How large are these users' mailboxes? I vaguely remember someone reporting
high cpu usage and it was because the account they were checking had so many
emails that it would slow down scanning the directory to find the number of
messages.
__________________
Chris Bolt
[EMAIL PROTECTED]
http://www.bolt.cx
> -----Original Message-----
> From: Tamer Hassan [mailto:[EMAIL PROTECTED]]
> Sent: February 21, 2001 7:31 AM
> To: [EMAIL PROTECTED]
> Subject: Re: Very high CPU Usage!!
>
>
> B>Check the list archive. There was a thread on this a few weeks
> ago. The
> B>apparent cause was compiler shipped with RH7.0, I believe.
>
> >-Bill
>
>
> Actually, I am not running RH7.0. I am running debian 2.2
> r2(glibc21+egcs).
> Now I had to kill the processes manually. I have only seen this problem
> once. It has been four days since I killed the processes and the
> problem has
> not reoccured. I am not running tcpserver with the -p option
> (reverse dns)
> and the dns's in /etc/resolv.conf are ok (local djb dnscaches). The
> mysterious processes where a total of three, each consuming
> 33.3%. I am only
> running 10 domains on that box, and using cdb. Some hints, at the
> moment I
> ran top and saw these crazy processes, they were running under command
> "domainname" which was one of the 10 domains hosted on that box. This is
> also the only domain that has roaming users popping in. All other
> 9 domains'
> users access their mail through sqwebmail. One of my local dnscache's was
> down a while before I noticed this because of a power surge, if
> that could
> be related (maybe the user popped in at the moment the dns was down). But
> again, I do not have the -p option to tcpserver in my rc.pop init
> script. I
> just hope to unveil this mysterious behavior before it burns me a cpu.
>
> TH
