[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [cobalt-users] Re: memory problems
- Subject: Re: [cobalt-users] Re: memory problems
- From: Leon Kyneur <lkyneur@xxxxxxxxxxxxxxx>
- Date: Mon Apr 19 16:29:01 2004
- List-id: Mailing list for users to share thoughts on Sun Cobalt products. <cobalt-users.list.cobalt.com>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Mon, 19 Apr 2004, Bruce Timberlake wrote:
> Are you getting any CPU overutilization emails from Active Monitor (e.g.,
> maybe you've got a job that's sucking up all the resources)? Do you have any
> customers on the server that might have poorly written CGIs or database
> searches that could be thrashing away for a long time?
All the sites on this box are static html sites.. very basic pages..
Active monitor only sends me the occational email that a user has gone
over their quota.. a couple of times it's sent me an email to say 'disk
useage is normal' but didn't state there was an error beforehand.
> That's normal AFAIK, although I can't put my finger on an explanation at the
> moment.
>
> What is causing the "serious load"? Could this be related to your lack of
> login capabilities?
A couple of times it seemed swatch was the culprit, all that showed in
messages and security was something about another swatch proccess is
already running .... lockfile in '/var/lock' - I just removed the lockfile
and all came good. Other than that we haven't yet determined what's doing
this, the company I work for has had it for about 9 months and it's never
been _good_. It's been rebuilt twice in the 5 months i've been here.
> No, that's normal Unix caching... It keeps data around in memory for quite a
> while, on the off chance something might need it again. It's not doing any
> harm; having RAM sitting around unused is "wasteful" of resources. If
> something does need RAM, older cached info will be dumped as needed to make
> room for it.
I thought as much.. what about it not, or very rarely using it's buffers?
Is there any way to better benchmark this thing?
Thanks,
Leon
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)
iD8DBQFAhF+/nNBGYRS6RLQRAqdJAJ9/Jd+G/2vKzlVJm8tWG0sNuaQLiQCgn1oV
WHQtlgmnh560JlqtuZT+p1A=
=3uLk
-----END PGP SIGNATURE-----