Matt schrieb:
[...]
Cpu(s): 10.1% us, 3.3% sy, 0.0% ni, 25.4% id, 61.0% wa, 0.2% hi, 0.0% si
61% wa? That's a true indicator for IO-problems.
[...]
Running bind on server and from I have heard bind does not create a
significant system load.
If you run it as a caching nameserver, it should
Actually, so far you failed to state a problem.
Load is something like number of processes waiting to be processed and
I have seen servers with load-values higher thann 1000 which still
reacted faster than my laptop.
Load is not a problem. It might indicate a problem, but load itself is
@exim.org
Subject: [exim] Speeding Up Exim
I just got a call from someone wanting to be able to send 50k email
messages a second. Most of it in bursts of a few seconds long. It has to
do with some kind of gaming application. I'm just trying to picture in
my mind what kind of hardware could do
Martin.Hepworth wrote:
Matt
Quickest (but not cheapest!) way is to get Spamassassin off the email
email onto a dedicated box. SA is a real CPU hog.
before doing that, make sure you're only running SpamAssassin against
thhose messages you can't reject first by other means. SpamAssassin is
http://wiki.mailscanner.info/doku.php?id=maq:index#getting_the_best_out_of_spamassassin
contains some tuning tips...
Its a bit soon to tell but I think this helped a bunch.
If running the bayes database in a file (DB_File) consider adding
bayes_learn_to_journal 1 to spam.assassin.prefs.conf
Am Samstag, den 01.03.2008, 12:48 -0600 schrieb Matt:
[...]
Hardware is:
AMD64 Dualcore 3800+ just updated to 5600+
4Gbyte DDR2
SATA2 500GB drive
Normally load average is like 8 or less but at peak times I am seeing
it spike to like 100 area. Upgrading the CPU gained very little. I
am
I just got a call from someone wanting to be able to send 50k email
messages a second. Most of it in bursts of a few seconds long. It has to
do with some kind of gaming application. I'm just trying to picture in
my mind what kind of hardware could do that.
I have an Directadmin based
Hello, Matt!
Matt wrote:
I wander if the SATA feature of NCQ would speed the disk I/O up but
there is likely no easy way to upgrade CentOS 4.x to do that.
AFAICT CentOS 4.5 kernel was originally based on 2.6.9. A long time
has passed since then and the 2.6.21/2.6.22 kernels where the new
I wander if the SATA feature of NCQ would speed the disk I/O up but
there is likely no easy way to upgrade CentOS 4.x to do that.
AFAICT CentOS 4.5 kernel was originally based on 2.6.9. A long time
has passed since then and the 2.6.21/2.6.22 kernels where the new
libata based disk
Matt wrote:
I wander if the SATA feature of NCQ would speed the disk I/O up but
there is likely no easy way to upgrade CentOS 4.x to do that.
AFAICT CentOS 4.5 kernel was originally based on 2.6.9. A long time
...
Anyway. I'd give it a shot if you have some minutes of downtime.
Matt schrieb:
I wander if the SATA feature of NCQ would speed the disk I/O up but
there is likely no easy way to upgrade CentOS 4.x to do that.
AFAICT CentOS 4.5 kernel was originally based on 2.6.9. A long time
has passed since then and the 2.6.21/2.6.22 kernels where the new
libata based
11 matches
Mail list logo