Re: [hlds_linux] cpu on i7 920
I done a full reset, and went over all the options one by one, and researching the ones I didn't knew 100% sure what they were doing. I am assuming I ended up with a different set of settings because of the massive improvements. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of gamead...@127001.org [gamead...@127001.org] Sent: 27 July 2011 01:48 To: 'Half-Life dedicated Linux server mailing list' Subject: Re: [hlds_linux] cpu on i7 920 Wow! Did you change any settings, or just reset to default then put them back how they were? -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 26 July 2011 12:58 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 We ran around 90-100% per core with 40-ish FPS. Players found that reasonably acceptable, so I'm guessing no one is going to complain now. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Andres Pozos [javato...@yahoo.es] Sent: 26 July 2011 13:59 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 More than 80% cpu usage = lag spikes Everyone thanks a lot for your feedback regarding this issue. I went into the datacenter today and had another good look at the BIOS settings. In the end I did a full BIOS reset and re-programmed all the options. Servers now run at a rock solid 250FPS at around 80-90% CPU load per 24max server per core (acceptable). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Svensk Ljud Ljus Produktion [i...@teaterljud.se] Sent: 25 July 2011 17:00 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 To get a true picture on the cpu load you have to monitor both the machines cpu-usage by core, and the srcds instances usage of cpu. This can be monitored by munin. Offcource it can also monitor fps, no.off players, uptime and the network traffic on the nic by port. But dont forget to monitor the memory usage on the machine. Its a easy way to find errors. Peter Sweden Eric Riemers skrev 2011-07-25 15:04: I only have issues on the box with the 32 slots server, and since tf2 is build for 24 i dont think there will be any performance increases in this area. The 24 slots sit around 60/70% and i've got no complaints from there (except for the few that always complain) Also if you have hlxce running or similar you can usually see your serverside fps too, not that this matters that much but if you see drops below 66 it could be cpu, at least a reference point to look at if people complain about lag. Still running an older distro of debian on that box though, perhaps a upgrade would do some good i presume too. On Mon, 25 Jul 2011 14:05:10 +0200, Saint K.sai...@specialattack.net wrote: Hi, My concern is not really to have more servers on there (although it be nice if possible), but I'd like to get at least 66+fps stable per server per core, and having some load left to have replay etc enabled, as we had to kill that as well to get things running on the edge of normal. I've also tried assigning it per core, but never found clear bennifits to do so. The keeping one core free is just the thought process, leaves some room for other processes to peak (if required). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of gamead...@127001.org [gamead...@127001.org] Sent: 25 July 2011 13:56 To: 'Half-Life dedicated Linux server mailing list' Subject: Re: [hlds_linux] cpu on i7 920 Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea
Re: [hlds_linux] cpu on i7 920
Everyone thanks a lot for your feedback regarding this issue. I went into the datacenter today and had another good look at the BIOS settings. In the end I did a full BIOS reset and re-programmed all the options. Servers now run at a rock solid 250FPS at around 80-90% CPU load per 24max server per core (acceptable). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Svensk Ljud Ljus Produktion [i...@teaterljud.se] Sent: 25 July 2011 17:00 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 To get a true picture on the cpu load you have to monitor both the machines cpu-usage by core, and the srcds instances usage of cpu. This can be monitored by munin. Offcource it can also monitor fps, no.off players, uptime and the network traffic on the nic by port. But dont forget to monitor the memory usage on the machine. Its a easy way to find errors. Peter Sweden Eric Riemers skrev 2011-07-25 15:04: I only have issues on the box with the 32 slots server, and since tf2 is build for 24 i dont think there will be any performance increases in this area. The 24 slots sit around 60/70% and i've got no complaints from there (except for the few that always complain) Also if you have hlxce running or similar you can usually see your serverside fps too, not that this matters that much but if you see drops below 66 it could be cpu, at least a reference point to look at if people complain about lag. Still running an older distro of debian on that box though, perhaps a upgrade would do some good i presume too. On Mon, 25 Jul 2011 14:05:10 +0200, Saint K.sai...@specialattack.net wrote: Hi, My concern is not really to have more servers on there (although it be nice if possible), but I'd like to get at least 66+fps stable per server per core, and having some load left to have replay etc enabled, as we had to kill that as well to get things running on the edge of normal. I've also tried assigning it per core, but never found clear bennifits to do so. The keeping one core free is just the thought process, leaves some room for other processes to peak (if required). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of gamead...@127001.org [gamead...@127001.org] Sent: 25 July 2011 13:56 To: 'Half-Life dedicated Linux server mailing list' Subject: Re: [hlds_linux] cpu on i7 920 Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea if that was the OS or the server, and we've not tested again since, but we've had little issue leaving them to roam free. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 25 July 2011 12:22 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozosjavato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower
Re: [hlds_linux] cpu on i7 920
More than 80% cpu usage = lag spikes Everyone thanks a lot for your feedback regarding this issue. I went into the datacenter today and had another good look at the BIOS settings. In the end I did a full BIOS reset and re-programmed all the options. Servers now run at a rock solid 250FPS at around 80-90% CPU load per 24max server per core (acceptable). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Svensk Ljud Ljus Produktion [i...@teaterljud.se] Sent: 25 July 2011 17:00 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 To get a true picture on the cpu load you have to monitor both the machines cpu-usage by core, and the srcds instances usage of cpu. This can be monitored by munin. Offcource it can also monitor fps, no.off players, uptime and the network traffic on the nic by port. But dont forget to monitor the memory usage on the machine. Its a easy way to find errors. Peter Sweden Eric Riemers skrev 2011-07-25 15:04: I only have issues on the box with the 32 slots server, and since tf2 is build for 24 i dont think there will be any performance increases in this area. The 24 slots sit around 60/70% and i've got no complaints from there (except for the few that always complain) Also if you have hlxce running or similar you can usually see your serverside fps too, not that this matters that much but if you see drops below 66 it could be cpu, at least a reference point to look at if people complain about lag. Still running an older distro of debian on that box though, perhaps a upgrade would do some good i presume too. On Mon, 25 Jul 2011 14:05:10 +0200, Saint K.sai...@specialattack.net wrote: Hi, My concern is not really to have more servers on there (although it be nice if possible), but I'd like to get at least 66+fps stable per server per core, and having some load left to have replay etc enabled, as we had to kill that as well to get things running on the edge of normal. I've also tried assigning it per core, but never found clear bennifits to do so. The keeping one core free is just the thought process, leaves some room for other processes to peak (if required). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of gamead...@127001.org [gamead...@127001.org] Sent: 25 July 2011 13:56 To: 'Half-Life dedicated Linux server mailing list' Subject: Re: [hlds_linux] cpu on i7 920 Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea if that was the OS or the server, and we've not tested again since, but we've had little issue leaving them to roam free. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 25 July 2011 12:22 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozosjavato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd
Re: [hlds_linux] cpu on i7 920
