Re: [hlds_linux] Crashing L4D2 fork killing entire machine
I can confirm the 100% cpu usage by one process. It does not respond anymore, but it uses the whole cpu. Any workaround for this? I have had this issue twice up till now, and it's rather disturbing. What happened is that a L4D2 fork gets in a crashing mode of some kind where it continuously puts 100% CPU load on a single core. I've seen these sort of crashes perhaps around 10 times so far. 8 times I've spotted it soon enough to kill -9 pid the process, and 2 times I've been too late and it been in such a state for several hours before it will entirely kill off the machine. The machine will only reply to icmp and nothing else (ssh, http, ftp, gameservers etc). This worries me a bit, because I do not have any reason to assume something is broken on OS or hardware level. Regards, Saint K. ___ 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] Crashing L4D2 fork killing entire machine
At 04:21 AM 11/30/2009, Pavilus Zirovski wrote: It's happening about 3-4 times per month for me on 32bit Gentoo hosting TF2 servers only. Im not using -debug in startup line. I am using default preempt kernel (not realtime), but i was using realtime priority scheduler on srcds processes. I think maybe that is the problem. Now i removed resched.sh script from crontab, cause i thought that maybe it happens at some specific moments when cpu overload is very high and rescheduling script changes realtime priority of all srcds_processes (chrt -f -p 98 processid), but im not sure. Now i change realtime priority manually only. Server hasnt crashed for about 2 weeks for now, but i think it might crash any time. This has been frustrating for me as well, cause i lost all remote accesses to server, all processes start to hang one by one and cpu usage is maximized (all taken by one srcds process) at that moment till i lost connection from server at all and i have to call my hosting company so they could physically restart server. At first i thought it was because of some hardware failure or something but now seeing that others have the same problem then im not sure. I hope someone could give some more clues.. TBH you don't need SCHED_FIFO, you only need SCHED_RR. Gameservers are not time sensitive enough to justify running them at the same priority as, say, the network packet scheduler. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] Crashing L4D2 fork killing entire machine
At 04:21 AM 11/30/2009, Pavilus Zirovski wrote: It's happening about 3-4 times per month for me on 32bit Gentoo hosting TF2 servers only. Im not using -debug in startup line. I am using default preempt kernel (not realtime), but i was using realtime priority scheduler on srcds processes. I think maybe that is the problem. Now i removed resched.sh script from crontab, cause i thought that maybe it happens at some specific moments when cpu overload is very high and rescheduling script changes realtime priority of all srcds_processes (chrt -f -p 98 processid), but im not sure. Now i change realtime priority manually only. Server hasnt crashed for about 2 weeks for now, but i think it might crash any time. This has been frustrating for me as well, cause i lost all remote accesses to server, all processes start to hang one by one and cpu usage is maximized (all taken by one srcds process) at that moment till i lost connection from server at all and i have to call my hosting company so they could physically restart server. At first i thought it was because of some hardware failure or something but now seeing that others have the same problem then im not sure. I hope someone could give some more clues.. TBH you don't need SCHED_FIFO, you only need SCHED_RR. Gameservers are not time sensitive enough to justify running them at the same priority as, say, the network packet scheduler. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] Crashing L4D2 fork killing entire machine
Do you realy need chrt on l4d and l4d2? In my opinion no. Unlike the other source games it is limited to tick 30 and fps max. You can not change this values. So where is the point using chrt? I use it only for games like tf2 dods css and never had the problem you described. Gary Stanley schrieb: At 04:21 AM 11/30/2009, Pavilus Zirovski wrote: It's happening about 3-4 times per month for me on 32bit Gentoo hosting TF2 servers only. Im not using -debug in startup line. I am using default preempt kernel (not realtime), but i was using realtime priority scheduler on srcds processes. I think maybe that is the problem. Now i removed resched.sh script from crontab, cause i thought that maybe it happens at some specific moments when cpu overload is very high and rescheduling script changes realtime priority of all srcds_processes (chrt -f -p 98 processid), but im not sure. Now i change realtime priority manually only. Server hasnt crashed for about 2 weeks for now, but i think it might crash any time. This has been frustrating for me as well, cause i lost all remote accesses to server, all processes start to hang one by one and cpu usage is maximized (all taken by one srcds process) at that moment till i lost connection from server at all and i have to call my hosting company so they could physically restart server. At first i thought it was because of some hardware failure or something but now seeing that others have the same problem then im not sure. I hope someone could give some more clues.. TBH you don't need SCHED_FIFO, you only need SCHED_RR. Gameservers are not time sensitive enough to justify running them at the same priority as, say, the network packet scheduler. ___ 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] tf2 stable fps
