vcard deleted ]
--

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] TF2 - High CPU Usage

2008-01-06 Thread Midnight
This is a multi-part message in MIME format.
--
What kind of CPU is it and how do you have the server configured?



Rian Brooks-Kane wrote:
 --
 [ Picked text/plain from multipart/alternative ]
 Hi My TF2 Server has nothing else on it. running a completely vanilla
 install of TF2.

 With just 6 People its running at 40% CPU.

 rcon stats reports 99.90 cpu usage with just 3 people.

 Any ideas?

 --
 Rian Brooks-Kane
 --

 ___
 To unsubscribe, edit your list preferences, or view the list archives, please 
 visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux


--
[ midowns.vcf of type text/x-vcard deleted ]
--

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Server hardware requirements

2007-09-15 Thread Midnight
This is a multi-part message in MIME format.
--
Should be fine for a 20 player server at default settings.  The real
question becomes how high can you run the tick rate and fps settings at
and still not overload the server.  I think one core should handle at
least 1 20 player server at 500fps and 100 tick without any problem.

James Gurney wrote:
 Hi folks, long time no email. (Been on and off this list for the past 6
 years or so).

 I tried searching the archives for this question, but it's hard to do,
 so apologies in advance if this has come up on a regular basis. If
 there's a thread or a wiki page somewhere I can read, I'd appreciate the
 link.

 We're getting ready to host a TF2 server on our now aging hardware, and
 I'd like to get a rough idea of how much performance we'll get with the
 Steam back-end. We have a dual Opteron 244 (1.8GHz). I'm well aware that
 the dual proc won't help a great deal, so considering a single one of
 these processors - is it sufficient to host a 20 player steam based
 server? Of course, TF2 will likely put more strain than a hldm or CS
 server, but I'm just looking for a rough estimate at this point.

 If the hardware we have is insufficient, what's a good level to be going
 for instead? We're running Linux, if that makes a significant difference.

 Thanks in advance,

 James

 ___
 To unsubscribe, edit your list preferences, or view the list archives,
 please visit:
 http://list.valvesoftware.com/mailman/listinfo/hlds_linux

--
[ midowns.vcf of type text/x-vcard deleted ]
--

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] RE: Server's got choke and loss ...

2007-07-25 Thread Midnight

Linux makes me LOL

Daniel Küspert wrote:

This is a multi-part message in MIME format.
--
[ Picked text/plain from multipart/alternative ]
Hi,

my problem isn't solved until now ...

Today I tried to build a new kernel (2.6.22) with CK Patch ...

But the Loss and Choke is still there ...

Now I'm back on the standard Etch Kernel (2.6.18)

Does anybody have an idea ?

--






___
Der frühe Vogel fängt den Wurm. Hier gelangen Sie zum neuen Yahoo! Mail: 
http://mail.yahoo.de


___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux




___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] CS 1.6 in game Ads

2007-03-07 Thread Midnight

If you CS guys want to quit and come play UT2004 we'd be happy to have
you.  I wouldn't put up with this if I were you.

John Sheu wrote:

/rant

Quite frankly, I am sick and tired of hearing people whine about thi_.  Don't
get me wrong: I am all against in-game advertisements.  I feel that they
detract from the gameplay experience and the congruity of the environment,
and are generally just a cheap trick to cash in on a popular game.

But quite equally, I am all against people who will whine and raise bloody
hell about boycotts and such, and promptly return to playing the damn game
in half an hour.  If you want a boycott, put your money where your mouth is
and *do it*.  Don't give me lame excuses about how everybody plays it,
or I'm in CAL and I can't quit, or such.  You're no more than a childish
bunch of brats until I see you stand up like real men for what you believe
in.  As of now, I can only conclude that you're just doing this because you
like attention.

And by you, I mean the vast mass of bitching CS players out there.  So
nothing personal.

-John Sheu

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux




___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Doors at Tickrate 100

2006-12-05 Thread Midnight

No offense, but this is wrong imo.

I think you are confusing bandwidth caps and updates/sec.

However I do agree that higher tick helps improve the feel of the game,
but only to a point.  At some point it starts to hurt things because the
server and the player's latency cannot keep up, making higher tickrates
useless.


storno.porno wrote:

Picture this :

10 players (5 on each team) are in the same part of the map.
All of them are shooting and moving.
By shooting around they hit a few props as cans, tons etc aswell, and they
start moving too.

How in hell is the server supposed to let everybody see the exact same scene
if it can't make more updates than 66 in one tick?


Interpolation.

Also, it only has to interpolate 5ms of extra positional delta, which is
pretty small.

There is a very good reason why leagues demand to use tick100 on official
matches.
And that reason is simply accurate gaming.


I would more likely call it psychological self-imposed setting.

If u think there's no way to tell the difference between a 100 tickrate
server and a 66 tickrate server while playing, then obviously your
requirements aren't very high (which is ok , I assume, for a fun player ...
but definitely not ok for a competitive semi-professional gamer).

FYI, in UT2004 the top players feel that 50-60 tickrate is totally
adequate and anything more is overkill and will only hurt server
performance by taxing the system or internet connection.


___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: SV: [hlds_linux] Counter-Strike: Source and Source Engine Update Released

2006-11-19 Thread Midnight
This is a multi-part message in MIME format.
--
[ Picked text/plain from multipart/alternative ]
X-MIME-Autoconverted: from 8bit to quoted-printable by ms-smtp-04.socal.rr.com 
id kAK0Aof2004137

It's a setting for pub servers, get over it.  Nobody is forced to use it
for competition.


