Long Local Delivery Delays

[Jamin A. Brown]

Hello,

We have just completed migrating a sendmail installation for roughly 15000
users to qmail. After doing so, we are experiencing longish delays on the
delivery of incoming messages.

We currently have qmail-smtpd starting from tcpserver using:

#!/bin/sh
QMAILDUID=`id -u qmaild`
NOFILESGID=`id -g qmaild`
exec /usr/local/bin/softlimit -m 200 \
 /usr/local/bin/tcpserver -c 512 -v -pR -x /usr/local/etc/tcp.smtp.cdb \
 -u $QMAILDUID -g $NOFILESGID 0 smtp \
 /usr/local/bin/rblsmtpd -b -r \
 "relays.mail-abuse.org:Open relay problem - see
 " \
 -r dialups.mail-abuse.org \
 -r rbl.maps.vix.com \
 /var/qmail/bin/qmail-smtpd 2>&1

And logging using:

#!/bin/sh
exec /usr/local/bin/setuidgid qmaill /usr/local/bin/multilog t s250
/var/log/qmail/qmail-smtpd


I have increased our conncurrencylocal and concurrencyremote to 50 each.
The qmail-send log we have seems to indicate that we aren't coming close
to this limit:

@40003a105ba216072edc status: local 1/50 remote 3/50
@40003a105ba21626288c end msg 695835
@40003a105ba2196aff04 new msg 695707
@40003a105ba21a78f464 info msg 695707: bytes 8004 from <> qp 18635 uid
59
@40003a105ba21dda6ee4 starting delivery 18024: msg 695707 to remote
[EMAIL PROTECTED]
@40003a105ba21ddd7454 status: local 1/50 remote 4/50


Our architecture is setup as:

mx0.gwi.net - Incoming server, BDS/OS, Dual PIII 550, 256 MB RAM. This
server writes incoming mail to a NetApp NFS server. Unfortunately also
running qmail-pop3d right now for some of our customers. Will be disabling
this ASAP.
pop0.gwi.net - POP server. (Same hardware). Just running qmail-pop3d and
reading from the NetApp.

The delay seems to occur from when mx0 accepts the message to when mx0
writes it to the user's Maildir. My guess would be that the queue is not
being processed fast enough.

We are NOT yet using the big-todo patch at this time.

We are seeing delays of up to a couple hours on messages, so any
assistance or pointers in the right direction would be appreciated.

If you would be so kind as to CC any responses to [EMAIL PROTECTED] (to
get to me directly rather than through the qmail server) it would be
appreciated.


-
Jamin A. Brown  Systems Operations Department
[EMAIL PROTECTED]   * Great Works Internet *   207.286.8686 x142
RSA PGP Key:  http://www.gwi.net/~jamin/pgp/jamin.asc

Re: Long Local Delivery Delays

[Ricardo Cerqueira]

Try disabling host resolution in tcpserver (by adding the -H option).
That's what usually stalls smtp deliveries with tcpserver.

