Re: [hlds_linux] team fortress 2 high cpu consumption
Do you mind if I ask where your servers are based out of? # Eric On Sun, Jan 17, 2010 at 8:26 PM, "[ЯтR] The-/ wrote: > Our community runs 6 tf2 servers and sadly for our setup they run better > on windows than nix, on nix a 32 player server would just eat a core > 100% and lag massively windows at least allows them to swap a little > more, just a thought to try... > > EVAgames Community wrote: > > Even DoD:s which is built on same engines, uses less CPU than > TeamFortress. > > So better solution is to change CPU. More GHz per core, more fun. > > > > -- > > From: "pet" > > Sent: Sunday, January 17, 2010 4:19 PM > > To: "Half-Life dedicated Linux server mailing list" > > > > Subject: Re: [hlds_linux] team fortress 2 high cpu consumption > > > > > >> This can't be, oder counter strike based games do not have so huge CPU > >> usage! Is there a solution to lower the CPU usage? > >> > >>> I have 2.8GHz Quad Core and I can run 1000fps TeamFortress with ~26 > >>> players > >>> on one core. Usage is about ~70-80%. > >>> So, doing some math... > >>> (26*2.0)/2.8=18.57 > >>> And I think that 18 players on 2.0GHz core isn't so bad. If you want > >>> more, > >>> try lowering your kernel settings. Do not use RT and lower fps to > >>> ~300-500. > >>> > >>> > >>> -- > >>> From: "pet" > >>> Sent: Sunday, January 17, 2010 3:14 PM > >>> To: > >>> Subject: [hlds_linux] team fortress 2 high cpu consumption > >>> > >>> > >>> > Hi > > I have some problem with my team fortress 2 server: > > Platform: > > System: OpenSUSE x32 11.1 > CPU: Xeon DP E5405 Quadcore 4x 2,0 GHz > memory: 4 GB DDR2-667 ECC Registered > hd: 2 x 250GB SATA II 7.200 obr./min. RAID 1 > network: 2 x 1000 Mbit > link to server: 100 Mbit Switchport > > Issue: Server has 18 slots with standard metamod and sourcemod and I > have serious problem with CPU utilization. When the server is full it > reaches 70% CPU usage, this is very high and I do not now why, *any > help > would be appreciate*! > > meta version > Metamod:Source version 1.8.0 > Build ID: 691:92ea98d22d75 > Loaded As: Valve Server Plugin > Compiled on: Dec 19 2009 > Plugin interface version: 15:14 > SourceHook version: 5:5 > http://www.metamodsource.net/ > > meta list > Listing 4 plugins: > [01] SourceMod (1.3.0) by AlliedModders LLC > [02] TF2 Tools (1.3.0) by AlliedModders LLC > [03] BinTools (1.3.0) by AlliedModders LLC > [04] SDK Tools (1.3.0) by AlliedModders LLC > > > regards > > Pet > ___ > 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 > >>> > >>> __ Informacja programu ESET NOD32 Antivirus, wersja bazy > sygnatur > >>> wirusow 4779 (20100117) __ > >>> > >>> Wiadomosc zostala sprawdzona przez program ESET NOD32 Antivirus. > >>> > >>> http://www.eset.pl lub http://www.eset.com > >>> > >>> > >>> > >>> > >>> > >>> > >> ___ > >> 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] CP for Servers
I'm all swiftpanel. Eric On Wed, Sep 16, 2009 at 9:51 AM, Maximilian L. wrote: > TekBase here > http://www.teklab.de/english.html > > > Swift panel is freaking awesome. Not to mention free right now! > > > > On Wed, Sep 16, 2009 at 8:40 AM, EGL Sales wrote: > > > > > >> I'll put my vote in for Swift Panel! It's great > >> > >> Brendon Gough, > >> General Manager > >> > >> Elysium Gamers League > >> Gold Coast, QLD > >> > >> -Original Message- > >> From: hlds_linux-boun...