mailscanner hangs on automatic restart {Scanned}

[Glenn Steen]

On 03/10/06, Henry Hollenberg <hgh at rcwm.com> wrote:
> Julian Field wrote:
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> >
> >
> >
> > Henry Hollenberg wrote:
> >
> >>It looks like mailscanner is hanging every time it does it's automatic
> >>restart at 14400 sec.
> >>
> >>If I do a manual restart
> >>
> >>/etc/init.d/mailscanner restart
> >>
> >>the logs look like this:
> >>
> >>Oct  1 17:32:06 bastion MailScanner[30537]: MailScanner E-Mail Virus
> >>Scanner version 4.41.3 starting...
> >>Oct  1 17:32:06 bastion postfix/smtpd[30539]: connect from
> >>c-69-138-210-131.hsd1.md.comcast.net[69.138.210.131]
> >>Oct  1 17:32:06 bastion MailScanner[30537]: Read 120 hostnames from
> >>the phishing whitelist
> >>Oct  1 17:32:09 bastion postfix/smtpd[30539]: NOQUEUE: reject: RCPT
> >>from c-69-138-210-131.hsd1.md.comcast.net[69.138.210.131]: 554 Service
> >>unavailable; Client host [69.138.210.131] blocked using
> >>bl.spamcop.net; Blocked - see
> >>http://www.spamcop.net/bl.shtml?69.138.210.131;
> >>from=<drkatzswfq at wingdriver.com.au> to=<domainudbr at rcwm.com>
> >>proto=SMTP helo=<c-69-138-210-131.hsd1.md.comcast.net>
> >>Oct  1 17:32:09 bastion postfix/smtpd[30539]: lost connection after
> >>RCPT from c-69-138-210-131.hsd1.md.comcast.net[69.138.210.131]
> >>Oct  1 17:32:09 bastion postfix/smtpd[30539]: disconnect from
> >>c-69-138-210-131.hsd1.md.comcast.net[69.138.210.131]
> >>Oct  1 17:32:09 bastion MailScanner[30537]: Using locktype = flock
> >
> > What do you think is wrong there? You get a startup notice from it,
> > followed by the locktype notice, all looks fine. Remember MailScanner
> > doesn't have anything to do with your SMTP service.
> >
> > Jules
> >
> > - --
> > Julian Field
> > www.MailScanner.info
> > Buy the MailScanner book at www.MailScanner.info/store
>
> Are you suggesting I change the locktype?
>
> It did hang again last night, and bayesian db is working now as is
> pyzor and razor, so I don't think they are hanging things up.
>
No, changing the locktype shouldn't affect your situation, since you
use Postfix...
What might be happening would be if some stray non-queue file end up
in the hold queue. Check that that isn't happening.
Depending on what you find, you should be able to determine if that is
it, and if so... what is responsible for putting it there:-).
Might be razor still being a bit confused where the logfile should go
(fix is to make sure it knows where too put it by way of the
razor-agent.conf file setting... and making sure the postfix user can
write where you say it should go), or perhaps the tnef expander
placing a file wrong... (don't remember the fix for that... Search the
archives, it has cropped up before... Perhaps switch to the internal
one).

HtH

-- 
-- Glenn
email: glenn < dot > steen < at > gmail < dot > com
work: glenn < dot > steen < at > ap1 < dot > se