> On Fri, 22 Dec 2000, Daniel Conlon wrote:
>
> > > > I have had qmail installed on a machine with < 100 users for 6
> > > months and all aspects of the installation are working correctly.
> > > However, it has recently been brought to my attention that the
> > > hard disk activity on the machine is excessive/abnormal - the
> > > disk is flat out.
> > > >
> > > > By systematically stopping processes I have established that it
> > > is qmail which is causing this. qmail is not under any load
> to speak of.
> > > >
> > > > Is this normal for qmail? If not, what could be causing this?
> > >
> > > Is it qmail-send or qmail-smtpd that is causing it?
> >
>
> Please reply to the list too. That way you'll get more eyes looking at
> your problem.
Sorry, thought I did.
> Also please wrap your lines to < 80 characters. It makes it easier to
> read.
I have this set to 76 characters in Outlook. Is it not working?
>
> > I think it is qmail-smtpd. I say this only because 'supervise
> qmail-smtpd' is being quite heavy on CPU usage (18%). How would I
> find out for sure?
>
> Looks like supervise is failing to start tcpserver correctly and
> looping. What does /service/qmail-smtpd/run (or wherever you put it)
> look like?
#!/bin/sh
QMAILDUID=`id -u qmaild`
NOFILESGID=`id -g qmaild`
exec /usr/local/bin/softlimit -m 2000000 \
/usr/local/bin/tcpserver -v -p -x /etc/tcp.smtp.cdb \
-u $QMAILUID -g $NOFILESGID 0 smtp /var/qmail/bin/qmail-smtpd 2>&1
Are you logging with multilog? If so what's the output of
> the current file in the directory as specified by
> /service/qmail-smtpd/log/run?
Yes, multilog. The output mainly consists:
@400000003a438b480b0ca6c4 tcpserver: fatal: no IP address for 1003
Looks like maybe I am passing it a GID where there should be something else? (1003 is
the GID of qmaild)