We ran around 90-100% per core with 40-ish FPS. Players found that reasonably acceptable, so I'm guessing no one is going to complain now. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andres Pozos [javato...@yahoo.es] Sent: 26 July 2011 13:59 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 More than 80% cpu usage = lag spikes Everyone thanks a lot for your feedback regarding this issue. I went into the datacenter today and had another good look at the BIOS settings. In the end I did a full BIOS reset and re-programmed all the options. Servers now run at a rock solid 250FPS at around 80-90% CPU load per 24max server per core (acceptable). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Svensk Ljud Ljus Produktion [i...@teaterljud.se] Sent: 25 July 2011 17:00 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 To get a true picture on the cpu load you have to monitor both the machines cpu-usage by core, and the srcds instances usage of cpu. This can be monitored by munin. Offcource it can also monitor fps, no.off players, uptime and the network traffic on the nic by port. But dont forget to monitor the memory usage on the machine. Its a easy way to find errors. Peter Sweden Eric Riemers skrev 2011-07-25 15:04: I only have issues on the box with the 32 slots server, and since tf2 is build for 24 i dont think there will be any performance increases in this area. The 24 slots sit around 60/70% and i've got no complaints from there (except for the few that always complain) Also if you have hlxce running or similar you can usually see your serverside fps too, not that this matters that much but if you see drops below 66 it could be cpu, at least a reference point to look at if people complain about lag. Still running an older distro of debian on that box though, perhaps a upgrade would do some good i presume too. On Mon, 25 Jul 2011 14:05:10 +0200, Saint K.sai...@specialattack.net wrote: Hi, My concern is not really to have more servers on there (although it be nice if possible), but I'd like to get at least 66+fps stable per server per core, and having some load left to have replay etc enabled, as we had to kill that as well to get things running on the edge of normal. I've also tried assigning it per core, but never found clear bennifits to do so. The keeping one core free is just the thought process, leaves some room for other processes to peak (if required). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of gamead...@127001.org [gamead...@127001.org] Sent: 25 July 2011 13:56 To: 'Half-Life dedicated Linux server mailing list' Subject: Re: [hlds_linux] cpu on i7 920 Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea if that was the OS or the server, and we've not tested again since, but we've had little issue leaving them to roam free. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 25 July 2011 12:22 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozosjavato...@yahoo.es wrote: Its not a bug its
Re: [hlds_linux] cpu on i7 920
Wow! Did you change any settings, or just reset to default then put them back how they were? -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 26 July 2011 12:58 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 We ran around 90-100% per core with 40-ish FPS. Players found that reasonably acceptable, so I'm guessing no one is going to complain now. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Andres Pozos [javato...@yahoo.es] Sent: 26 July 2011 13:59 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 More than 80% cpu usage = lag spikes Everyone thanks a lot for your feedback regarding this issue. I went into the datacenter today and had another good look at the BIOS settings. In the end I did a full BIOS reset and re-programmed all the options. Servers now run at a rock solid 250FPS at around 80-90% CPU load per 24max server per core (acceptable). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Svensk Ljud Ljus Produktion [i...@teaterljud.se] Sent: 25 July 2011 17:00 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 To get a true picture on the cpu load you have to monitor both the machines cpu-usage by core, and the srcds instances usage of cpu. This can be monitored by munin. Offcource it can also monitor fps, no.off players, uptime and the network traffic on the nic by port. But dont forget to monitor the memory usage on the machine. Its a easy way to find errors. Peter Sweden Eric Riemers skrev 2011-07-25 15:04: I only have issues on the box with the 32 slots server, and since tf2 is build for 24 i dont think there will be any performance increases in this area. The 24 slots sit around 60/70% and i've got no complaints from there (except for the few that always complain) Also if you have hlxce running or similar you can usually see your serverside fps too, not that this matters that much but if you see drops below 66 it could be cpu, at least a reference point to look at if people complain about lag. Still running an older distro of debian on that box though, perhaps a upgrade would do some good i presume too. On Mon, 25 Jul 2011 14:05:10 +0200, Saint K.sai...@specialattack.net wrote: Hi, My concern is not really to have more servers on there (although it be nice if possible), but I'd like to get at least 66+fps stable per server per core, and having some load left to have replay etc enabled, as we had to kill that as well to get things running on the edge of normal. I've also tried assigning it per core, but never found clear bennifits to do so. The keeping one core free is just the thought process, leaves some room for other processes to peak (if required). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of gamead...@127001.org [gamead...@127001.org] Sent: 25 July 2011 13:56 To: 'Half-Life dedicated Linux server mailing list' Subject: Re: [hlds_linux] cpu on i7 920 Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea if that was the OS or the server, and we've not tested again since, but we've had little issue leaving them to roam free. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 25 July 2011 12:22 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint
Re: [hlds_linux] cpu on i7 920
What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 05:15 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I have a AMD Phenom x6 1055T doing multiple servers at the same time. TF2 causes the active core to go to 100% for about two seconds during map changes, but otherwise I've never seen it go that high for extended periods of time. Average during full 24-player usage is about 40%. I also have an older AMD Athlon64 x2 4600+ that runs a single 24-player TF2 quickplay server. It averages 70% usage when full and gameplay is great. Both of these are desktop class boards with DDR2 PC800M RAM. Linux 2.6.39. Try watching your CPU usage at (relatively) high resolution with something like htop -d 1 No clue why you are maxing out like that. Eric Riemers wrote: All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ 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] cpu on i7 920
Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 05:15 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I have a AMD Phenom x6 1055T doing multiple servers at the same time. TF2 causes the active core to go to 100% for about two seconds during map changes, but otherwise I've never seen it go that high for extended periods of time. Average during full 24-player usage is about 40%. I also have an older AMD Athlon64 x2 4600+ that runs a single 24-player TF2 quickplay server. It averages 70% usage when full and gameplay is great. Both of these are desktop class boards with DDR2 PC800M RAM. Linux 2.6.39. Try watching your CPU usage at (relatively) high resolution with something like htop -d 1 No clue why you are maxing out like that. Eric Riemers wrote: All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ 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] cpu on i7 920
The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 05:15 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I have a AMD Phenom x6 1055T doing multiple servers at the same time. TF2 causes the active core to go to 100% for about two seconds during map changes, but otherwise I've never seen it go that high for extended periods of time. Average during full 24-player usage is about 40%. I also have an older AMD Athlon64 x2 4600+ that runs a single 24-player TF2 quickplay server. It averages 70% usage when full and gameplay is great. Both of these are desktop class boards with DDR2 PC800M RAM. Linux 2.6.39. Try watching your CPU usage at (relatively) high resolution with something like htop -d 1 No clue why you are maxing out like that. Eric Riemers wrote: All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ 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] cpu on i7 920
