I am posting this because I was unable to find a suitable solution on
google or the mailing list archives. I also sending it to the Debian
maintainer in case he wants to show a warning during the install.
We have recently begun deploying SM as the standard MUA. We were having
some problems which were being attributed to SM. Knowing that many other
people/organizations use SM, I was hesistant to post a bug report. So I
looked further into it.
Here is the setup:
Dell PowerEdge
Dual Intel P4 Xeon 2.4 GHz
2 GB RAM
280 GB SCSI HDD
Debian 3.0
inetd from netkit 0.10 (netkit-inetd_0.10-9)
uw-imapd 2001a (uw-imapd-ssl_2001adebian-6)
postfix 1.1.11 (postfix_1.1.11-0.woody2)
squirrelmail (squirrelmail_1.2.6-1)
~300 users
The symptom:
Seemingly at random times everyone who was logged on to SM would start
receiving the same errors in their Folders frame (most people have the
folders being automatically updated every 3-5 minutes). If they tried to
refresh their main frame, the errors would appear there. The errors
complained about invalid File-Handle:
Warning: Supplied argument is not a valid File-Handle resource in \
/usr/share/squirrelmail/functions/imap_general.php on line N
The lines that had complaints were 38, 62, 81, and 137 (most likely
version specific).
The search:
Searching the web (google) and the mailing list archives (sf), I was
unable to find a suitable solution for this problem. I looked through
/var/log/syslog, but postfixes verbosity made it difficult to find the
needle in the haystack that was causing the problem.
The solution:
Since I had the source code, I was able to see that all these lines were
reading (fputs) or writing (fgets) to the $imap_stream. This helped me to
figure that problably something was causing the imap connection to die,
which lead me away from squirrelmail and to imapd and inetd. Looking more
closely through /var/log/syslog, I saw that at the times the problems
occurred, there were a lot of imap connections. Knowing this was handled
by inetd, I grep'ed the logs for inetd and came up with lines like the
following:
inetd[200]: imap2/tcp server failing (looping), service terminated
at the exact times the problems were appearing. Googling on this I found
out about the maximum connection limit for a server per limit in
inetd.conf. The default is 40 for the version of inetd that I am using.
This can be increased in /etc/inetd.conf by appending a .MAX to the nowait
field for imap2:
imap2 stream tcp nowait.10000 root /usr/sbin/tcpd /usr/sbin/imapd
and restarting inetd (# /etc/init.d/inetd restart).
I think that should fix the problem and hopefully this email help anyone
else who might be having the same problem.
Finally, thanks to the SM developers for a great product.
dd
--
David Dooling
-------------------------------------------------------
This sf.net email is sponsored by: See the NEW Palm
Tungsten T handheld. Power & Color in a compact size!
http://ads.sourceforge.net/cgi-bin/redirect.pl?palm0001en
--
squirrelmail-users mailing list
List Address: [EMAIL PROTECTED]
List Archives: http://sourceforge.net/mailarchive/forum.php?forum_id=2995
List Info: https://lists.sourceforge.net/lists/listinfo/squirrelmail-users