@list.valvesoftware.com > >> [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Al > >> Sent: Wednesday, 16 September 2009 5:49 AM > >> To: Half-Life dedicated Linux server mailing list > >> Subject: Re: [hlds_linux] CP for Servers > >> > >> The panel i use is gamecp - gamecp.com it has lots of features and good > >> support and has an active development project. > >> > >> Alex wrote: > >> > >>> There is multiple different control panels out there, > >>> > >>> cPGS (cPanel Game Server Addon (still in beta, but really nice - great > >>> support as well (Run's off cPanel/WHM)). > >>> gPanel (Game Panel (Fairly decent, but support isn't that great, it's > >>> freeware though)). > >>> Swift Panel (Decent control panel (from what I've seen) looks to be > good > >>> support as well). > >>> > >>> -- > >>> From: "Gregg Hanpeter" > >>> Sent: Wednesday, September 16, 2009 5:14 AM > >>> To: "Half-Life dedicated Linux server mailing list" > >>> > >>> Subject: Re: [hlds_linux] CP for Servers > >>> > >>> > >>> > Take a look at this: http://www.swiftpanel.com/ -Never tried it but I > heard > good things at WHT. > > -G > > On Tue, Sep 15, 2009 at 3:02 PM, -=Acid=Rain=- > wrote: > > > > > Hi there guys, > > > > Is there any CP for dedicated servers running on linux, good like > > > >> TCADMIN > >> > > for Windows? > > > > Thanks! > > > > > > Rodrigo "Acid Rain" Kroehn > > -=Acid=Rain=- Games > > www.acidrain.com.br > > > > ___ > > > > AVISO LEGAL: Esta mensagem e arquivo(s) podem conter informações > > confidenciais e/ou legalmente protegidas. > > Caso tenha recebido por engano, favor devolvê-la ao remetente e > > eliminá-la > > do seu sistema, não divulgando > > ou utilizando a totalidade ou parte desta mensagem ou dos documentos > a > > ela > > anexados. > > > > LEGAL NOTICE: This message and attached document(s) may contain > > information > > of confidential nature and/or > > legally protected. If you have received this message by mistake, > please > > reply to the sender, eliminate it > > from your system and do not disclose or use this message or the > > > >> attached > >> > > documents, in whole or in part. > > _ > > ___ > > 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 > >> > >> > > ___ > > To unsubscribe, edit your list preferences, or view the list archives, > please visit: > > http://list.valvesoftware.com/mailman/listinfo/hlds_linux > > > > __ Information from ESET NOD32 Antivirus, version of virus > signature database 4427 (20090915) __ > > > > The message was checked by ESET NOD32 Antivirus. > > > > http://www.eset.com > > > > > > > > > > > > > -- > Mailing List Conversations - mail...@ml86.de - Please don´t spam :) > > > ___ > 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:
Re: [hlds_linux] Very unstable FPS in hlds
What CPU do you have in it? Eric On Wed, Sep 9, 2009 at 11:32 AM, Nevermore 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 In Out Uptime Users FPS Players > 0.50 0.00 0.00 6 0 103.98 0 > stats > CPU In Out Uptime Users FPS Players > 1.00 0.00 0.00 6 0 214.50 0 > stats > CPU In Out Uptime Users FPS Players > 1.20 0.00 0.00 6 0 136.95 0 > stats > CPU In Out Uptime Users FPS Players > 1.17 0.00 0.00 6 0 123.05 0 > stats > CPU In Out Uptime Users FPS Players > 2.00 0.00 0.00 6 0 141.34 0 > stats > CPU In Out Uptime Users FPS Players > 2.00 0.00 0.00 6 0 112.69 0 > stats > CPU In Out Uptime Users FPS Players > 2.00 0.00 0.00 6 0 302.02 0 > > Same configuration in a windows box: > > CPU In Out Uptime Users FPS Players > 1.56 0.00 0.00 2651 978 250.84 0 > stats > CPU In Out Uptime Users FPS Players > 1.56 0.00 0.00 2651 978 233.92 0 > stats > CPU In Out Uptime Users FPS Players > 3.13 0.00 0.00 2651 978 252.14 0 > stats > CPU In Out Uptime Users FPS Players > 3.13 0.00 0.00 2651 978 231.70 0 > stats > CPU In Out Uptime Users FPS Players > 3.13 0.00 0.00 2651 978 238.05 0 > stats > CPU In Out Uptime Users FPS Players > 3.13 0.00 0.00 2651 978 236.71 0 > stats > CPU In Out Uptime Users FPS Players > 3.13 0.00 0.00 2651 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] 1000 FPS CentOS Servers?