I wouldn't expect problems with. But I happily admit to being no expert. Try ethtool -k interface and see what's what? A On 25/07/2011 08:42, Saint K. wrote: The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 05:15 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I have a AMD Phenom x6 1055T doing multiple servers at the same time. TF2 causes the active core to go to 100% for about two seconds during map changes, but otherwise I've never seen it go that high for extended periods of time. Average during full 24-player usage is about 40%. I also have an older AMD Athlon64 x2 4600+ that runs a single 24-player TF2 quickplay server. It averages 70% usage when full and gameplay is great. Both of these are desktop class boards with DDR2 PC800M RAM. Linux 2.6.39. Try watching your CPU usage at (relatively) high resolution with something like htop -d 1 No clue why you are maxing out like that. Eric Riemers wrote: All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] cpu on i7 920
Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: off udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off Does this appear to be good? I've checked the chipset specs and it supports checksum offloading. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 10:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 I wouldn't expect problems with. But I happily admit to being no expert. Try ethtool -k interface and see what's what? A On 25/07/2011 08:42, Saint K. wrote: The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 05:15 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I have a AMD Phenom x6 1055T doing multiple servers at the same time. TF2 causes the active core to go to 100% for about two seconds during map changes, but otherwise I've never seen it go that high for extended periods of time. Average during full 24-player usage is about 40%. I also have an older AMD Athlon64 x2 4600+ that runs a single 24-player TF2 quickplay server. It averages 70% usage when full and gameplay is great. Both of these are desktop class boards with DDR2 PC800M RAM. Linux 2.6.39. Try watching your CPU usage at (relatively) high resolution with something like htop -d 1 No clue why you are maxing out like that. Eric Riemers wrote: All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ 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
Re: [hlds_linux] cpu on i7 920
The network traffic that srcds uses is mostly udp, not tcp, so all the tcp driver offloading stuff is not involved here, for the most part. Any old network card should do. Not that the quality of the driver doesn't affect udp traffic. Some drivers can have trouble with large amounts of small udp traffic, or large udp traffic. The game server traffic here is quite modest though. I don't think anything networking related would be involved with the CPU problems (I'm a Cisco/Juniper network engineer and do high-bandwidth supercomputing stuff) Andrew Armitage wrote: Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] cpu on i7 920
Thanks for that. Are there any non-kernel tips people can give to look at? Because if I hear the loads of other people somehow ours are much higher, regardless of the kernels used. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 11:42 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 The network traffic that srcds uses is mostly udp, not tcp, so all the tcp driver offloading stuff is not involved here, for the most part. Any old network card should do. Not that the quality of the driver doesn't affect udp traffic. Some drivers can have trouble with large amounts of small udp traffic, or large udp traffic. The game server traffic here is quite modest though. I don't think anything networking related would be involved with the CPU problems (I'm a Cisco/Juniper network engineer and do high-bandwidth supercomputing stuff) Andrew Armitage wrote: Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ 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] cpu on i7 920
map used? is the problem happening on stock maps with no mods too? on a 100% stock 32slot server running at 500fps on a realtime preempt kernel max single core cpu usage i saw was 45%... Il 25/07/2011 11:49, Saint K. ha scritto: Thanks for that. Are there any non-kernel tips people can give to look at? Because if I hear the loads of other people somehow ours are much higher, regardless of the kernels used. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 11:42 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 The network traffic that srcds uses is mostly udp, not tcp, so all the tcp driver offloading stuff is not involved here, for the most part. Any old network card should do. Not that the quality of the driver doesn't affect udp traffic. Some drivers can have trouble with large amounts of small udp traffic, or large udp traffic. The game server traffic here is quite modest though. I don't think anything networking related would be involved with the CPU problems (I'm a Cisco/Juniper network engineer and do high-bandwidth supercomputing stuff) Andrew Armitage wrote: Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ 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] cpu on i7 920
Hi, Yes - default stock maps, we run no custom maps. There is no real difference between running it with sourcemod or without. We have our SM configured very lightly primarily just for administration purposes. They are so called Vanilla servers (defaults), 24 slots. We generally see high loads, and low FPS values. The machine is installed with Debian Squeeze (64bit), build on (imo) proper hardware, Tyan mobo's with Intel's 5400 chipset, 2 Quadcore E5410's in the case of the 90+% load, and E5420's in case of the 70-80% load per core. Machines are equipped with 16GB ECC 1333Mhz memory, using Cheetah 15K.6 SAS drives dedicated to the gameserver installs. Nothing else but TF2 servers run off the machine with the 5410's, the 5420's also run our databases etc. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Marco Padovan [e...@evcz.tk] Sent: 25 July 2011 12:12 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 map used? is the problem happening on stock maps with no mods too? on a 100% stock 32slot server running at 500fps on a realtime preempt kernel max single core cpu usage i saw was 45%... Il 25/07/2011 11:49, Saint K. ha scritto: Thanks for that. Are there any non-kernel tips people can give to look at? Because if I hear the loads of other people somehow ours are much higher, regardless of the kernels used. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 11:42 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 The network traffic that srcds uses is mostly udp, not tcp, so all the tcp driver offloading stuff is not involved here, for the most part. Any old network card should do. Not that the quality of the driver doesn't affect udp traffic. Some drivers can have trouble with large amounts of small udp traffic, or large udp traffic. The game server traffic here is quite modest though. I don't think anything networking related would be involved with the CPU problems (I'm a Cisco/Juniper network engineer and do high-bandwidth supercomputing stuff) Andrew Armitage wrote: Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ 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] cpu on i7 920
Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24-player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: off udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off Does this appear to be good? I've checked the chipset specs and it supports checksum offloading. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 10:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 I wouldn't expect problems with. But I happily admit to being no expert. Try ethtool -kinterface and see what's what? A On 25/07/2011 08:42, Saint K. wrote: The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 05:15 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I have a AMD Phenom x6 1055T doing multiple servers at the same time. TF2 causes the active core to go to 100% for about two seconds during map changes, but otherwise I've never seen it go that high for extended periods of time. Average during full 24-player usage is about 40%. I also have an older AMD Athlon64 x2 4600+ that runs a single 24-player TF2 quickplay server. It averages 70% usage when full and gameplay is great. Both of these are desktop class boards with DDR2 PC800M RAM. Linux 2.6.39. Try watching your CPU usage at (relatively) high resolution with something like htop -d 1 No clue why you are maxing out like that. Eric Riemers wrote: All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full
