TF2 is most definitely not multi-threaded.
[ЯтR] The-/ Sorry for reviving an old thread but.
> dual AMD opteron 2382 (shanghai) @ 2.6ghz each
> 16gb kingston ddr2 667
> CentOS 5.3 with kernel at 1000hz
> CSS 32slot 100tick maybe 50% cpu usage when full, is multithreaded though?
> DoDS 32slot
When I first updated (when update just came out) on 4 different
installs, it only changed 4 files.
I just went back and run the updater again and there was another 5 to 8
that were updated.
So how files were changed exactly?
- Dan
Jason Ruymen wrote:
> Try re-running the update command. We m
L 03/10/2009 - 20:02:02: "[redacted
username<29>" changed role to "spy"
./srcds_run: line 351: 5479 Segmentation fault $HL_CMD
Add "-debug" to the ./srcds_run command line to generate a debug.log to
help with solving this problem
Tue Mar 10 20:02:02 PDT 2009: Server restart in 5 seconds
I'
Your sure? I'm on a Q9300 with a 1000hz prempt kernel and bind each 32
man server to1 core and haven't gotten any complaints.
What kind of issues have you experienced with that?
On Mar 8, 2009, at 12:50 PM, Cc2iscooL wrote:
> Don't bind tf2 to one core. it creates issues.
>
> On 3/8/09, Ric
Oops, thats right.
Taavi wrote:
> The Universes wrote:
>> cat hpet > /sys/devices/system/clocksource/clocksource0/current_clocksourc
>
> You mean:
>
> echo hpet > /sys/devices/system/clocksource/clocksource0/current_cloc
cat hpet > /sys/devices/system/clocksource/clocksource0/current_clocksource
Ben B wrote:
> I feel like such a linux nub... how do you set your clocksource?
>
> On Fri, Mar 6, 2009 at 11:56 PM, The Universes wrote:
>
>> Do you think dropping it down to 500HZ is a be
Do you think dropping it down to 500HZ is a better choice or do you
suggest something else? If yes, does 500HZ still maintain 500FPS?
Thanks!
Gary Stanley wrote:
> At 10:33 PM 3/6/2009, The Universes wrote:
>> I'm running 4 32 man TF2 servers at 500fps (66 tick) and I'm won
Would changing to HPET help at all?
ServerAlex wrote:
> Its not the timing that wastes the CPU time. You can't lower it with that.
> 90% is absolutly normal for such a big server.
>
> 2009/3/7 The Universes :
>> I'm running 4 32 man TF2 servers at 500fps (66 t
I'm running 4 32 man TF2 servers at 500fps (66 tick) and I'm wondering
if having kernel at 1000HZ (not tickless) with PREEMPT is necessary?
Is that putting unnecessary strain on my CPU (Q9300)?
My second question is if you guys use HPET as your clocksource
$ cat /sys/devices/system/clocksource/cl
It's alway said .dll for that particular one on Linux.
On Mar 5, 2009, at 6:24 PM, Pawel wrote:
> Okay it loaded. But now starting the server is so long.
> But it works ;)
>
>
>
> On Fri, Mar 6, 2009 at 3:14 AM, Rick Payton
> wrote:
>
>> I thought srcds in linux always referenced the dll fi
from ~980 to ~60))
>>
>> 2009/3/5 DontWannaName!
>>
>>> I would say that those numbers sound about right, more or less. 90
>> max.
>>> On Thu, Mar 5, 2009 at 12:59 AM, The Universes
>> >>> wrote:
>>>> Is it normal for a 32 man TF2
Is it normal for a 32 man TF2 (Badwater) server to take up 80-90% of the
CPU on Q9300 (Quad 2.5ghz) when the server is full and theres a lot of
action going on?
I recompiled my kernel according to the Fragaholics wiki guide.
Kernel 2.6.28.7 #1 SMP PREEMPT Wed Mar 4 17:36:14 PST 2009 x86_64 x86_6
Try rebooting your machine. I've had the segfault issue since that one
update that started it all and rebooting seems to help (for the recent
updates).
