ldap_result() failed: Can't contact LDAP server (AGAIN)

[Charlie Wyse <charlie.wyse [at] gmail.com>]

I saw the thread around March of 11.  But from reading over it, it looked like the issue was never resolved.  (http://lists.arthurdejong.org/nss-pam-ldapd-users/2011/msg00049.html)

Since we have moved to RHEL6.  We are noticing these errors very often on all of our systems.  However as others have reported.  Everything appears to work fine, so it looks like it's just a little to verbose.  Is there a way to limit verbosity so we don't see this message, or any solutions that actually just fix this issue?

One of our developers said this is caused by ldap recieving the timeout, then displaying this message before reconnecting instead of just reconnecting and if that fails, displaying this message.  However I tried to look through the nss-pam source and I couldn't really see this error message in the code, making me believe it's another library that's making the call?

Anyhow, we have a ticket with Red Hat to help resolve this but I wanted to also speak directly to the list to see if there is more information, ie. It's based on the fact that we use SUN Netscape ldap vs. openldap.  Or if anyone has found a magic idle_timelimit setting, or version of openldap libraries that will make this work?  I tried the latest version of nss-pam-ldapd and I notcied it's still there.  It's hard for us to get signoff to push RHEL6 out when the messages are just filled with the same thing over and over.  It looks like we get 4 messages every time someone uses sudo after 5 minutes,  Then just a few messages every 10 minutes or so.  I've adjusted every timelimit option I could find in nslcd but it just doesn't go away.  Any help would be appreciated.

-- 
To unsubscribe send an email to
nss-pam-ldapd-users-unsubscribe@lists.arthurdejong.org or see
http://lists.arthurdejong.org/nss-pam-ldapd-users/