Re: [hlds_linux] cpu on i7 920
I've got a question about your kernel, is it patched with the RT-patch? I know that RT causes more stable fps but a lot higher cpu load. Date: Mon, 25 Jul 2011 03:36:10 -0700 From: je...@opendreams.net To: hlds_linux@list.valvesoftware.com CC: sai...@specialattack.net Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24-player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: off udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off Does this appear to be good? I've checked the chipset specs and it supports checksum offloading. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 10:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 I wouldn't expect problems with. But I happily admit to being no expert. Try ethtool -kinterface and see what's what? A On 25/07/2011 08:42, Saint K. wrote: The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 05:15 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I have a AMD Phenom x6 1055T doing multiple servers at the same time. TF2 causes the active core to go to 100% for about two seconds during map changes, but otherwise I've never seen it go that high for extended periods of time. Average during full 24-player usage is about 40%. I also have an older AMD Athlon64 x2 4600+ that runs a single 24-player TF2 quickplay server. It averages 70% usage when full and gameplay is great. Both of these are desktop class boards with DDR2 PC800M RAM
Re: [hlds_linux] cpu on i7 920
Hi, Not that I am aware of. We're currently running the stock kernel again; 2.6.32-5-amd64 I've tried several kernels suggested here before, but that didn't change anything either. I don't care much for high FPS, currently our servers are running at a very low 40-ish FPS on high load, and you can't notice too much performance issues. So I'd much rather run around 66-ish FPS and have a lower CPU load. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Michael Johansen [michs...@live.no] Sent: 25 July 2011 12:44 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 I've got a question about your kernel, is it patched with the RT-patch? I know that RT causes more stable fps but a lot higher cpu load. Date: Mon, 25 Jul 2011 03:36:10 -0700 From: je...@opendreams.net To: hlds_linux@list.valvesoftware.com CC: sai...@specialattack.net Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24-player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: off udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off Does this appear to be good? I've checked the chipset specs and it supports checksum offloading. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 10:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 I wouldn't expect problems with. But I happily admit to being no expert. Try ethtool -kinterface and see what's what? A On 25/07/2011 08:42, Saint K. wrote: The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf
Re: [hlds_linux] cpu on i7 920
Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24-player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: off udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off Does this appear to be good? I've checked the chipset specs and it supports checksum offloading. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 10:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 I wouldn't expect problems with. But I happily admit to being no expert. Try ethtool -kinterface and see what's what? A On 25/07/2011 08:42, Saint K. wrote: The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 05:15 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I have a AMD Phenom x6 1055T doing multiple servers at the same time. TF2 causes the active core to go to 100% for about two seconds during map changes, but otherwise I've never seen it go that high for extended periods of time. Average
Re: [hlds_linux] cpu on i7 920
Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24-player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: off udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off Does this appear to be good? I've checked the chipset specs and it supports checksum offloading. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 10:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 I wouldn't expect problems with. But I happily admit to being no expert. Try ethtool -kinterface and see what's what? A On 25/07/2011 08:42, Saint K. wrote: The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 05:15 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I have a AMD Phenom x6 1055T doing multiple servers at the same time. TF2 causes the active
Re: [hlds_linux] cpu on i7 920
Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozos javato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24-player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: off udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off Does this appear to be good? I've checked the chipset specs and it supports checksum offloading. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 10:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 I wouldn't expect problems with. But I happily admit to being no expert. Try ethtool -kinterface and see what's what? A On 25/07/2011 08:42, Saint K. wrote: The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out there. Is there perhaps certain BIOS settings which could benefit when running gameservers on them? Ours surely should perform much better then that? Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf
Re: [hlds_linux] cpu on i7 920
I tried to run 32 slots x2 on my Core2Duo dedicated server. It used 80% cpu and nothing more. However, fps drops were to 1 and shit, so i cba to host them. Date: Mon, 25 Jul 2011 12:14:45 +0100 From: javato...@yahoo.es To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24-player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: off udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off Does this appear to be good? I've checked the chipset specs and it supports checksum offloading. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 10:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 I wouldn't expect problems with. But I happily admit to being no expert. Try ethtool -kinterface and see what's what? A On 25/07/2011 08:42, Saint K. wrote: The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core for 24 players, and our Xeon E5410's 90%+, where you say your older 4600+ does 70%. Tried all sorts of different kernels out
Re: [hlds_linux] cpu on i7 920
I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozos javato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24-player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: off udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off Does this appear to be good? I've checked the chipset specs and it supports checksum offloading. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 10:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 I wouldn't expect problems with. But I happily admit to being no expert. Try ethtool -kinterface and see what's what? A On 25/07/2011 08:42, Saint K. wrote: The servers are build on Tyan Tempest i5400 motherboards, based on the Intel 5400B chipset platform, the Gbit nic's used on this board are Intel 82563EB chips. I've never really figured the load could be related to the networking chip as our throughput tests never really show any issues when tested (with all sorts of packet sizes, tcp/udp) Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Andrew Armitage [and...@thirdlife.org] Sent: 25 July 2011 09:36 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. A On 25/07/2011 07:28, Saint K. wrote: What I am still not getting is that our Xeon E5420's are doing like 70-80% load on a single core
Re: [hlds_linux] cpu on i7 920