RC


On Mon, Nov 13, 2000 at 04:39:11PM -0500, Jamin A. Brown wrote:
> Hello,
> 
> We have just completed migrating a sendmail installation for roughly 15000
> users to qmail. After doing so, we are experiencing longish delays on the
> delivery of incoming messages.
> 
> We currently have qmail-smtpd starting from tcpserver using:
> 
> #!/bin/sh
> QMAILDUID=`id -u qmaild`
> NOFILESGID=`id -g qmaild`
> exec /usr/local/bin/softlimit -m 200 \
>  /usr/local/bin/tcpserver -c 512 -v -pR -x /usr/local/etc/tcp.smtp.cdb \
>  -u $QMAILDUID -g $NOFILESGID 0 smtp \
>  /usr/local/bin/rblsmtpd -b -r \
>  "relays.mail-abuse.org:Open relay problem - see
>  " \
>  -r dialups.mail-abuse.org \
>  -r rbl.maps.vix.com \
>  /var/qmail/bin/qmail-smtpd 2>&1
> 
> And logging using:
> 
> #!/bin/sh
> exec /usr/local/bin/setuidgid qmaill /usr/local/bin/multilog t s250
> /var/log/qmail/qmail-smtpd
> 
> 
> I have increased our conncurrencylocal and concurrencyremote to 50 each.
> The qmail-send log we have seems to indicate that we aren't coming close
> to this limit:
> 
> @40003a105ba216072edc status: local 1/50 remote 3/50
> @40003a105ba21626288c end msg 695835
> @40003a105ba2196aff04 new msg 695707
> @40003a105ba21a78f464 info msg 695707: bytes 8004 from <> qp 18635 uid
> 59
> @40003a105ba21dda6ee4 starting delivery 18024: msg 695707 to remote
> [EMAIL PROTECTED]
> @40003a105ba21ddd7454 status: local 1/50 remote 4/50
> 
> 
> Our architecture is setup as:
> 
> mx0.gwi.net - Incoming server, BDS/OS, Dual PIII 550, 256 MB RAM. This
> server writes incoming mail to a NetApp NFS server. Unfortunately also
> running qmail-pop3d right now for some of our customers. Will be disabling
> this ASAP.
> pop0.gwi.net - POP server. (Same hardware). Just running qmail-pop3d and
> reading from the NetApp.
> 
> The delay seems to occur from when mx0 accepts the message to when mx0
> writes it to the user's Maildir. My guess would be that the queue is not
> being processed fast enough.
> 
> We are NOT yet using the big-todo patch at this time.
> 
> We are seeing delays of up to a couple hours on messages, so any
> assistance or pointers in the right direction would be appreciated.
> 
> If you would be so kind as to CC any responses to [EMAIL PROTECTED] (to
> get to me directly rather than through the qmail server) it would be
> appreciated.
> 
> 
> -
> Jamin A. Brown  Systems Operations Department
> [EMAIL PROTECTED]   * Great Works Internet *   207.286.8686 x142
> RSA PGP Key:  http://www.gwi.net/~jamin/pgp/jamin.asc
> 

-- 
+---
| Ricardo Cerqueira  
| PGP Key fingerprint  -  B7 05 13 CE 48 0A BF 1E  87 21 83 DB 28 DE 03 42 
| Novis  -  Engenharia ISP / Rede Técnica 
| Pç. Duque Saldanha, 1, 7º E / 1050-094 Lisboa / Portugal
| Tel: +351 2 1010  - Fax: +351 2 1010 4459

 PGP signature

Re: Long Local Delivery Delays

[markd]

On Mon, Nov 13, 2000 at 04:39:11PM -0500, Jamin A. Brown wrote:
> Hello,
> 
> We have just completed migrating a sendmail installation for roughly 15000
> users to qmail. After doing so, we are experiencing longish delays on the
> delivery of incoming messages.

What sort of passwd technology are you using? /etc/passwd? NIS, NIS+?

> The delay seems to occur from when mx0 accepts the message to when mx0
> writes it to the user's Maildir. My guess would be that the queue is not
> being processed fast enough.

If you showed us some log entries from start of delivery to completion
we'd be able to tell you whether they seem slow or not.

> We are seeing delays of up to a couple hours on messages, so any
> assistance or pointers in the right direction would be appreciated.

Show us the specific log entries for some of these deliveries. We can only
speculate in the absence of information.


Regards.

RE: Long Local Delivery Delays

[Greg Owen]

> The delay seems to occur from when mx0 accepts the message to when mx0
> writes it to the user's Maildir. My guess would be that the 
> queue is not being processed fast enough.

Have you checked the trigger?

http://web.infoave.net/~dsill/lwq.html#trigger

Sounds like a classic case of a bad trigger.

-- 
gowen -- Greg Owen -- [EMAIL PROTECTED]
 

Re: Long Local Delivery Delays

[Jamin A. Brown]

On Mon, 13 Nov 2000 [EMAIL PROTECTED] wrote:

> On Mon, Nov 13, 2000 at 04:39:11PM -0500, Jamin A. Brown wrote:
> > Hello,
> >
> > We have just completed migrating a sendmail installation for roughly 15000
> > users to qmail. After doing so, we are experiencing longish delays on the
> > delivery of incoming messages.
>
> What sort of passwd technology are you using? /etc/passwd? NIS, NIS+?

/etc/passwd at this point.

> > The delay seems to occur from when mx0 accepts the message to when mx0
> > writes it to the user's Maildir. My guess would be that the queue is not
> > being processed fast enough.

It seems that the problem may have been related to the number of open
files that BSD/OS allows by default. After increasing this number by quite
a bit, qmail-send is finally showing us approach our concurrency limits.

If things continue to seem slow, I will email the list with logfile
entries.

Jamin


-
Jamin A. Brown  Systems Operations Department
[EMAIL PROTECTED]   * Great Works Internet *   207.286.8686 x142
RSA PGP Key:  http://www.gwi.net/~jamin/pgp/jamin.asc