Hello, I'm running 18 slots TF2 server on Debian Lenny stock kernel. And I wonder is it normal if server fps drops from 240 to 70-80 for few milliseconds? I see this with net_graph 4 enabled, while var doesn't show big difference always about 5-6msec. Does sourcetv affect this? I know recording STV demo is increase cpu load and I can see drops below 30 fps, but what about just broadcasting for additional 16 slots? ___ 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 stable fps
*/Does sourcetv affect this? I know recording STV demo is increase cpu load and I can see drops below 30 fps, but what about just broadcasting for additional 16 slots? /*It affects stable fps alot. Now i dont know your cpu and other hardware/internet limitation. But almost 99% of your unstable fps is probably from sourcetv. Run the fps meeter with and without sourcetv. And you will notice a big difference. //Daniel */ /* Nikolay Shopik skrev: Hello, I'm running 18 slots TF2 server on Debian Lenny stock kernel. And I wonder is it normal if server fps drops from 240 to 70-80 for few milliseconds? I see this with net_graph 4 enabled, while var doesn't show big difference always about 5-6msec. Does sourcetv affect this? I know recording STV demo is increase cpu load and I can see drops below 30 fps, but what about just broadcasting for additional 16 slots? ___ 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 4651 (20091201) __ The message was checked by ESET NOD32 Antivirus. http://www.eset.com __ Information from ESET NOD32 Antivirus, version of virus signature database 4651 (20091201) __ 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] tf2 stable fps
My hardware is core duo 6700 usually load up to 60-70% with 100Mbit connection. Fps meter, you mean net_graph 4? On 01.12.2009 16:29, Daniel Nilsson wrote: It affects stable fps alot. Now i dont know your cpu and other hardware/internet limitation. But almost 99% of your unstable fps is probably from sourcetv. Run the fps meeter with and without sourcetv. And you will notice a big difference ___ 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 stable fps
What kernel are you running? On Tue, 2009-12-01 at 17:54 +0300, Nikolay Shopik wrote: My hardware is core duo 6700 usually load up to 60-70% with 100Mbit connection. Fps meter, you mean net_graph 4? On 01.12.2009 16:29, Daniel Nilsson wrote: It affects stable fps alot. Now i dont know your cpu and other hardware/internet limitation. But almost 99% of your unstable fps is probably from sourcetv. Run the fps meeter with and without sourcetv. And you will notice a big difference ___ 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 stable fps
2.6.26-2-686 #1 SMP - This is stock Debian Lenny kernel On 01.12.2009 18:27, f0rkz wrote: What kernel are you running? ___ 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 stable fps
When it is stock I guess you did not disable cool and quiet yet (dunno what it is called with intel) As root try: cpufreq-set -g performance This sets the cpu to maximal standart Hz. Why do this? For the load peaks the cpufreq-utils do not scale up the CPU Hz fast enough. So it is better to disable them. Nikolay Shopik schrieb: 2.6.26-2-686 #1 SMP - This is stock Debian Lenny kernel On 01.12.2009 18:27, f0rkz wrote: What kernel are you running? ___ 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 stable fps
You could also look into compiling a dynamic kernel with 1000 hz frequency. On Tue, 2009-12-01 at 17:15 +0100, Ulrich Block wrote: When it is stock I guess you did not disable cool and quiet yet (dunno what it is called with intel) As root try: cpufreq-set -g performance This sets the cpu to maximal standart Hz. Why do this? For the load peaks the cpufreq-utils do not scale up the CPU Hz fast enough. So it is better to disable them. Nikolay Shopik schrieb: 2.6.26-2-686 #1 SMP - This is stock Debian Lenny kernel On 01.12.2009 18:27, f0rkz wrote: What kernel are you running? ___ 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 stable fps