How load averages and cpu usage depends a lot from the kernel. I believe individual process usage more than core load or load averages. For example, if you look with top program, you might see something like this: load average: 0.34, 0.42, 0.23 Cpu0 : 29.0%us, 0.5%sy, 0.0%ni, 66.8%id, 0.5%wa, 0.0%hi, 2.2%si, 1.0%st Cpu1 : 26.0%us, 0.7%sy, 0.0%ni, 70.4%id, 1.1%wa, 0.0%hi, 0.7%si, 1.1%st and when you look at the process list below that stuff, you might see the real values: PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 32094 game 20 0 493m 356m 11m R 61 17.8 638:30.00 srcds_linux 1499 game 20 0 487m 359m 11m S 54 17.9 516:56.58 srcds_linux So i'd rather believe the values in process list than the summary above. Then again, on our other bigger machine, the values are a bit different. The summary again: load average: 4.63, 5.07, 5.35 Cpu0 : 16.3%us, 0.0%sy, 0.0%ni, 82.4%id, 0.3%wa, 0.0%hi, 1.0%si, 0.0%st Cpu1 : 73.1%us, 0.7%sy, 0.0%ni, 25.6%id, 0.0%wa, 0.0%hi, 0.7%si, 0.0%st Cpu2 : 0.0%us, 0.0%sy, 0.0%ni, 97.7%id, 2.0%wa, 0.0%hi, 0.3%si, 0.0%st Cpu3 : 68.1%us, 1.0%sy, 0.0%ni, 30.2%id, 0.0%wa, 0.0%hi, 0.7%si, 0.0%st And the process info: PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 7339 game 20 0 392m 258m 21m S 78 3.3 91:41.36 srcds_linux 10288 game 20 0 399m 275m 21m R 73 3.5 38:46.18 srcds_linux 7262 game 20 0 317m 201m 15m S 18 2.5 195:01.39 srcds_linux 5686 game20 0 341m 157m 12m R2 2.0 11:07.86 srcds_linux As you can see, the differences are caused by different kernels. We had 2.6.37 or something where there was constant 16-20 load average while the cpu stats where the same, just load went bonkers. Never seen 100% unless the server is stuck like it has been couple of times since the lazor update. It might be i7 issue, we have older machines. -ics 25.7.2011 14:22, Saint K. kirjoitti: I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozosjavato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24-player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives
Re: [hlds_linux] cpu on i7 920
Very weird... I'm having some boxes running cheap desktop hardware (i7 930 for example) and doing very good (even with very extreme realtime kernels...) Never used debian in gameservers environment tho centos only here... Il 25/07/2011 12:28, Saint K. ha scritto: Hi, Yes - default stock maps, we run no custom maps. There is no real difference between running it with sourcemod or without. We have our SM configured very lightly primarily just for administration purposes. They are so called Vanilla servers (defaults), 24 slots. We generally see high loads, and low FPS values. The machine is installed with Debian Squeeze (64bit), build on (imo) proper hardware, Tyan mobo's with Intel's 5400 chipset, 2 Quadcore E5410's in the case of the 90+% load, and E5420's in case of the 70-80% load per core. Machines are equipped with 16GB ECC 1333Mhz memory, using Cheetah 15K.6 SAS drives dedicated to the gameserver installs. Nothing else but TF2 servers run off the machine with the 5410's, the 5420's also run our databases etc. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Marco Padovan [e...@evcz.tk] Sent: 25 July 2011 12:12 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 map used? is the problem happening on stock maps with no mods too? on a 100% stock 32slot server running at 500fps on a realtime preempt kernel max single core cpu usage i saw was 45%... Il 25/07/2011 11:49, Saint K. ha scritto: Thanks for that. Are there any non-kernel tips people can give to look at? Because if I hear the loads of other people somehow ours are much higher, regardless of the kernels used. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 11:42 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 The network traffic that srcds uses is mostly udp, not tcp, so all the tcp driver offloading stuff is not involved here, for the most part. Any old network card should do. Not that the quality of the driver doesn't affect udp traffic. Some drivers can have trouble with large amounts of small udp traffic, or large udp traffic. The game server traffic here is quite modest though. I don't think anything networking related would be involved with the CPU problems (I'm a Cisco/Juniper network engineer and do high-bandwidth supercomputing stuff) Andrew Armitage wrote: Maybe the issue is the network hardware? HLDS is very network intensive. I believe that some cards support checksum offloading and some don't. -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] cpu on i7 920
I use a combination of top and htop to monitor the servers usage. I generally ignore the load av. value, as you point out, per kernel this can be completely different. Generally I use top to view al the servers load per process, here's an example output of top with 5 loaded TF2 servers; 25668 tf2 -21 -20 486m 349m 21m R 97 4.4 177:05.31 srcds_linux 26408 tf2 -21 -20 647m 475m 21m R 93 5.9 696:37.88 srcds_linux 32345 tf2 -21 -20 442m 306m 21m R 90 3.8 133:28.53 srcds_linux 3297 tf2 -21 -20 397m 269m 21m R 85 3.4 38:07.04 srcds_linux 25736 tf2 -21 -20 533m 383m 21m S 68 4.8 180:19.68 srcds_linux Cpu0 : 72.3%us, 0.4%sy, 0.0%ni, 26.3%id, 0.0%wa, 0.0%hi, 0.9%si, 0.0%st Cpu1 : 73.3%us, 0.0%sy, 0.0%ni, 26.2%id, 0.0%wa, 0.0%hi, 0.4%si, 0.0%st Cpu2 : 66.1%us, 0.0%sy, 0.0%ni, 33.5%id, 0.0%wa, 0.0%hi, 0.5%si, 0.0%st Cpu3 : 70.2%us, 0.0%sy, 0.0%ni, 29.8%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Cpu4 : 76.2%us, 0.0%sy, 0.0%ni, 23.8%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Cpu5 : 71.6%us, 0.5%sy, 0.0%ni, 28.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Cpu6 : 68.5%us, 0.5%sy, 0.0%ni, 31.1%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Cpu7 : 67.1%us, 2.3%sy, 0.0%ni, 30.2%id, 0.0%wa, 0.0%hi, 0.5%si, 0.0%st load average: 6.84 Overall seen per cpu this is around 70-80% load steady, grouped together this comes to 73% overall load. htop shows 100+ CPU load per process, so not sure what to think of that, the overall load output per CPU however reads the same as top does. Mind you, this when 5 of the 7 servers are full. I generally assign 1 server per CPU core, and leave 1 core free for all OS stuff to be handled on. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of ics [i...@ics-base.net] Sent: 25 July 2011 13:42 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 How load averages and cpu usage depends a lot from the kernel. I believe individual process usage more than core load or load averages. For example, if you look with top program, you might see something like this: load average: 0.34, 0.42, 0.23 Cpu0 : 29.0%us, 0.5%sy, 0.0%ni, 66.8%id, 0.5%wa, 0.0%hi, 2.2%si, 1.0%st Cpu1 : 26.0%us, 0.7%sy, 0.0%ni, 70.4%id, 1.1%wa, 0.0%hi, 0.7%si, 1.1%st and when you look at the process list below that stuff, you might see the real values: PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 32094 game 20 0 493m 356m 11m R 61 17.8 638:30.00 srcds_linux 1499 game 20 0 487m 359m 11m S 54 17.9 516:56.58 srcds_linux So i'd rather believe the values in process list than the summary above. Then again, on our other bigger machine, the values are a bit different. The summary again: load average: 4.63, 5.07, 5.35 Cpu0 : 16.3%us, 0.0%sy, 0.0%ni, 82.4%id, 0.3%wa, 0.0%hi, 1.0%si, 0.0%st Cpu1 : 73.1%us, 0.7%sy, 0.0%ni, 25.6%id, 0.0%wa, 0.0%hi, 0.7%si, 0.0%st Cpu2 : 0.0%us, 0.0%sy, 0.0%ni, 97.7%id, 2.0%wa, 0.0%hi, 0.3%si, 0.0%st Cpu3 : 68.1%us, 1.0%sy, 0.0%ni, 30.2%id, 0.0%wa, 0.0%hi, 0.7%si, 0.0%st And the process info: PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 7339 game 20 0 392m 258m 21m S 78 3.3 91:41.36 srcds_linux 10288 game 20 0 399m 275m 21m R 73 3.5 38:46.18 srcds_linux 7262 game 20 0 317m 201m 15m S 18 2.5 195:01.39 srcds_linux 5686 game20 0 341m 157m 12m R2 2.0 11:07.86 srcds_linux As you can see, the differences are caused by different kernels. We had 2.6.37 or something where there was constant 16-20 load average while the cpu stats where the same, just load went bonkers. Never seen 100% unless the server is stuck like it has been couple of times since the lazor update. It might be i7 issue, we have older machines. -ics 25.7.2011 14:22, Saint K. kirjoitti: I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozosjavato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS
Re: [hlds_linux] cpu on i7 920
Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea if that was the OS or the server, and we've not tested again since, but we've had little issue leaving them to roam free. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 25 July 2011 12:22 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozos javato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24- player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: off udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off Does this appear to be good? I've checked the chipset specs and it supports checksum offloading. Saint K
Re: [hlds_linux] cpu on i7 920
I forgot to mention that some of the default maps are also pretty cpu itensive comparing to the others. Hoodoo, Frontier, Thundermountain, Hydro. There's couple of examples. Propably this is due to all of them having very large open areas within them like let's say, gold rush, dustbowl, etc narrow maps. It also depends on what effects the maps use and how they are built. Difference between dustbowl and hoodoo cpu usage can be 5-10% or even more. -ics 25.7.2011 14:56, gamead...@127001.org kirjoitti: Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea if that was the OS or the server, and we've not tested again since, but we've had little issue leaving them to roam free. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 25 July 2011 12:22 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozosjavato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24- player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi, I am getting these values returned, not entirely sure what I am looking at; mrblonde:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming
