[imp] IMAP server failing, IMP users booted out

[andrew morgan]

On Tue, 23 Oct 2001, Craig Richardson wrote:

> Hi Everyone,
>
> I suspect this is more of a problem with my IMAP server than IMP, which is
> awesome by the way :) but I was wondering if anyone could help me out.  I ask
> this question here because sometimes it is only my IMP users that have this
> problem.  ie - other IMAP clients (Outlook, Netscape etc) dont.
>
> The problem is every now and again (about once per day) I get the following
> message in /var/log/messages
>
> Oct 23 12:13:56 currawong inetd[417]: imap/tcp server failing (looping),
> service terminated
> Oct 23 12:13:56 currawong IMP[32392]: FAILED 137.111.xx.xxx to localhost:143 as
> jenny
> Oct 23 12:14:04 currawong IMP[30616]: FAILED 137.111.xx.xxx to localhost:143 as
> anna
> Oct 23 12:14:11 currawong IMP[32398]: FAILED 137.111.xx.xxx to localhost:143 as
> anna
> Oct 23 12:14:48 currawong IMP[1824]: FAILED 137.111.xx.xxx to localhost:143 as
> carl
>
> After an arbitary amount of time it seems imap is restarted, otherwise if I am
> around a quick kill -HUP inetd will fix the problem.
>
> I have the following config
> RH 6.1
> apache 1.3.14
> imap 4.6
> php 4.0.4
> mysql 3.22.25
> IMP 2.2.4
>
> any suggestions as to what to check next would be greatly appreciated.

Check the man pages for your inetd server (man inetd).  This is a feature
of inetd to prevent too many connections from overwhelming your server.
Unfortunately, the default value for the maximum number of connections per
minute is usually one appropriate for hardware ten years old.  Most
servers these days can handle MUCH more.  The default for Solaris 8's
inetd is 40 connections, I've bumped that up to 500 on my imap server for
IMP.

	Andy