Well, "slapd -d" and/or gdb should show you where it's hanging, which would be Very Useful Information.
We've found that CentOS 5 really, really, REALLY likes:
The bind_policy soft option forbids nss_ldap from retrying failed LDAP queries. If the default bind policy is used, LDAP will retry a query several times when the LDAP server is not present. This can cause a pause of several seconds during routine operations.
Maybe you can try that. I can document the "pause of several seconds" extending for hours. No, this should not be the case. No, this isn't the case on my Solaris boxes.
On Tue, 1 Apr 2008, Rick Stevens wrote:
Hi. I've got a weird one for you.
System: CentOS 5, kernel 2.6.18-53.1.4 64-bit OpenLDAP: openldap-2.3.27 from RPM (openldap-servers-2.3.27-8.el5_1.3)
I'm using OpenLDAP for authentication via pam_ldap and this is going to be the secondary server (if I can ever get replication to work).
However, that's not the issue right now. This machine won't start LDAP unless I remove the "ldap" part from the "group" line of /etc/nsswitch.conf.
Won't start: group: files ldap
Will start: group: files
There are no log messages indicating a problem. slapd simply won't start. Has anyone ever heard of such a screwup and if so, any ideas as to what the fark is going on?
- Rick Stevens, Unix Geek rps2@socal.rr.com -
-If this is the first day of the rest of my life... -I'm in BIG trouble! -