I recently had a guy complain about ping from Chicago to Montreal. His ping was 30 and he proclaimed his internet was 'badass' and he pings 80 to china so my servers had a problem. I explained exactly how the internet works but he insisted my server had issues. I further explained it takes 300ms to blink your eye. They named their clan '300ms or FAIL' and dedicated it to hating me :-) I feel honored :-) Eric On Tue, Sep 8, 2009 at 7:55 PM, Gregg Hanpeter wrote: > On a side note, whenever anyone complains about bad registration I usually > ask them for a traceroute and most often that player is shitty at the > moment. But anyways, with the many seasons I have spent in cal and cevo, I > know a cruddy server when I join it, especially when the ip begins with an 8 > (hehe). > > On Tue, Sep 8, 2009 at 7:51 PM, Gregg Hanpeter wrote: > >> Yes, I probably would if there were 9 other players in the server.. But you >> have to keep in mind that I am on comcast and they blow. More often than not >> I blame the isp or anyone that is located in giganet. >> >> >> >> >> On Tue, Sep 8, 2009 at 2:25 PM, Eric Greer wrote: >> >>> I think servers above 350 FPS are just one of those "What the customer >>> wants, the customer gets" things. >>> >>> Do you think you would notice a difference with tick rate at 100 and >>> FPS over 350 Gregg? >>> >>> Eric >>> >>> >>> >>> On Tue, Sep 8, 2009 at 9:48 AM, Gregg Hanpeter >>> wrote: >>> > I would be complaining. I hate cruddy registry. If you take this >>> discussion >>> > to CEVO forums you would get a lot of negative opinions to low fps >>> rates. >>> > Sure the average joe won't notice but I sure would. >>> > >>> > On Mon, Sep 7, 2009 at 1:57 PM, Ulrich Block wrote: >>> > >>> >> Stcok fps ist ~245 even if the value is set to fps_max 300 ^^ And no >>> our >>> >> players didnt even realise setting that i set fps_max to 75 @ tick 66. >>> >> I was just curious if they would realise and they did not... >>> >> >>> >> Eric Greer schrieb: >>> >> > I was reading about 'if its really worth it' today. >>> >> > Apparently 500 FPS servers add 2ms of delay while 1000 FPS servers >>> add >>> >> 1ms >>> >> > delay. Also, 2000 FPS adds only .5ms. >>> >> > >>> >> > Considering it takes the human being 300ms to blink his eye, and i >>> can do >>> >> > that while typing accurately... talking about anything over 350fps >>> >> literally >>> >> > becomes trivial. Include in tick rates deprecating this math and >>> you're >>> >> > literally down to nothing more than the CPU calculating the same crap >>> >> over >>> >> > and over and over. >>> >> > >>> >> > The most important thing is tick rate. In my opinion, if a blind >>> study >>> >> was >>> >> > done NOBODY would be able to accurately tell the difference between a >>> 100 >>> >> > tick 350fps server compared to anything higher - ever. >>> >> > >>> >> > But that never stopped people from selling what the public wants to >>> pay >>> >> for >>> >> > :-). Plus, the amount of placebo effect involved with 'pro' PC >>> gamers is >>> >> so >>> >> > intense I have a hard time comprehending it. Simply telling them >>> >> they've >>> >> > switched to a 1000 FPS server should make them rave about how great >>> it >>> >> is. >>> >> > The only REAL difference is what number comes up in tests. >>> >> > >>> >> > I'm sure gamers will never believe this and you'll find lines of >>> people >>> >> > waiting to tell you that they can see the difference between 350, >>> 500, >>> >> 1000 >>> >> > and 2000 FPS servers. *sigh* >>> >> > >>> >> > Eric >>> >> > >>> >> > >>> >> > On Mon, Sep 7, 2009 at 1:04 PM, Daniel Nilsson < >>> daniel.joki...@gmail.com >>> >> >wrote: >>> >> > >>> >> > >>> >> >> It´s very true that many do not see th
Re: [hlds_linux] 1000 FPS CentOS Servers?