Re: [hlds_linux] cpu on i7 920
Hi, My concern is not really to have more servers on there (although it be nice if possible), but I'd like to get at least 66+fps stable per server per core, and having some load left to have replay etc enabled, as we had to kill that as well to get things running on the edge of normal. I've also tried assigning it per core, but never found clear bennifits to do so. The keeping one core free is just the thought process, leaves some room for other processes to peak (if required). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of gamead...@127001.org [gamead...@127001.org] Sent: 25 July 2011 13:56 To: 'Half-Life dedicated Linux server mailing list' Subject: Re: [hlds_linux] cpu on i7 920 Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea if that was the OS or the server, and we've not tested again since, but we've had little issue leaving them to roam free. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 25 July 2011 12:22 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozos javato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24- player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have
Re: [hlds_linux] cpu on i7 920
Knowing that each new cpu generation have more cores and less cpu cycles its pointless to keep using a engine that doesnt support multicore. I forgot to mention that some of the default maps are also pretty cpu itensive comparing to the others. Hoodoo, Frontier, Thundermountain, Hydro. There's couple of examples. Propably this is due to all of them having very large open areas within them like let's say, gold rush, dustbowl, etc narrow maps. It also depends on what effects the maps use and how they are built. Difference between dustbowl and hoodoo cpu usage can be 5-10% or even more. -ics 25.7.2011 14:56, gamead...@127001.org kirjoitti: Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea if that was the OS or the server, and we've not tested again since, but we've had little issue leaving them to roam free. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 25 July 2011 12:22 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozosjavato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24- player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers. I really have no idea what I'm talking about. I've only been messing with srcds servers for the last nine months or so. Then again... Seagate did ship me all of those SATA 300 drives with the 150-limiting jumpers on by default. Saint K. wrote: Hi
Re: [hlds_linux] cpu on i7 920
I only have issues on the box with the 32 slots server, and since tf2 is build for 24 i dont think there will be any performance increases in this area. The 24 slots sit around 60/70% and i've got no complaints from there (except for the few that always complain) Also if you have hlxce running or similar you can usually see your serverside fps too, not that this matters that much but if you see drops below 66 it could be cpu, at least a reference point to look at if people complain about lag. Still running an older distro of debian on that box though, perhaps a upgrade would do some good i presume too. On Mon, 25 Jul 2011 14:05:10 +0200, Saint K. sai...@specialattack.net wrote: Hi, My concern is not really to have more servers on there (although it be nice if possible), but I'd like to get at least 66+fps stable per server per core, and having some load left to have replay etc enabled, as we had to kill that as well to get things running on the edge of normal. I've also tried assigning it per core, but never found clear bennifits to do so. The keeping one core free is just the thought process, leaves some room for other processes to peak (if required). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of gamead...@127001.org [gamead...@127001.org] Sent: 25 July 2011 13:56 To: 'Half-Life dedicated Linux server mailing list' Subject: Re: [hlds_linux] cpu on i7 920 Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea if that was the OS or the server, and we've not tested again since, but we've had little issue leaving them to roam free. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 25 July 2011 12:22 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozos javato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24
Re: [hlds_linux] cpu on i7 920
On our servers we have noticed that cpu0 is used by the OP (Debian/Fedora) and accordingly we cant use that core for any gameservers. Peter Sweden ics skrev 2011-07-25 13:42: How load averages and cpu usage depends a lot from the kernel. I believe individual process usage more than core load or load averages. For example, if you look with top program, you might see something like this: load average: 0.34, 0.42, 0.23 Cpu0 : 29.0%us, 0.5%sy, 0.0%ni, 66.8%id, 0.5%wa, 0.0%hi, 2.2%si, 1.0%st Cpu1 : 26.0%us, 0.7%sy, 0.0%ni, 70.4%id, 1.1%wa, 0.0%hi, 0.7%si, 1.1%st and when you look at the process list below that stuff, you might see the real values: PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 32094 game 20 0 493m 356m 11m R 61 17.8 638:30.00 srcds_linux 1499 game 20 0 487m 359m 11m S 54 17.9 516:56.58 srcds_linux So i'd rather believe the values in process list than the summary above. Then again, on our other bigger machine, the values are a bit different. The summary again: load average: 4.63, 5.07, 5.35 Cpu0 : 16.3%us, 0.0%sy, 0.0%ni, 82.4%id, 0.3%wa, 0.0%hi, 1.0%si, 0.0%st Cpu1 : 73.1%us, 0.7%sy, 0.0%ni, 25.6%id, 0.0%wa, 0.0%hi, 0.7%si, 0.0%st Cpu2 : 0.0%us, 0.0%sy, 0.0%ni, 97.7%id, 2.0%wa, 0.0%hi, 0.3%si, 0.0%st Cpu3 : 68.1%us, 1.0%sy, 0.0%ni, 30.2%id, 0.0%wa, 0.0%hi, 0.7%si, 0.0%st And the process info: PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 7339 game 20 0 392m 258m 21m S 78 3.3 91:41.36 srcds_linux 10288 game 20 0 399m 275m 21m R 73 3.5 38:46.18 srcds_linux 7262 game 20 0 317m 201m 15m S 18 2.5 195:01.39 srcds_linux 5686 game20 0 341m 157m 12m R2 2.0 11:07.86 srcds_linux As you can see, the differences are caused by different kernels. We had 2.6.37 or something where there was constant 16-20 load average while the cpu stats where the same, just load went bonkers. Never seen 100% unless the server is stuck like it has been couple of times since the lazor update. It might be i7 issue, we have older machines. -ics 25.7.2011 14:22, Saint K. kirjoitti: I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozosjavato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect that to cause such significant problems. I have no clue. Grab an old cruddy desktop, set it up on your home network, do a quickplay qualified server, and then watch how it runs. Use the same OS and versions you are using on your server and then start twiddling. You only need about a 2Mbps upstream rate for a 24-player server. I would blame software way before I started blaming hardware. Actually, I'd blame something you did unknowingly, first, but just because I'm a BOFH. People like to throw switches in the desperate hope that one of them was put there by system developers specifically just to slow things down, like a turbo switch on an old 486DX. Surely, my sheer desire to make things go faster and by randomly throwing every bios setting, recompiling my kernel with obscure realtime patches I found, enabling weird sysctrl parameters, and buying a Bigfoot gaming NIC will make things go faster! And that's why we have fps_max 1000 and 100PPS update/cmd rates on servers
Re: [hlds_linux] cpu on i7 920
