What do you have your openldap server's idletimeout set to in 
/etc/openldap/slapd.conf?

How about the setting on your client server for idle_timelimit in 
/etc/ldap.conf?

Kevin

-----Original Message-----
From: [email protected] [mailto:[email protected]] On 
Behalf Of Rich Graves
Sent: Thursday, September 08, 2011 9:21 AM
To: [email protected]
Subject: Re: [rhelv6-list] Intermittent Mozilla NSS/LDAP client failures in PHP

> I came across this once before where the LDAP server couldn't accept more
> connections from a particular host because there weren't any available
> ports.   Basically there were a lot (huge numbers) of sockets on the LDAP
> server in a CLOSE_WAIT or FIN_WAIT (or similar) state.

Nope, not it. I've been looking at netstat/lsof and actually forked off a 
dedicated LDAP server just to eliminate outside noise.

By the way: the LDAP servers and the web server/LDAP client are all RHEL6 Xen 
guests on top of RHEL5 DomUs.

I thought maybe we could be running low on entropy, but scanning 
/proc/sys/random shows that's not it either... and I'd expect NSS to use 
urandom anyway.

_______________________________________________
rhelv6-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/rhelv6-list

_______________________________________________
rhelv6-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/rhelv6-list

Reply via email to