I think servers above 350 FPS are just one of those "What the customer wants, the customer gets" things. Do you think you would notice a difference with tick rate at 100 and FPS over 350 Gregg? Eric On Tue, Sep 8, 2009 at 9:48 AM, Gregg Hanpeter wrote: > I would be complaining. I hate cruddy registry. If you take this discussion > to CEVO forums you would get a lot of negative opinions to low fps rates. > Sure the average joe won't notice but I sure would. > > On Mon, Sep 7, 2009 at 1:57 PM, Ulrich Block wrote: > >> Stcok fps ist ~245 even if the value is set to fps_max 300 ^^ And no our >> players didnt even realise setting that i set fps_max to 75 @ tick 66. >> I was just curious if they would realise and they did not... >> >> Eric Greer schrieb: >> > I was reading about 'if its really worth it' today. >> > Apparently 500 FPS servers add 2ms of delay while 1000 FPS servers add >> 1ms >> > delay. Also, 2000 FPS adds only .5ms. >> > >> > Considering it takes the human being 300ms to blink his eye, and i can do >> > that while typing accurately... talking about anything over 350fps >> literally >> > becomes trivial. Include in tick rates deprecating this math and you're >> > literally down to nothing more than the CPU calculating the same crap >> over >> > and over and over. >> > >> > The most important thing is tick rate. In my opinion, if a blind study >> was >> > done NOBODY would be able to accurately tell the difference between a 100 >> > tick 350fps server compared to anything higher - ever. >> > >> > But that never stopped people from selling what the public wants to pay >> for >> > :-). Plus, the amount of placebo effect involved with 'pro' PC gamers is >> so >> > intense I have a hard time comprehending it. Simply telling them >> they've >> > switched to a 1000 FPS server should make them rave about how great it >> is. >> > The only REAL difference is what number comes up in tests. >> > >> > I'm sure gamers will never believe this and you'll find lines of people >> > waiting to tell you that they can see the difference between 350, 500, >> 1000 >> > and 2000 FPS servers. *sigh* >> > >> > Eric >> > >> > >> > On Mon, Sep 7, 2009 at 1:04 PM, Daniel Nilsson > >wrote: >> > >> > >> >> It´s very true that many do not see the difference from different >> >> ticrates. Mainly because thare are no big difference (Especialy from 66 >> >> to 100). Now several of you will protest. But the most important is the >> >> following >> >> >> >> 1. The fps must in must be stable. >> >> >> >> 2. The box must have an stabile networking to avoid changes in packages >> >> sent and recived. This part i have noticed on my own when playing >> >> online. Playing on a server for 5-10 min... hitting nothing. Take an >> >> look at the in and out on the netgraph and voila. They drop from 60 to >> >> 40, up to 50 and so on. >> >> >> >> Swithces to an more stable server and suddenly i hit everything, almost. >> >> But the funny part imo is that is it unstable for everyone or youst me >> >> because of my isp peering to that server isp or is it the same for >> >> everyone on that server. If it is the latest part. Why does some hit on >> >> those servers at all. >> >> >> >> >> >> //Daniel >> >> >> >> >> >> >> >> Ulrich Block skrev: >> >> >> >>> Same here. The most players can not even notice if they are playing on >> >>> tick 33, 66 or even 100. >> >>> Remember the valve defaults ar updaterate 20 cmdrate 30. Many of the >> >>> players out there do not know to alter this cvars, so that they can not >> >>> notice the tick >> >>> I run some public Servers at tick 33 with fps_max 300 and nobody ever >> >>> complained about bad hit registration. >> >>> The 1000fps thing is only for the wannabe pro players that are not able >> >>> to say its their fault they died. Besides there are no real 1000fps >> >>> only values close to that number like 950-990 fps. But again nobody >> will >> >>> notice the difference between 500 and 1000fps Servers. >> >>> Maybe some players from 250 to 500... >> >>> W
Re: [hlds_linux] Killing floor query
I second this. What is the PHP to list a server, of any game type on a website with an up to date status?? Eric On Tue, Sep 8, 2009 at 1:59 PM, Saint K. wrote: > Hi, > > Does anyone happen to have sampel code (php or so) to query KF servers to be > listed on a website? > > Cheers, > > ___ > 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] 1000 FPS CentOS Servers?