To get a true picture on the cpu load you have to monitor both the machines cpu-usage by core, and the srcds instances usage of cpu. This can be monitored by munin. Offcource it can also monitor fps, no.off players, uptime and the network traffic on the nic by port. But dont forget to monitor the memory usage on the machine. Its a easy way to find errors. Peter Sweden Eric Riemers skrev 2011-07-25 15:04: I only have issues on the box with the 32 slots server, and since tf2 is build for 24 i dont think there will be any performance increases in this area. The 24 slots sit around 60/70% and i've got no complaints from there (except for the few that always complain) Also if you have hlxce running or similar you can usually see your serverside fps too, not that this matters that much but if you see drops below 66 it could be cpu, at least a reference point to look at if people complain about lag. Still running an older distro of debian on that box though, perhaps a upgrade would do some good i presume too. On Mon, 25 Jul 2011 14:05:10 +0200, Saint K.sai...@specialattack.net wrote: Hi, My concern is not really to have more servers on there (although it be nice if possible), but I'd like to get at least 66+fps stable per server per core, and having some load left to have replay etc enabled, as we had to kill that as well to get things running on the edge of normal. I've also tried assigning it per core, but never found clear bennifits to do so. The keeping one core free is just the thought process, leaves some room for other processes to peak (if required). Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of gamead...@127001.org [gamead...@127001.org] Sent: 25 July 2011 13:56 To: 'Half-Life dedicated Linux server mailing list' Subject: Re: [hlds_linux] cpu on i7 920 Regardless of CPU usage you're going to struggle running more servers than that. The thing is, it's not that the servers use a lot of CPU*, it's just that when they need it, they need it _NOW_, or else they miss their window and the framerate drops. The more things running per core, the higher the chance that something else will stop a server getting its slice when it needs it. *actually, they _DO_, but the point would stand even if they didn't This is why I find load average useful; it can point out overloading (if LA number of threads your server has) even when the CPU load isn't showing 100% Re: assigning servers to cores - this is going back about 3 years, but I noticed that gameservers liked to jump cores about every 30 seconds or so... but if I pinned them to a core, they'd lag for a split-second around every 30 seconds or so instead. No idea if that was the OS or the server, and we've not tested again since, but we've had little issue leaving them to roam free. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Saint K. Sent: 25 July 2011 12:22 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I could live with that, at least knowing a reason why my 3000,- euro hardware can only sustain 7-ish 24 slots TF2 servers. Saint K. From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux- boun...@list.valvesoftware.com] On Behalf Of James Botting [bottswan...@googlemail.com] Sent: 25 July 2011 13:18 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Everytime a new weapon goes 'pew pew' the laser destroys a section of your CPU. It's a new feature. On 25/07/2011 12:14, Andres Pozosjavato...@yahoo.es wrote: Its not a bug its a feature, jk. Orangebox take teh 100% cpu usage in a core with more than 25 slots, i tested it on many linux distros, many kernel configurations and many cpus. So welcome to the club. Hi, Thanks for the reply. The servers are build according to Tyans best practise, so everything is inserted in the correct slots. I've recently also updated the BIOS versions to be sure. One thing I notice, at the memory settings is a snooping option - If this is disabled, the overall load is slightly lower (as it is now). I'd generally wouldn't blame the hardware either, however, seeing I can't find anything software wise, it's the logical next thing to look at. Any more tips are welcome! Saint K. From: Jesse Molina [je...@opendreams.net] Sent: 25 July 2011 12:36 To: Half-Life dedicated Linux server mailing list Cc: Saint K. Subject: Re: [hlds_linux] cpu on i7 920 Looks fine, don't mess with it. generic-receive-offload would be good if you were doing 10G networking, but otherwise forget about it. Make sure that your RAM is in the right slots as recommended by Tyan. That can slow things down sometimes but I would not expect
Re: [hlds_linux] cpu on i7 920
I thought maybe I misunderstood or you mis-typed previously. You are saying that your server-side FPS is very unstable, below 100? I thought maybe you meant 66 packets-per-second/ticks/cmdrate/updaterate. FYI, I believe that the standard fps_max for tf2 is 300. My cruddy old AMD x2 4600+ holds steady with very little variance while loaded. There is something seriously wrong with your setup if that's true. I have no idea what it might be. Just for fun, apt-get install adjtimex and do adjtimex -p And maybe try running a few phoronics tests; sudo apt-cache show phoronix-test-suite install and do some memory and CPU tests. Compare with similar hardware to see what you are getting. As for setting process affinity, I doubt that there would be any benefit for an application like srcds. AMD processors for each core have their own cache, unlike Intel, so sometimes cache misses can hurt you for cache-intensive applications, but modern Linux schedulers take that into account and are pretty good about it (if I'm remembering right). There are also some cases where if you want maximum IO/network throughput (10G+ networking), it can be useful, but srcds isn't one of them. Per-CPU licensed apps (Oracle?) can also be a good reason to bind an app to a particular CPU. The issue with setting a process to a particular CPU affinity is that it does not provide exclusivity. Any other process will still run on that CPU too, potentially fighting for resources. Saint K. wrote: Hi, My concern is not really to have more servers on there (although it be nice if possible), but I'd like to get at least 66+fps stable per server per core, and having some load left to have replay etc enabled, as we had to kill that as well to get things running on the edge of normal. I've also tried assigning it per core, but never found clear bennifits to do so. The keeping one core free is just the thought process, leaves some room for other processes to peak (if required). Saint K. -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] cpu on i7 920
The cpu usage in Linux is not correct. Do uptime and see what terminal value or something like that you get after you've ran it a few times. From: riem...@binkey.nl To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 13:00:11 +0200 Subject: [hlds_linux] cpu on i7 920 All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ 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] cpu on i7 920
Ok, from console then I ran a few stats after each other. CPU InOut Uptime Users FPSPlayers 99.03 117488.39 310790.97 491 0 95.55 31 So I see a 99% cpu in this one, surely it might not be the best way to check it but if I do this on other servers I don't really have an issue. In some cases I see less cpu usage but also below 66 values of fps. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Michael Johansen Sent: zondag 24 juli 2011 13:06 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 The cpu usage in Linux is not correct. Do uptime and see what terminal value or something like that you get after you've ran it a few times. From: riem...@binkey.nl To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 13:00:11 +0200 Subject: [hlds_linux] cpu on i7 920 All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ 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] cpu on i7 920
First off, to save CPU-usage, set fps_max to 66.67. This will ensure optimal hitreg. It's the best to run with atm. And don't care about the cpu readings. They are incorrect. From: riem...@binkey.nl To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 13:20:02 +0200 Subject: Re: [hlds_linux] cpu on i7 920 Ok, from console then I ran a few stats after each other. CPU InOut Uptime Users FPSPlayers 99.03 117488.39 310790.97 491 0 95.55 31 So I see a 99% cpu in this one, surely it might not be the best way to check it but if I do this on other servers I don't really have an issue. In some cases I see less cpu usage but also below 66 values of fps. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Michael Johansen Sent: zondag 24 juli 2011 13:06 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 The cpu usage in Linux is not correct. Do uptime and see what terminal value or something like that you get after you've ran it a few times. From: riem...@binkey.nl To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 13:00:11 +0200 Subject: [hlds_linux] cpu on i7 920 All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ 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] cpu on i7 920