Nullbit wrote:
 Agree in this post! The business impact of this failure is devastating!

 For 10 years have I been working with SAP! They also make mistakes and
 sometimes - sadly to admit - big mistakes!

 But they who are the customers are - They know that the key factor to
 success in the long run is satisfied customers!

 After 24 hours would Alfred's successor (or the successor responsible for
 the mistake) been apologising for the mistake and in the same moment
 announced a world-wide roll back of the update has already been initiated on
 his initiative!

 And they would (my guess) announce a programme to prevent this types of
 mistakes together with an explanation of the ideas behind the disaster!

 If I was the devils advocate and I would like to sabotage the HL2 based
 servers in order to move players to other platforms would I say that this is
 the most successful attempt that I ever could have hope to achieve... And I
 believe a serious investor would see the update in the same perspective.

 On the stock exchanges would they say, that the management is trying to
 destroy value with updates like this one - coming without a warning and a
 default value of 1 instead of 0! Because if they could code 3 different
 options - why on earth would they make the default value of 1 - leading to
 major drop in the number of servers and players - why not make the default
 value 0 and combine it with a kind of pop-up that a new CVR is introduced
 and explain the impact of this CVAR!

 The investors would probably listen to explanations from Valve but they
 would see the default value of 1 as a deliberate attempt of destroying their
 value.

 My only hope is that an investor with influence would read this thread and
 punish the Valve top management for this serious mistake!

 Br
 nullbit

 -Oprindelig meddelelse-
 Fra: [EMAIL PROTECTED]
 [mailto:[EMAIL PROTECTED] På vegne af Ghost
 Sendt: 18. november 2006 00:12
 Til: hlds_linux@list.valvesoftware.com
 Emne: Re: [hlds_linux] Counter-Strike: Source and Source Engine Update
 Released

 Valve - instead of ignoring us, can  you give us some insight?  Can you
 say something?  Why are you always so silent?

 Specifically, I would like to hear your thoughts on:

 - Is this update _really_  worth breaking so many mods and server add-ons?
 - Why is CS:S not considered good enough for the CPL?
 - Why are you adding features that the community did not ask for instead
 of fixing long-time bugs?
 - Why do you think that CS:S is played less then CS1.6 even though 1.6
 is much older?
 - Why are you ignoring the community instead of using them to your
 advantage? (testing, betas, ideas, sales, etc.)
 - If the community is not important to you what is?
 - What are your long term plans for CS:S?
 - Why do you think your solutions are the best ones?

 I am not trolling for Valve bashing answers from the court.  These are
 honest questions that I have for Valve.

 I run 3 public servers that together get 60,000+ connection attempts per
 month and a CS:S community with a website that gets 350,000+ hits per
 month.

 My communities users spend money by renting servers (game, web, vent),
 buying STEAM games (The Ship, etc.) and convincing others to buy CS:S.

 Yes we are just a small voice but the point I am trying to make is that
 we are part of the overall community. We are your end users.

 I will not try and hide my disappointment in the last update but this
 really sucks.  The add-ons provided by Mani make my servers fun. They
 add choice and switch up the same old thing. Now that is gone (by
 default).

 For the first time I actually started thinking about either just giving
 up and shutting everything down or migrating to a new game. My decision
 has not been made yet, I'm still thinking.

 I guess I just don't understand and am hoping for some honest
 enlightenment from Valve themselves.

 Am I asking for too much?

 Fred Harju
 aka: Ghost++;
 aka: very frustrated, disappointed and confused



 Jason Ruymen wrote:

 Updates to Counter-Strike: Source and the Source Engine have been
 released.  Release run hldsupdatetool to receive these updates.  The
 changes are:

 ...
 Jason

 ___
 To unsubscribe, edit your list preferences, or view the list archives,

 please visit:

 http://list.valvesoftware.com/mailman/listinfo/hlds_linux







--

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] Half-Life 1 Dedicated server update

2006-10-26 Thread Midnight

Scott Pettit wrote:


Because we don't have any x86 swervers : 

amd64 has always worked nicely for me.

-Scott




except it's not secure.

___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux


Re: [hlds_linux] CS:S - Net-Code FIX - open letter

2006-09-30 Thread Midnight

I didn't see any explination of how to fix the netcode, or even what it
should do differently.  All this guy is doing is pointing out some
problems that are inherient in games played over the internet.  The
problem is, there is no fix.  Latency is not going away, ever.  With
latency in the picture, all you can do is find ways to optimize around
it, which is what is currently done with the netcode.  If that guy is so
smart he should at least provide a description of how the new netcode
should work.

Rónai György wrote:


[ Converted text/html to text/plain ]
I dont think that fixing the netcode is just that simple.
Its something naturally, like if you would say stop the wind blowing.
If you play on the internet, there will be always a latency, and an amount of
lag, wich must be corrected somehow. The server collects infos from the
players, calculates the new world frame, and then sends it back to the users,
in updates. All the steps of this needs time, and this time is latency, that
must be corrected. Since all net-connections are different, there must be a
buffer, to collect, correct, and then display changes. Therefore, a perfect
hitreg will never be, with this system.
Of course, they could make client side hitreg, where if you someone, the hit
info sent by your client to the server, that it must be a hit. But it would
make impossible, to filter out cheaters, since anyone could send false hitreg
updates, with a cheat, or even if there is a client side anticheat, from a 3rd
computer.
This is not just a problem, where we can say, that 'Valve fix it asap!'. Try
explain them how...
Locutus
Michael McKoy írta:





___
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux