[Exim] Some advice on system hangs

Top Page
Delete this message
Reply to this message
Author: Ron McKeating
Date:  
To: Exim-Users (E-mail)
Subject: [Exim] Some advice on system hangs
We have a very irregular problem of the mailserver suddenly going slow
for about 15-20 seconds. Here is an example from the log

Email comes in to mail server

2003-11-13 14:05:10 1AKI5L-0001AX-4Z <= T.Warren@???
H=(bill.lut.ac.uk) [158.125.1.193] P=esmtp S=1589
id=018201c3a9ef$0e1a5330$afd4e783@pc2000adll

Email gets delivered to local account (it is delivered to 6 local users
in all)

2003-11-13 14:05:43 1AKI5L-0001AX-4Z => admee
<admee@???> R=localuser T=local_delivery

Then many more log entries until we hit
2003-11-13 14:08:18 1AKI8E-0001Jf-KA Spool file is locked (another
process is handling this message)
2003-11-13 14:11:04 1AKI8E-0001Im-Dv Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001KB-TJ Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Ju-NB Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Js-Mp Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jo-Kj Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jm-Ki Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jc-K9 Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001JP-Dw Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jn-Ki Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jp-Kk Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jt-Mq Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jv-NW Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jx-NZ Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001KC-TK Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jr-Mp Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jg-KA Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001Jq-Kl Spool file is locked (another
process is handling this message)
2003-11-13 14:11:10 1AKI8E-0001JO-Dv Spool file is locked (another
process is handling this message)

And then it completes and all is well again

2003-11-13 14:11:51 1AKI5L-0001AX-4Z Completed

Notice there are not log entries for the one between 14:11:10 and the
completed one at 14:11:51 40 seconds with no activity. That is unheard
of on our system.

One significant fact is that one of the addressees has the largest inbox
on our server

8948 in /home/ad/admee/Maildir/./cur
9209 in /home/ad/admee/Maildir/.Sent Items/cur

Exim 4.24 and a 500G volume using reiserfs

Any thoughts as this problem has been plaguing us for months, and lots
of staff complaining about sudden freezes of email for about 20 seconds.

/Ron

--
Ron McKeating
Senior IT Services Specialist
Internet Services and Software Solutions
Loughborough University
01509 222329