I have a question about that. Does setting the fps_max affect cpu usage? I've tried setting various fps_max settings and cpu usage doesn't change. I mean that on HT kernel there is no difference between 66 and 500 or 950fps (when it was available). On Sun, 24 Jul 2011 14:26:28 +0300, Michael Johansen michs...@live.no wrote: First off, to save CPU-usage, set fps_max to 66.67. This will ensure optimal hitreg. It's the best to run with atm. And don't care about the cpu readings. They are incorrect. From: riem...@binkey.nl To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 13:20:02 +0200 Subject: Re: [hlds_linux] cpu on i7 920 Ok, from console then I ran a few stats after each other. CPU InOut Uptime Users FPSPlayers 99.03 117488.39 310790.97 491 0 95.55 31 So I see a 99% cpu in this one, surely it might not be the best way to check it but if I do this on other servers I don't really have an issue. In some cases I see less cpu usage but also below 66 values of fps. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Michael Johansen Sent: zondag 24 juli 2011 13:06 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 The cpu usage in Linux is not correct. Do uptime and see what terminal value or something like that you get after you've ran it a few times. From: riem...@binkey.nl To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 13:00:11 +0200 Subject: [hlds_linux] cpu on i7 920 All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ 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 -- Evaldas, EVAgames community www.evagames.eu ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] cpu on i7 920
I don't know tbh. I do know that there is a difference between all those values. To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 15:19:00 +0300 From: hlds_li...@evagames.eu Subject: Re: [hlds_linux] cpu on i7 920 I have a question about that. Does setting the fps_max affect cpu usage? I've tried setting various fps_max settings and cpu usage doesn't change. I mean that on HT kernel there is no difference between 66 and 500 or 950fps (when it was available). On Sun, 24 Jul 2011 14:26:28 +0300, Michael Johansen michs...@live.no wrote: First off, to save CPU-usage, set fps_max to 66.67. This will ensure optimal hitreg. It's the best to run with atm. And don't care about the cpu readings. They are incorrect. From: riem...@binkey.nl To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 13:20:02 +0200 Subject: Re: [hlds_linux] cpu on i7 920 Ok, from console then I ran a few stats after each other. CPU InOut Uptime Users FPSPlayers 99.03 117488.39 310790.97 491 0 95.55 31 So I see a 99% cpu in this one, surely it might not be the best way to check it but if I do this on other servers I don't really have an issue. In some cases I see less cpu usage but also below 66 values of fps. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Michael Johansen Sent: zondag 24 juli 2011 13:06 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 The cpu usage in Linux is not correct. Do uptime and see what terminal value or something like that you get after you've ran it a few times. From: riem...@binkey.nl To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 13:00:11 +0200 Subject: [hlds_linux] cpu on i7 920 All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ 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 -- Evaldas, EVAgames community www.evagames.eu ___ 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] cpu on i7 920
I see no difference either in setting fps_max on any kernel I've tried. CPU load always remains high. We noticed after last updates CPU usage going up as well, but haven't really gotten any reports on degraded performance, so I'm just keeping it quiet hoping noone notices :) From: hlds_linux-boun...@list.valvesoftware.com [hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Evaldas [hlds_li...@evagames.eu] Sent: 24 July 2011 14:19 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 I have a question about that. Does setting the fps_max affect cpu usage? I've tried setting various fps_max settings and cpu usage doesn't change. I mean that on HT kernel there is no difference between 66 and 500 or 950fps (when it was available). On Sun, 24 Jul 2011 14:26:28 +0300, Michael Johansen michs...@live.no wrote: First off, to save CPU-usage, set fps_max to 66.67. This will ensure optimal hitreg. It's the best to run with atm. And don't care about the cpu readings. They are incorrect. From: riem...@binkey.nl To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 13:20:02 +0200 Subject: Re: [hlds_linux] cpu on i7 920 Ok, from console then I ran a few stats after each other. CPU InOut Uptime Users FPSPlayers 99.03 117488.39 310790.97 491 0 95.55 31 So I see a 99% cpu in this one, surely it might not be the best way to check it but if I do this on other servers I don't really have an issue. In some cases I see less cpu usage but also below 66 values of fps. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of Michael Johansen Sent: zondag 24 juli 2011 13:06 To: hlds_linux@list.valvesoftware.com Subject: Re: [hlds_linux] cpu on i7 920 The cpu usage in Linux is not correct. Do uptime and see what terminal value or something like that you get after you've ran it a few times. From: riem...@binkey.nl To: hlds_linux@list.valvesoftware.com Date: Sun, 24 Jul 2011 13:00:11 +0200 Subject: [hlds_linux] cpu on i7 920 All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ 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 -- Evaldas, EVAgames community www.evagames.eu ___ 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] cpu on i7 920
I can confirm that a 32 slots TF2 server maxes out a core of a i7 920 or a xeon w3520 on linux On Sun, Jul 24, 2011 at 1:00 PM, Eric Riemers riem...@binkey.nl wrote: All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ 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] cpu on i7 920
Our i5 750 (same clockspeed) runs around 40% when full at 24 slots, though, as people have said, uptime can give you a better idea if the box is struggling or not. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Eric Riemers Sent: 24 July 2011 12:00 To: 'Half-Life dedicated Linux server mailing list' Subject: [hlds_linux] cpu on i7 920 All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ 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] cpu on i7 920
Uptime? You mean the same load averages as top shows? -ics 24.7.2011 20:27, gamead...@127001.org kirjoitti: Our i5 750 (same clockspeed) runs around 40% when full at 24 slots, though, as people have said, uptime can give you a better idea if the box is struggling or not. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Eric Riemers Sent: 24 July 2011 12:00 To: 'Half-Life dedicated Linux server mailing list' Subject: [hlds_linux] cpu on i7 920 All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ 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] cpu on i7 920
I presume so, perhaps they mean the total amount of the box. On average the load is 3 on the box, since it's a i7 I roughly estimate 8 is the max (and yes memory free, and 4gb cached, and iotop shows me nothing struggling) But since Beretta also confirms this I think I am out of luck, I do have another 24/7 running on the same hardware with 32 slots but that runs fine but it's a small map (trade map) which doesn't hit the 100% .. combination of map size + players + pew pew that makes it a 100 I presume. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux-boun...@list.valvesoftware.com] On Behalf Of ics Sent: zondag 24 juli 2011 19:37 To: Half-Life dedicated Linux server mailing list Subject: Re: [hlds_linux] cpu on i7 920 Uptime? You mean the same load averages as top shows? -ics 24.7.2011 20:27, gamead...@127001.org kirjoitti: Our i5 750 (same clockspeed) runs around 40% when full at 24 slots, though, as people have said, uptime can give you a better idea if the box is struggling or not. -Original Message- From: hlds_linux-boun...@list.valvesoftware.com [mailto:hlds_linux- boun...@list.valvesoftware.com] On Behalf Of Eric Riemers Sent: 24 July 2011 12:00 To: 'Half-Life dedicated Linux server mailing list' Subject: [hlds_linux] cpu on i7 920 All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ 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] cpu on i7 920
I have a AMD Phenom x6 1055T doing multiple servers at the same time. TF2 causes the active core to go to 100% for about two seconds during map changes, but otherwise I've never seen it go that high for extended periods of time. Average during full 24-player usage is about 40%. I also have an older AMD Athlon64 x2 4600+ that runs a single 24-player TF2 quickplay server. It averages 70% usage when full and gameplay is great. Both of these are desktop class boards with DDR2 PC800M RAM. Linux 2.6.39. Try watching your CPU usage at (relatively) high resolution with something like htop -d 1 No clue why you are maxing out like that. Eric Riemers wrote: All, I run a 32 slots server on a i7 920 @ 2.67ghz, but i can see with top and such that its potentially maxing out at 100% at times since it only uses one core. Is it really now doing so much cpu that even a i7 core isn't enough? Didn't have much complaints before the pew pew. 24 slots on the same server do around 60% when full. Eric ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- # Jesse Molina # Mail = je...@opendreams.net # Page = page-je...@opendreams.net # Cell = 1.602.323.7608 # Web = http://www.opendreams.net/jesse/ ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux