I just more or less want to post a "Me too" mail.
hlds uses more resources than before. I am running them on FreeBSD 4.x,
5.x and 6.x with different types of linux_bases. Most of them are now
using Gentoo 2005.x and 2006.x for the linuxulator.
/Bjorn
On Sun, 28 May 2006, Mike Noordermeer wrote:
Well, status here is that serverside FPS is quite ok without pingboost
for ticrate < 150, with -pingboost 2 for ticrate < 400 and unstable with
any other combo. But well, that's the HLDS.
Distro doesn't seem to matter, only Ubuntu shows somewhat unstable FPS
with any combo. Also Ubuntu 5.10/6.06
My testing RHEL5 box running 64bit multilib does exhibit this problem
with -pingboost. Note it does have both nptl and linuxthreads support
built into glibc. It also uses a 2.6.16 kernel. My gentoo boxes are
using glibc2.4 which only support nptl and are using 64bit multilib on a
2.6.16 kernel.
[EMAIL PROTECTED] wrote:
Exactly!
I also think acutally alot of people have that issue, but wont notice or
report it, as they for example
got enough CPU Resources Free, so their Server ran at 40% before, and is
now running at 70%, so they won't notice
Pingsspikes or lags.
Yeah well, currently
Exactly!
I also think acutally alot of people have that issue, but wont notice or
report it, as they for example
got enough CPU Resources Free, so their Server ran at 40% before, and is
now running at 70%, so they won't notice
Pingsspikes or lags.
cheers
SecurityFox
I personally think most
peop
Rikard,
http://cdimage.debian.org/debian-cd/3.1_r2/i386/iso-cd/debian-31r2-i386-netinst.iso
download install and enjoy i say or you have some crappy hardware.
Please keep any flaming on our hardware off-list. (Btw, the machines are
al Dual AMD Opteron, Tyan Thunder K8S(-D) Pro, 2+ GB ECC REG r
This is a multi-part message in MIME format.
--
[ Picked text/plain from multipart/alternative ]
http://cdimage.debian.org/debian-cd/3.1_r2/i386/iso-cd/debian-31r2-i386-netinst.iso
download install and enjoy i say or you have some crappy hardware.
--
//Rikard 'Zapy' Bremark
[EMAIL PROTECTED]
Wa
Btw, on one server we haven't been able to run any ping boosted server
for a while. We thought the problem was hardware bound, but I digged
some logs now and the complaining started at 13 March, exactly after the
HLDS update of 9 March.
Seems quite clear to me now there really are issues with the
Hi,
"I'll try to plan some maintenance to switch to a 32 bit OS on some
machines, we'll see if that helps. "
Please tell us the result!
Well, I migrated one box with some servers on it, but don't really see a
difference. Some customers are still whining and FPS on some servers is
still instab
Hi
"I'll try to plan some maintenance to switch to a 32 bit OS on some
machines, we'll see if that helps. "
Please tell us the result!
cheers
SecurityFox
Mike Noordermeer schrieb:
Hi,
First of all, thanks for the reply Alfred.
Well, the problems are quite vague here. I'm not seeing really
Hi,
First of all, thanks for the reply Alfred.
Well, the problems are quite vague here. I'm not seeing really high CPU
usage numbers, except for on some servers. I do see some increased
instability and a less stable serverside FPS.
Currently we're using a x86_64 kernels on all our boxes, mostly
--
[ Picked text/plain from multipart/alternative ]
We experienced the same cpu issues on p4 2.8Ghz boxes with 1-2Gb of ram
running Fedora Core 4.
I fixed the problems by taking them offline!
--
___
To unsubscribe, edit your list preferences, or view th
On May 23, 2006, at 5:53 PM, Andrew Forsberg wrote:
On Tue, 2006-05-23 at 18:41 -0400, Gary wrote:
You should really try and use a 32bit system instead of bleeding edge
64bit stuff. I don't see any high cpu usage everyone that everyone
says, but then again, I stick to using x86 stuff until I h
On Tue, 2006-05-23 at 18:41 -0400, Gary wrote:
> You should really try and use a 32bit system instead of bleeding edge
> 64bit stuff. I don't see any high cpu usage everyone that everyone
> says, but then again, I stick to using x86 stuff until I have the
> need for more than 4 gigs of memory.
Wel
Is there a big difference in performance, between using HLDS/SRDS in
32/64 mode?
What takes more CPU, 32 or 64 bit?
Gary wrote:
At 04:38 PM 5/23/2006, [EMAIL PROTECTED] wrote:
But we also had the problem when I installed Debian 3.1 64 bit, to test.
So it's no Operating System Problem aswell
At 04:38 PM 5/23/2006, [EMAIL PROTECTED] wrote:
But we also had the problem when I installed Debian 3.1 64 bit, to test.
So it's no Operating System Problem aswell.
Things I think we know till now:
Its not:
- a CPU problem - we got reports from both, Intel and AMD Systems
- a Operation System P
This is a multi-part message in MIME format.
--
[ Picked text/plain from multipart/alternative ]
so what is suse then? windows?
i dont know and i dont mean to find out, not when im writing this email,
its liunx / unix same " shit " diffrent names.
try install a 32 bit kernel instead and you will s
erver uses
~ 20-30% CPU. And ~24slot full CSS server uses ~60-70% CPU. A gerat start
for me. :).
- Original Message -
From: "Stuart Stegall" <[EMAIL PROTECTED]>
To:
Sent: Tuesday, May 23, 2006 11:59 PM
Subject: Re: [hlds_linux] *Alfred* Very high CPU usage since HLDS Engine
u
[hlds_linux] *Alfred* Very high CPU usage since HLDS Engine
upgrade 35
I have been monitoring the list. I have not been able to reproduce the
problem and it seems that only a subset of users are effected. I wonder
if it is related to multi-core machines (hence my suggestion a while ago
to lock th
How many people who have this problem, are running a regular x86 32bit
system and how many who have this are running 64bit-multilib? (I assume
no one is using a 32bit chroot inside a 64bit pure enviroment)
(Several people who have mentioned this problem have also mentioned
running 64bit multilib.
Hello Alfred
Thanks for your reply.
I don't think that it's a multi-core CPU Problem, as I got the same
problem on our single core Server (AMD Athlon 64 3700+).
Our both Servers, one Athlon 4200+ dualcore and one Athlon 64 3700+ are
running Suse 10.0 64 bit.
But we also had the problem when I ins
This is a multi-part message in MIME format.
--
[ Picked text/plain from multipart/alternative ]
like iwe said, this is not a problem with HLDS its something with just
that linux distor, running
Debian Sarge here with latest 686 kernel, 3 servers 14 slots on a dual
2.2 opteron 1 gb ram
and no lag s
I have been monitoring the list. I have not been able to reproduce the
problem and it seems that only a subset of users are effected. I wonder
if it is related to multi-core machines (hence my suggestion a while ago
to lock the process to a single CPU to see how that helps) but that is
just a guess
Alfred,
Could you please respond to this matter? Is there being worked on, is
the problem unknown, are we just stupid, just a little "I've read it and
xyz" would be nice.
--
Regards,
Mike Noordermeer
[EMAIL PROTECTED]
___
To unsubscribe, edit your li
24 matches
Mail list logo