We had this problem, and found that our tcp wrappers were sending ident (rfc 931) queries. The intermediate server from which the IMAP requests were coming dropped those packets rather than sending a TCP reset, which caused long delays while the ident query timed out. The same thing happens with some firewalls.
Thanks Mark (both Marks) for your replies. We're currently not running imapd with tcp wrappers wrapping it. A little more background on what we have - we have a Solaris 9 server and run imapd out of /etc/inetd.conf like so:
imaps stream tcp nowait root /usr/local/imap/libexec/imapd imapd
Is there any other way imapd or inetd can be issuing ident queries? thanks, --tom _______________________________________________ Imap-uw mailing list [email protected] https://mailman1.u.washington.edu/mailman/listinfo/imap-uw