I was reading about 'if its really worth it' today. Apparently 500 FPS servers add 2ms of delay while 1000 FPS servers add 1ms delay. Also, 2000 FPS adds only .5ms. Considering it takes the human being 300ms to blink his eye, and i can do that while typing accurately... talking about anything over 350fps literally becomes trivial. Include in tick rates deprecating this math and you're literally down to nothing more than the CPU calculating the same crap over and over and over. The most important thing is tick rate. In my opinion, if a blind study was done NOBODY would be able to accurately tell the difference between a 100 tick 350fps server compared to anything higher - ever. But that never stopped people from selling what the public wants to pay for :-). Plus, the amount of placebo effect involved with 'pro' PC gamers is so intense I have a hard time comprehending it. Simply telling them they've switched to a 1000 FPS server should make them rave about how great it is. The only REAL difference is what number comes up in tests. I'm sure gamers will never believe this and you'll find lines of people waiting to tell you that they can see the difference between 350, 500, 1000 and 2000 FPS servers. *sigh* Eric On Mon, Sep 7, 2009 at 1:04 PM, Daniel Nilsson wrote: > It´s very true that many do not see the difference from different > ticrates. Mainly because thare are no big difference (Especialy from 66 > to 100). Now several of you will protest. But the most important is the > following > > 1. The fps must in must be stable. > > 2. The box must have an stabile networking to avoid changes in packages > sent and recived. This part i have noticed on my own when playing > online. Playing on a server for 5-10 min... hitting nothing. Take an > look at the in and out on the netgraph and voila. They drop from 60 to > 40, up to 50 and so on. > > Swithces to an more stable server and suddenly i hit everything, almost. > But the funny part imo is that is it unstable for everyone or youst me > because of my isp peering to that server isp or is it the same for > everyone on that server. If it is the latest part. Why does some hit on > those servers at all. > > > //Daniel > > > > Ulrich Block skrev: > > Same here. The most players can not even notice if they are playing on > > tick 33, 66 or even 100. > > Remember the valve defaults ar updaterate 20 cmdrate 30. Many of the > > players out there do not know to alter this cvars, so that they can not > > notice the tick > > I run some public Servers at tick 33 with fps_max 300 and nobody ever > > complained about bad hit registration. > > The 1000fps thing is only for the wannabe pro players that are not able > > to say its their fault they died. Besides there are no real 1000fps > > only values close to that number like 950-990 fps. But again nobody will > > notice the difference between 500 and 1000fps Servers. > > Maybe some players from 250 to 500... > > What i noticed is the fact, that the ping is slightly reduced when > > running at higher fps. > > > > Ferenc Kovacs schrieb: > > > >> If you was fine before you heard about the 1000fps thing, I think you > >> shouldnt bother. > >> The difference between stable ~250 and 1000 fps is almost zero from > >> the players perspective, but requires higher cpu cost, so you can have > >> fewer server on the same box. > >> just my 2c > >> > >> Tyrael > >> > >> On Mon, Sep 7, 2009 at 2:15 PM, Nevermore wrote: > >> > >> > >>> Very insteresting thread! > >>> > >>> I run a lot of servers in the same box with stock debian kernel. > >>> Wich optimization is suitable for me? > >>> > >>> > >>> Thank you all! > >>> > >>> > >>> > >>> ___ > >>> 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 > > > > > > > > __ Information from ESET NOD32 Antivirus, version of virus > signature database 4402 (20090907) __ > > > > The message was checked by ESET NOD32 Antivirus. > > > > http://www.eset.com > > > > > > > > > > __ Information from ESET NOD32 Antivirus, version of virus > signature database 4403 (20090907) __ > > The message was checked by ESET NOD32 Antivirus. > > http://www.eset.com > > > > ___ > To unsubscribe, edit your list preferences, or view the list archives, > please visit: > http://list.valvesoftware.com/mailman/listinfo/hlds_linux > _
Re: [hlds_linux] 1000 FPS CentOS Servers?
I JUST finished following the great walk through written at fragaholics.de here: http://wiki.fragaholics.de/index.php/EN:Linux_Kernel_Optimization <http://wiki.fragaholics.de/index.php/EN:Linux_Kernel_Optimization>IT WORKS!! http://www.fpsmeter.org/p,view;17127.html Thanks everyone! Eric On Mon, Sep 7, 2009 at 12:14 AM, Gary Stanley wrote: > At 11:04 PM 9/6/2009, Eric Greer wrote: > >Thanks everyone, especially Ulrich. You seem to be an expert with 1000 > FPS > >servers. > >I made the changes beyond modifying the kernel and ran some tests but I am > >only getting 400 FPS a the highest right now. I'm scared to do anything > >involving the kernel because blowing it up would be a horrible thing for > me. > > > >I've checked and the server processes are running at -99 priority and > >max_fps is set to 1000. > > > >[r...@atom ~]# rpm -qa | grep kernel > >kernel-PAE-2.6.18-128.el5 > >kernel-PAE-2.6.18-128.7.1.el5 > >kernel-PAE-devel-2.6.18-128.el5 > >kernel-headers-2.6.18-128.7.1.el5 > >kernel-PAE-devel-2.6.18-128.7.1.el5 > >[r...@atom ~]# > > > Your not going to be able to get ultra accurate timers with stock > CentOS kernels. You will need something with hrtimers merged into > them, and hardware > that support high res timers. > > > > ___ > 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] 1000 FPS CentOS Servers?
