Once upon a time, John Haxby <[email protected]> said: > On 5 April 2010 19:48, Chris Adams <[email protected]> wrote: > > I ran slapd with -d2 to gather debugging during the 3+ second delay, and > > I saw the following line repeated 37,370 times during the delay: > > > A delay of about three seconds is often down to a DNS problem. Check that > the servers listed in /etc/resolv.conf are working properly. Running > wireshark to see what's going on during the delay might also shed some light > if you're not seeing any DNS servers aren't right.
The delay is after the connection is open and TLS has been started, and it also happens on the ldapi (local socket) connection, so it isn't DNS (also, a DNS problem wouldn't come and go with just changing the openldap packages). There's a patch in the RHEL 5.5 openldap packages that changes TLS negotiation, and to me, just looking at it, it looks like it could cause a busy loop; I recompiled without that patch and the problem went away. I think the problem that the patch was supposed to fixed needs to be re-investigated (that's BZ 509230). Red Hat support has reproduced the problem and has passed it on to engineering. -- Chris Adams <[email protected]> Systems and Network Administrator - HiWAAY Internet Services I don't speak for anybody but myself - that's enough trouble. _______________________________________________ rhelv5-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/rhelv5-list