Enabled, first test shows server drop fps from 247 to 245fps but this with just one player. Will see how this affect with fully loaded server. On 01.12.2009 19:15, Ulrich Block wrote: When it is stock I guess you did not disable cool and quiet yet (dunno what it is called with intel) As root try: cpufreq-set -g performance This sets the cpu to maximal standart Hz. Why do this? For the load peaks the cpufreq-utils do not scale up the CPU Hz fast enough. So it is better to disable them. Nikolay Shopik schrieb: 2.6.26-2-686 #1 SMP - This is stock Debian Lenny kernel On 01.12.2009 18:27, f0rkz wrote: What kernel are you running? ___ 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 stable fps
test this url http://www.fpsmeter.org/ And do a measurement there with sourcetv off. It requires rcon so it has to be enabled. Post the result of that measurement here... //Daniel Nikolay Shopik skrev: My hardware is core duo 6700 usually load up to 60-70% with 100Mbit connection. Fps meter, you mean net_graph 4? On 01.12.2009 16:29, Daniel Nilsson wrote: It affects stable fps alot. Now i dont know your cpu and other hardware/internet limitation. But almost 99% of your unstable fps is probably from sourcetv. Run the fps meeter with and without sourcetv. And you will notice a big difference ___ 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 4651 (20091201) __ The message was checked by ESET NOD32 Antivirus. http://www.eset.com __ Information from ESET NOD32 Antivirus, version of virus signature database 4651 (20091201) __ 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
[hlds_linux] sv_steamgroup = terribly pain
So my, as I see it, very simple server.cfg fails utterly to do what I want it to. I can't stop public people joining the server and no one in the steam group can ever see it. Anyone any ideas? Show tags has some interesting properties...notice the i at the end of the grp: also the Public always has coop set. The empty tag disappears when public peeps enjoy my paid for server! sv_showtags Tags: Public : coop,empty Private: g:l4d2,coop,realism,survival,versus,teamversus,scavenge,teamscavenge,grp:23762i, //Server Info hostname Dancingfrog rcon_password pass //Client Config sv_alltalk 0 //Lan, region, and network sv_lan 0 sv_region 3 //Steam group //sv_search_key df sv_steamgroup 23762 sv_steamgroup_exclusive 1 sv_allow_lobby_connect_only 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] sv_steamgroup = terribly pain
If you dont want public people joining, then give your server a search_key. That'll make it available only to lobbies that have that search key set. On Tue, Dec 1, 2009 at 1:38 PM, Matt Ford matt wrote: So my, as I see it, very simple server.cfg fails utterly to do what I want it to. I can't stop public people joining the server and no one in the steam group can ever see it. Anyone any ideas? Show tags has some interesting properties...notice the i at the end of the grp: also the Public always has coop set. The empty tag disappears when public peeps enjoy my paid for server! sv_showtags Tags: Public : coop,empty Private: g:l4d2,coop,realism,survival,versus,teamversus,scavenge,teamscavenge,grp:23762i, //Server Info hostname Dancingfrog rcon_password pass //Client Config sv_alltalk 0 //Lan, region, and network sv_lan 0 sv_region 3 //Steam group //sv_search_key df sv_steamgroup 23762 sv_steamgroup_exclusive 1 sv_allow_lobby_connect_only 1 ___ 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] sv_steamgroup = terribly pain
True, but sv_steamgroup_exclusive should make it invisible to public lobbies unless Steam group members are already on the server. I'm running two L4D2 Steam group servers with sv_steamgroup_exclusive = 1 and no search key set, and I do not get any public players on these servers, while my four public servers are full most of the time. I'm not sure why Steam group exclusivity seems to work for some people but not for others. Maybe this is a region problem? - Dave Brent Veal wrote: If you dont want public people joining, then give your server a search_key. That'll make it available only to lobbies that have that search key set. On Tue, Dec 1, 2009 at 1:38 PM, Matt Ford matt wrote: So my, as I see it, very simple server.cfg fails utterly to do what I want it to. I can't stop public people joining the server and no one in the steam group can ever see it. Anyone any ideas? Show tags has some interesting properties...notice the i at the end of the grp: also the Public always has coop set. The empty tag disappears when public peeps enjoy my paid for server! sv_showtags Tags: Public : coop,empty Private: g:l4d2,coop,realism,survival,versus,teamversus,scavenge,teamscavenge,grp:23762i, //Server Info hostname Dancingfrog rcon_password pass //Client Config sv_alltalk 0 //Lan, region, and network sv_lan 0 sv_region 3 //Steam group //sv_search_key df sv_steamgroup 23762 sv_steamgroup_exclusive 1 sv_allow_lobby_connect_only 1 ___ 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 -- Dave Parker Utica College Integrated Information Technology Services (315) 792-3229 Registered Linux User #408177 ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
[hlds_linux] L4D2 update (yeah again, right now)