Thanks everyone, especially Ulrich. You seem to be an expert with 1000 FPS servers. I made the changes beyond modifying the kernel and ran some tests but I am only getting 400 FPS a the highest right now. I'm scared to do anything involving the kernel because blowing it up would be a horrible thing for me. I've checked and the server processes are running at -99 priority and max_fps is set to 1000. [r...@atom ~]# rpm -qa | grep kernel kernel-PAE-2.6.18-128.el5 kernel-PAE-2.6.18-128.7.1.el5 kernel-PAE-devel-2.6.18-128.el5 kernel-headers-2.6.18-128.7.1.el5 kernel-PAE-devel-2.6.18-128.7.1.el5 [r...@atom ~]# Anyone have any ideas what the safest way to modify my CentOS kernel is? Thanks a million everyone. Eric On Sun, Sep 6, 2009 at 3:40 PM, John wrote: > My experience differs from Monk's on this. I started using the CFS patches > before they were integrated into the mainline, and based on my (hundreds of > hours of) testing, CFS and newer kernels run just fine -- in fact, better > than older ones. But, I do apply my own patchset to them that affects their > behavior, and likely use a different .config than others, so YMMV. > > We do agree on staying away from RT kernels. > > -John > > - Original Message - > From: "Crazy Canucks" > To: "Half-Life dedicated Linux server mailing list" > > Sent: Sunday, September 06, 2009 4:09 AM > Subject: Re: [hlds_linux] 1000 FPS CentOS Servers? > > > > This discussion has me wondering. Why am I reading the newer kernels > > getting dissed? From what I read not too long ago the scheduler in the > > older kernels wasn't that great, and the new scheduler is much better. > > Is this not true? > > > > I'm not picking a fight here, I'm genuinely curious. > > > > Drek > > > > Gary Stanley wrote: > >> At 09:05 PM 9/5/2009, Eric Greer wrote: > >> > >>> This is all really awesome information everyone and I am very > >>> appreciative > >>> of all your ime and knowledge... however... > >>> > >>> What does this mean to the guy who hasn't recompiled a linux kenel > >>> before? > >>> Right now I'm seting fps_max on the command line to 500. Can I get more > >>> than > >>> 500 fps without recompiling? What settings would that require? > >>> > >>> If I do have to recompile, where do I start learning for that? How > >>> dangerous is it? > >>> > >>> Thanks again everyone, > >>> > >>> Eric > >>> > >> > >> The stock centos kernels do not have hrtimers, so you aren't able to > >> get low latency sleeping. The 2.6.18 kernels are very good latency > >> wise, so unless you want to patch them in you're going > >> to have to build something newer. The newer kernels with CFS are > >> pretty much crap, in my opinion, due to the scheduler changes and > >> other .. things, but mostly you will need HPET support in the BIOS > >> enabled with hrtimers. > >> > >> As Laws stated eariler, the older kernels are generally better > >> overall, and newer kernels with RT have too much overhead in most of > >> the codepath the game does not use. > >> > >> -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 > > > > > ___ > 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] 1000 FPS CentOS Servers?