On a side note, even with Spy disabled, I still get occasional crashes
on my servers. I checked the logs, there was just normal activity, and
boom
Confirmed, same here.
Bruce Potter wrote:
> As a follow up, I'm still having the issue where the server segfaults
> repeatedly at start time and eventually takes. But now I also have an
> issue where it segfaults once it's running. Goes for about 5
> minutes.. Get a pretty full server, th
I'm getting this message in the console a lot:
CreateFragmentsFromFile: 'downloads/644b6ec3.dat' doesn't exist.
JäKë T wrote:
> I thought you're using windows. :)
>
> This is for linux,
>
>
>
> Anyways people are still getting disconnect from my servers at random.
>
> I'm seeing a lot of
Same here.
Fyren wrote:
> As ever since the January 28th update, my Linux server still segfaults
> on startup more often than it goes up.
>
> -Fyren
>
> ___
> To unsubscribe, edit your list preferences, or view the list archives, please
> visit:
> htt
wrote:
> I find this even more annoying since they were asking for straces / dump
> logs, and once it's been given, no sign...
>
> On Wed, Feb 18, 2009 at 1:25 AM, The Universes wrote:
>
>> Seriously, thats a huge annoyance for those of us affected by it.
>>
>> I
Seriously, thats a huge annoyance for those of us affected by it.
It still starts up, but only after a few seg faults.
On Feb 17, 2009, at 4:03 PM, Flubber wrote:
> Not everyone i guess since there is still the seg fault, report 1
> month ago.
>
> 2009/2/18 Ryan Devonshire
>
>> Did Valve ac
Still no fix for the seg faults on Linux. Chris Green, you got my emails?
2009/2/16 Arg! :
> Thats odd, i got notification via mailing list.
>
> 2009/2/17 Pawel
>
>> So mailing list is dead.
>> There is new update. And they only informed about it on main site :/
>>
>> On Fri, Feb 13, 2009 at 8:2
Is there any information as to when this will be fixed? Its obviously
still happening, and Valve, if you need a machine that produces these
results, feel free to email me.
2009/2/4 Thibaut PERRIN :
> Hi,
>
> I'm helping Flubber on his servers, and here we go :
> First, we realized that the server
;30"
> L 02/03/2009 - 13:33:36: server_cvar: "sv_tags" "cp"
> L 02/03/2009 - 13:33:36: server_cvar: "sv_tags" "cp"
> Unknown command "mp_limitteams"
> Unknown command "mp_autokick"
> L 02/03/2009 - 13:33:36: server_cv
So this doesn't fix the seg fault issue?
On Mon, Feb 2, 2009 at 5:39 PM, Zuko wrote:
> ./srcds_run: line 352: 19086 Segmentation fault $HL_CMD
> ...
>
> 2009/2/3 Jason Ruymen
>
>> A required update for Team Fortress 2 has been released. Please run
>> hldsupdatetool to receive the update.
I played around with it some more, I was "able" to get the servers to
start on CentOS. If you just let it error (seg fault), it will
automatically restart, mine was able to startup correctly after like
the 3rd auto-retry.
On Wed, Jan 28, 2009 at 8:31 PM, The Universes wrote:
> I
I just tested a clean install of Fedora 10, same error as on CentOS5.
I'm guessing the problem is related to Red Hat based distros?
On Wed, Jan 28, 2009 at 8:22 PM, Anthal wrote:
> I've got 3 servers up and running on linux. 2 rented (was a pain
> updating them), and one I run myself.
> The one I
Same here, CentOS5 = seg fault
Just reloaded with Fedora 10, will see if that works.
On Wed, Jan 28, 2009 at 7:58 PM, Ben B wrote:
> I am on centos5, i notice some other linux servers up and running too.
>
> On Wed, Jan 28, 2009 at 8:45 PM, Steven Sumichrast wrote:
>
>> I just did the update to m
25 matches
Mail list logo