Since no one posted official info about this, just for your information, there is a new update for left 4 dead 2 available for you all server running folks. * Fixed case where game failed to launch in certain machine configurations Fixed bug where a burning witch could get into a state where she's invincible * Fixed bug where bots would not shoot at Jockeys and Chargers while approaching to save a teammate -ics ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] sv_steamgroup = terribly pain
Well after todays update I can see my server back in the steam group list! Also by setting sv_steamgroup_exclusive to 2 (rather than 1) I can stop public people from joining the game at least until someone from the steam group is in first (it might be the case that setting it to 2 requires a steam group admin to start the game) This shouldn't be soo hard...or left so unexplained ___ 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 update (yeah again, right now)
Is Jason on vacation or something?! On Tuesday, December 1, 2009, ics i...@ics-base.net wrote: Since no one posted official info about this, just for your information, there is a new update for left 4 dead 2 available for you all server running folks. * Fixed case where game failed to launch in certain machine configurations Fixed bug where a burning witch could get into a state where she's invincible * Fixed bug where bots would not shoot at Jockeys and Chargers while approaching to save a teammate -ics ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- Thanks, - Saul. ___ 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 stable fps
I really dont want to get involved in this, but beware basically all the advice you're receiving (including mine). There's tons of BS regarding tickrate, RT kernels, fpsmeter.org, 'hit registration', and a bunch of other useless nonsense by people who really just are echoing what they read in that one old article about high FPS and what that one wiki says about linux kernel configuration. If I were you: Setup a kernel with hi res timers and a 300hz interrupt, enable HPET if your system doesn't have a stable TSC. Bind each srcds instance to a core. Set it to sched_fifo (sudo chrt -f -p 98 pid). Set fps_max to 0. Use host_profile 1 to watch effective FPS. Fill server. Play with other tweaks and compare them to your test case. Just because someone says that X or Y will make your server better dont believe it until you see it. net_graph 4 is the best measure of how well your server is doing. If its a solid graph with no gaps, getting 66/s updates, and low var, I would say its near perfect. Others might whine. Decide for yourself. Miscellaneous nonsense: - On a system with hi res timers and TSC/HPET, sleep() will return independent of the interrupt timer, enabling 1000FPS to be hit regardless of system ticrate. In this case, a 1000hz interrupt timer will not have any effect, and possibly a negative one. - On linux/tf2, the stats command calculates fps in a very useless manner. A single slow frame will make it show '40fps', while the engine's own internal counter (what you see in the green banner in those windows srcds windows) as well as host_profile disagree. - fpsmeter.org uses the stats command. - I've talked to and worked with many people and never seen a linux TF2 server above 20 slots get 'stable' FPS, much less according to fpsmeter. I've seen many TF2 linux servers that perform very well and lag free. - RT kernels chug CPU like no tomorrow for very little benefit, vs FIFO scheduling and hi-res timers. - If your var is 10ms and your updaterate is stable 66, to hell with anyone whining about FPS (flamewar lol). Its worth noting that windows servers are tuned to run at 66fps originally. By valve. The 'booster' came later. - My linux TF2 servers are among the best stability in updaterate and var i've seen anywhere, yet many people have 'more stable FPS' than me. See previous point. - SourceTV is a massive buggy resource hog. - Anyone that brings up 'hit registration' probably doesn't know wtf they're talking about and read some old article about it with questionable logic. - neph On 12/01/2009 11:21 AM, Daniel Nilsson wrote: test this url http://www.fpsmeter.org/ And do a measurement there with sourcetv off. It requires rcon so it has to be enabled. Post the result of that measurement here... //Daniel Nikolay Shopik skrev: My hardware is core duo 6700 usually load up to 60-70% with 100Mbit connection. Fps meter, you mean net_graph 4? On 01.12.2009 16:29, Daniel Nilsson wrote: It affects stable fps alot. Now i dont know your cpu and other hardware/internet limitation. But almost 99% of your unstable fps is probably from sourcetv. Run the fps meeter with and without sourcetv. And you will notice a big difference ___ 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 4651 (20091201) __ The message was checked by ESET NOD32 Antivirus. http://www.eset.com __ Information from ESET NOD32 Antivirus, version of virus signature database 4651 (20091201) __ 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 ___ 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 stable fps
useful comment as always. ty. Tyrael On Wed, Dec 2, 2009 at 12:07 AM, Nephyrin Zey nephy...@doublezen.net wrote: I really dont want to get involved in this, but beware basically all the advice you're receiving (including mine). There's tons of BS regarding tickrate, RT kernels, fpsmeter.org, 'hit registration', and a bunch of other useless nonsense by people who really just are echoing what they read in that one old article about high FPS and what that one wiki says about linux kernel configuration. If I were you: Setup a kernel with hi res timers and a 300hz interrupt, enable HPET if your system doesn't have a stable TSC. Bind each srcds instance to a core. Set it to sched_fifo (sudo chrt -f -p 98 pid). Set fps_max to 0. Use host_profile 1 to watch effective FPS. Fill server. Play with other tweaks and compare them to your test case. Just because someone says that X or Y will make your server better dont believe it until you see it. net_graph 4 is the best measure of how well your server is doing. If its a solid graph with no gaps, getting 66/s updates, and low var, I would say its near perfect. Others might whine. Decide for yourself. Miscellaneous nonsense: - On a system with hi res timers and TSC/HPET, sleep() will return independent of the interrupt timer, enabling 1000FPS to be hit regardless of system ticrate. In this case, a 1000hz interrupt timer will not have any effect, and possibly a negative one. - On linux/tf2, the stats command calculates fps in a very useless manner. A single slow frame will make it show '40fps', while the engine's own internal counter (what you see in the green banner in those windows srcds windows) as well as host_profile disagree. - fpsmeter.org uses the stats command. - I've talked to and worked with many people and never seen a linux TF2 server above 20 slots get 'stable' FPS, much less according to fpsmeter. I've seen many TF2 linux servers that perform very well and lag free. - RT kernels chug CPU like no tomorrow for very little benefit, vs FIFO scheduling and hi-res timers. - If your var is 10ms and your updaterate is stable 66, to hell with anyone whining about FPS (flamewar lol). Its worth noting that windows servers are tuned to run at 66fps originally. By valve. The 'booster' came later. - My linux TF2 servers are among the best stability in updaterate and var i've seen anywhere, yet many people have 'more stable FPS' than me. See previous point. - SourceTV is a massive buggy resource hog. - Anyone that brings up 'hit registration' probably doesn't know wtf they're talking about and read some old article about it with questionable logic. - neph On 12/01/2009 11:21 AM, Daniel Nilsson wrote: test this url http://www.fpsmeter.org/ And do a measurement there with sourcetv off. It requires rcon so it has to be enabled. Post the result of that measurement here... //Daniel Nikolay Shopik skrev: My hardware is core duo 6700 usually load up to 60-70% with 100Mbit connection. Fps meter, you mean net_graph 4? On 01.12.2009 16:29, Daniel Nilsson wrote: It affects stable fps alot. Now i dont know your cpu and other hardware/internet limitation. But almost 99% of your unstable fps is probably from sourcetv. Run the fps meeter with and without sourcetv. And you will notice a big difference ___ 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 4651 (20091201) __ The message was checked by ESET NOD32 Antivirus. http://www.eset.com __ Information from ESET NOD32 Antivirus, version of virus signature database 4651 (20091201) __ 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 ___ 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 stable fps
At 06:07 PM 12/1/2009, Nephyrin Zey wrote: I really dont want to get involved in this, but beware basically all the advice you're receiving (including mine). There's tons of BS regarding tickrate, RT kernels, fpsmeter.org, 'hit registration', and a bunch of other useless nonsense by people who really just are echoing what they read in that one old article about high FPS and what that one wiki says about linux kernel configuration. I agree. You do not need RT kernels at all.. they are for real time systems, since gameservers estimate everything anyways it's pointless to run a RT kernel on something that gives the best guess. The engine can only respond as fast as the users ping to and from the server, with socket latency etc.. If I were you: Setup a kernel with hi res timers and a 300hz interrupt, enable HPET if your system doesn't have a stable TSC. Bind each srcds instance to a core. Set it to sched_fifo (sudo chrt -f -p 98 pid). Set fps_max to 0. Use host_profile 1 to watch effective FPS. Fill server. Play with other tweaks and compare them to your test case. Just because someone says that X or Y will make your server better dont believe it until you see it. net_graph 4 is the best measure of how well your server is doing. If its a solid graph with no gaps, getting 66/s updates, and low var, I would say its near perfect. Others might whine. Decide for yourself. SCHED_RR gives the same latency as SCHED_FIFO, in my tests. Under load, this will be different, though. (send me a message and i'll send you some code) Miscellaneous nonsense: - On a system with hi res timers and TSC/HPET, sleep() will return independent of the interrupt timer, enabling 1000FPS to be hit regardless of system ticrate. In this case, a 1000hz interrupt timer will not have any effect, and possibly a negative one. AFAIK select()/poll() on older kernels do not use hrtimers at all. Only nanosleep()/usleep() do. You don't need 1000hz anyways as that can cause cacheline pingpongs etc (Hurt NUMA performance) - On linux/tf2, the stats command calculates fps in a very useless manner. A single slow frame will make it show '40fps', while the engine's own internal counter (what you see in the green banner in those windows srcds windows) as well as host_profile disagree. - fpsmeter.org uses the stats command. - I've talked to and worked with many people and never seen a linux TF2 server above 20 slots get 'stable' FPS, much less according to fpsmeter. I've seen many TF2 linux servers that perform very well and lag free. - RT kernels chug CPU like no tomorrow for very little benefit, vs FIFO scheduling and hi-res timers. - If your var is 10ms and your updaterate is stable 66, to hell with anyone whining about FPS (flamewar lol). Its worth noting that windows servers are tuned to run at 66fps originally. By valve. The 'booster' came later. - My linux TF2 servers are among the best stability in updaterate and var i've seen anywhere, yet many people have 'more stable FPS' than me. See previous point. - SourceTV is a massive buggy resource hog. - Anyone that brings up 'hit registration' probably doesn't know wtf they're talking about and read some old article about it with questionable logic. As long as you run Low Latency in the kernel with little interrupt behavior, and without CPUspeed/ACPI Processor you should be okay. Realtime kernels chug too much cpu because the scheduler has more overhead etc etc - ___ 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 stable fps
At 06:07 PM 12/1/2009, Nephyrin Zey wrote: I really dont want to get involved in this, but beware basically all the advice you're receiving (including mine). There's tons of BS regarding tickrate, RT kernels, fpsmeter.org, 'hit registration', and a bunch of other useless nonsense by people who really just are echoing what they read in that one old article about high FPS and what that one wiki says about linux kernel configuration. I agree. You do not need RT kernels at all.. they are for real time systems, since gameservers estimate everything anyways it's pointless to run a RT kernel on something that gives the best guess. The engine can only respond as fast as the users ping to and from the server, with socket latency etc.. If I were you: Setup a kernel with hi res timers and a 300hz interrupt, enable HPET if your system doesn't have a stable TSC. Bind each srcds instance to a core. Set it to sched_fifo (sudo chrt -f -p 98 pid). Set fps_max to 0. Use host_profile 1 to watch effective FPS. Fill server. Play with other tweaks and compare them to your test case. Just because someone says that X or Y will make your server better dont believe it until you see it. net_graph 4 is the best measure of how well your server is doing. If its a solid graph with no gaps, getting 66/s updates, and low var, I would say its near perfect. Others might whine. Decide for yourself. SCHED_RR gives the same latency as SCHED_FIFO, in my tests. Under load, this will be different, though. (send me a message and i'll send you some code) Miscellaneous nonsense: - On a system with hi res timers and TSC/HPET, sleep() will return independent of the interrupt timer, enabling 1000FPS to be hit regardless of system ticrate. In this case, a 1000hz interrupt timer will not have any effect, and possibly a negative one. AFAIK select()/poll() on older kernels do not use hrtimers at all. Only nanosleep()/usleep() do. You don't need 1000hz anyways as that can cause cacheline pingpongs etc (Hurt NUMA performance) - On linux/tf2, the stats command calculates fps in a very useless manner. A single slow frame will make it show '40fps', while the engine's own internal counter (what you see in the green banner in those windows srcds windows) as well as host_profile disagree. - fpsmeter.org uses the stats command. - I've talked to and worked with many people and never seen a linux TF2 server above 20 slots get 'stable' FPS, much less according to fpsmeter. I've seen many TF2 linux servers that perform very well and lag free. - RT kernels chug CPU like no tomorrow for very little benefit, vs FIFO scheduling and hi-res timers. - If your var is 10ms and your updaterate is stable 66, to hell with anyone whining about FPS (flamewar lol). Its worth noting that windows servers are tuned to run at 66fps originally. By valve. The 'booster' came later. - My linux TF2 servers are among the best stability in updaterate and var i've seen anywhere, yet many people have 'more stable FPS' than me. See previous point. - SourceTV is a massive buggy resource hog. - Anyone that brings up 'hit registration' probably doesn't know wtf they're talking about and read some old article about it with questionable logic. As long as you run Low Latency in the kernel with little interrupt behavior, and without CPUspeed/ACPI Processor you should be okay. Realtime kernels chug too much cpu because the scheduler has more overhead etc etc - ___ 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 stable fps
Thanks for clearing that up Neph :) On Tuesday, December 1, 2009, Nephyrin Zey nephy...@doublezen.net wrote: I really dont want to get involved in this, but beware basically all the advice you're receiving (including mine). There's tons of BS regarding tickrate, RT kernels, fpsmeter.org, 'hit registration', and a bunch of other useless nonsense by people who really just are echoing what they read in that one old article about high FPS and what that one wiki says about linux kernel configuration. If I were you: Setup a kernel with hi res timers and a 300hz interrupt, enable HPET if your system doesn't have a stable TSC. Bind each srcds instance to a core. Set it to sched_fifo (sudo chrt -f -p 98 pid). Set fps_max to 0. Use host_profile 1 to watch effective FPS. Fill server. Play with other tweaks and compare them to your test case. Just because someone says that X or Y will make your server better dont believe it until you see it. net_graph 4 is the best measure of how well your server is doing. If its a solid graph with no gaps, getting 66/s updates, and low var, I would say its near perfect. Others might whine. Decide for yourself. Miscellaneous nonsense: - On a system with hi res timers and TSC/HPET, sleep() will return independent of the interrupt timer, enabling 1000FPS to be hit regardless of system ticrate. In this case, a 1000hz interrupt timer will not have any effect, and possibly a negative one. - On linux/tf2, the stats command calculates fps in a very useless manner. A single slow frame will make it show '40fps', while the engine's own internal counter (what you see in the green banner in those windows srcds windows) as well as host_profile disagree. - fpsmeter.org uses the stats command. - I've talked to and worked with many people and never seen a linux TF2 server above 20 slots get 'stable' FPS, much less according to fpsmeter. I've seen many TF2 linux servers that perform very well and lag free. - RT kernels chug CPU like no tomorrow for very little benefit, vs FIFO scheduling and hi-res timers. - If your var is 10ms and your updaterate is stable 66, to hell with anyone whining about FPS (flamewar lol). Its worth noting that windows servers are tuned to run at 66fps originally. By valve. The 'booster' came later. - My linux TF2 servers are among the best stability in updaterate and var i've seen anywhere, yet many people have 'more stable FPS' than me. See previous point. - SourceTV is a massive buggy resource hog. - Anyone that brings up 'hit registration' probably doesn't know wtf they're talking about and read some old article about it with questionable logic. - neph On 12/01/2009 11:21 AM, Daniel Nilsson wrote: test this url http://www.fpsmeter.org/ And do a measurement there with sourcetv off. It requires rcon so it has to be enabled. Post the result of that measurement here... //Daniel Nikolay Shopik skrev: My hardware is core duo 6700 usually load up to 60-70% with 100Mbit connection. Fps meter, you mean net_graph 4? On 01.12.2009 16:29, Daniel Nilsson wrote: It affects stable fps alot. Now i dont know your cpu and other hardware/internet limitation. But almost 99% of your unstable fps is probably from sourcetv. Run the fps meeter with and without sourcetv. And you will notice a big difference ___ 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 4651 (20091201) __ The message was checked by ESET NOD32 Antivirus. http://www.eset.com __ Information from ESET NOD32 Antivirus, version of virus signature database 4651 (20091201) __ 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 ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- Thanks, - Saul. ___ 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 stable fps