This post helps an ass-load. So my next question is: Whats the best way to reliably tell what FPS a server is running at? Is there a way to monitor this? Thanks! Eric On Sun, Sep 6, 2009 at 2:23 AM, Ulrich Block wrote: > First you can try the fps your kernel can hanlde out of the box by using > fps_max 0 . The Debian kernel for example gets you the 1000fps out of > the box. > For security reasons i rebuild it with only the drivers i need and > loadable Modules support disabled (In my opinion the less code you can > execute on a server the better). > The most important thing is not to get high fps. The most important > thing is to get them stable. The most player do not realise a difference > between a 300 500 or even 1000 fps server. What they will easy notice is > when the serverfps will drop from 950-50 etc > > Here is a howto for a RT patched kernel: > http://wiki.fragaholics.de/index.php/EN:Linux_Kernel_Optimization > Just try out your kernel with the resheduling and idler part of this > tutorial. Maybe you get already the results you need with the standart > kernel. > > Eric Greer schrieb: > > This is all really awesome information everyone and I am very > appreciative > > of all your ime and knowledge... however... > > > > What does this mean to the guy who hasn't recompiled a linux kenel > before? > > Right now I'm seting fps_max on the command line to 500. Can I get more > than > > 500 fps without recompiling? What settings would that require? > > > > If I do have to recompile, where do I start learning for that? How > > dangerous is it? > > > > Thanks again everyone, > > > > Eric > > > > > > On Sat, Sep 5, 2009 at 7:46 PM, Gary Stanley >wrote: > > > > > >> At 09:50 AM 9/5/2009, Joseph Laws wrote: > >> > >>> I've never cared for the RT patches...but the hi-res timers pre 2.6.24 > >>> are very solid. > >>> > >> RT patches try and reduce the latency of a great multitude of things, > >> but the only ones that really count > >> are the scheduler latency. The 2.6.22 kernels without CFS are better > >> than the newer ones :) > >> > >> The best mainline kernels are the 2.4 series, because nanosleep will > >> busy wait. > >> > >> > >> > >> ___ > >> 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] 1000 FPS CentOS Servers?
This is all really awesome information everyone and I am very appreciative of all your ime and knowledge... however... What does this mean to the guy who hasn't recompiled a linux kenel before? Right now I'm seting fps_max on the command line to 500. Can I get more than 500 fps without recompiling? What settings would that require? If I do have to recompile, where do I start learning for that? How dangerous is it? Thanks again everyone, Eric On Sat, Sep 5, 2009 at 7:46 PM, Gary Stanley wrote: > At 09:50 AM 9/5/2009, Joseph Laws wrote: > >I've never cared for the RT patches...but the hi-res timers pre 2.6.24 > >are very solid. > > > RT patches try and reduce the latency of a great multitude of things, > but the only ones that really count > are the scheduler latency. The 2.6.22 kernels without CFS are better > than the newer ones :) > > The best mainline kernels are the 2.4 series, because nanosleep will > busy wait. > > > > ___ > 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
[hlds_linux] 1000 FPS CentOS Servers?
This goes to you too, Jason :-D So I've recently had some trouble getting CentOS to run 1000 FPS servers. I read online people seem to think you need a custom kernel compile to make this happen. What kind of adjustments need to be made and what settings must be in the config or command line to make this feat possible? Thanks everyone, Eric ___ 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 a server using FTP
just add -autoupdate to the command line at reboot the server Eric On Fri, Sep 4, 2009 at 6:44 AM, Nightbox wrote: > Is it possible ? > ___ > 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
Strange! I hadn't heard of or experienced any disk problems with ESXi yet. Maybe I just haven't loaded it down when I'm using it enough to notice. Eric On Thu, Aug 27, 2009 at 5:58 PM, Valtteri Kiviniemi < valtteri.kivini...@dataproof.fi> wrote: > Hi, > > ESXi 3.5, havent tested the 4.0 because Areca only has a beta driver for > it. > > - Valtteri Kiviniemi > > Eric Greer kirjoitti: > > vmware server or esxi? > > Eric > > > > > > On Thu, Aug 27, 2009 at 5:35 PM, Valtteri Kiviniemi < > > valtteri.kivini...@dataproof.fi> 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 se
Re: [hlds_linux] srcds virtualized
vmware server or esxi? Eric On Thu, Aug 27, 2009 at 5:35 PM, Valtteri Kiviniemi < valtteri.kivini...@dataproof.fi> 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 operati
Re: [hlds_linux] srcds virtualized
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 suitable to be used for game servers because they > should be > the > ones that introduce the lower overhead and response delay. > Having a minor performance loss is fine, as long as no noticeable > jitter is > introduced or ping is increased.Has anyone had a chance to test > these > products and compare srcds performance on the same machine when > virtualized > and when running on the bare metal? > Provided that the machine can handle it, do you know if it is > poss