On 12/01/2009 03:47 PM, Gary Stanley wrote: SCHED_RR gives the same latency as SCHED_FIFO, in my tests. Under load, this will be different, though. (send me a message and i'll send you some code) I use a multicore system with one core 'dedicated' to each instance. In this case, FIFO basically gives that core to the process - it gets to use it before anything else can think about being scheduled. SCHED_RR would be very similar in this case, if not identical on one process-per-core. RR might make more sense if two or more srcds's are sharing the core. Miscellaneous nonsense: - On a system with hi res timers and TSC/HPET, sleep() will return independent of the interrupt timer, enabling 1000FPS to be hit regardless of system ticrate. In this case, a 1000hz interrupt timer will not have any effect, and possibly a negative one. AFAIK select()/poll() on older kernels do not use hrtimers at all. Only nanosleep()/usleep() do. You don't need 1000hz anyways as that can cause cacheline pingpongs etc (Hurt NUMA performance) Indeed, this is all relevant to modern kernels. With older kernels, YMMV. - On linux/tf2, the stats command calculates fps in a very useless manner. A single slow frame will make it show '40fps', while the engine's own internal counter (what you see in the green banner in those windows srcds windows) as well as host_profile disagree. - fpsmeter.org uses the stats command. - I've talked to and worked with many people and never seen a linux TF2 server above 20 slots get 'stable' FPS, much less according to fpsmeter. I've seen many TF2 linux servers that perform very well and lag free. - RT kernels chug CPU like no tomorrow for very little benefit, vs FIFO scheduling and hi-res timers. - If your var is10ms and your updaterate is stable 66, to hell with anyone whining about FPS (flamewar lol). Its worth noting that windows servers are tuned to run at 66fps originally. By valve. The 'booster' came later. - My linux TF2 servers are among the best stability in updaterate and var i've seen anywhere, yet many people have 'more stable FPS' than me. See previous point. - SourceTV is a massive buggy resource hog. - Anyone that brings up 'hit registration' probably doesn't know wtf they're talking about and read some old article about it with questionable logic. As long as you run Low Latency in the kernel with little interrupt behavior, and without CPUspeed/ACPI Processor you should be okay. Realtime kernels chug too much cpu because the scheduler has more overhead etc etc - ___ 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 update (yeah again, right now)
Why o why is this not being announced anymore -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of ics Sent: Tuesday, December 01, 2009 11:44 PM To: Half-Life dedicated Linux server mailing list Subject: [hlds_linux] L4D2 update (yeah again, right now) Since no one posted official info about this, just for your information, there is a new update for left 4 dead 2 available for you all server running folks. * Fixed case where game failed to launch in certain machine configurations Fixed bug where a burning witch could get into a state where she's invincible * Fixed bug where bots would not shoot at Jockeys and Chargers while approaching to save a teammate -ics ___ 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: 9.0.709 / Virus Database: 270.14.87/2536 - Release Date: 12/01/09 08:59:00 ___ 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 update (yeah again, right now)
Probably the same reason they don't announce Killing Floor updates or similar games. I wish there was some consistency, but then again, that would make life too easy. If only we lived in a perfect world. Xavier ___ 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 update (yeah again, right now)
By the looks of it, they are client-side updates. Not really something that needs to be announced on a hlds_srcds list. There is actually a proper announce list if that's what you guys want. Saint K. wrote: Why o why is this not being announced anymore -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of ics Sent: Tuesday, December 01, 2009 11:44 PM To: Half-Life dedicated Linux server mailing list Subject: [hlds_linux] L4D2 update (yeah again, right now) Since no one posted official info about this, just for your information, there is a new update for left 4 dead 2 available for you all server running folks. * Fixed case where game failed to launch in certain machine configurations Fixed bug where a burning witch could get into a state where she's invincible * Fixed bug where bots would not shoot at Jockeys and Chargers while approaching to save a teammate -ics ___ 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: 9.0.709 / Virus Database: 270.14.87/2536 - Release Date: 12/01/09 08:59:00 ___ 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
Re: [hlds_linux] L4D2 update (yeah again, right now)
Well, this is a list that VALVE hosts, so it really only makes sense that updates to Valve games would be announced. Killing Floor is the only exception, since Tripwire feels that they should announce their updates here. Jonah Hirsch --- On Wed, Dec 2, 2009 at 12:44 AM, Xavier Kerestesy xav...@kerestesy.comwrote: Probably the same reason they don't announce Killing Floor updates or similar games. I wish there was some consistency, but then again, that would make life too easy. If only we lived in a perfect world. Xavier ___ 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
