On Mar 3, 2022, at 1:39 AM, Rahman Duran <rahman.duran@erzurum.edu.tr> wrote:
Long story short, the following configuration works in default virtual server but it does not in inner tunnel. In inner-tunnel, if the first ldap section could not find user, then it does not go on and check second ldap;
The "redundant" key word doesn't know (or care) if it's running in "default" versus "inner-tunnel" If you want us to figure out why it's behaving differently in the two cases, then it would help to post the debug output for both cases.
authorize {
We don't need to see the configuration files. All of the documentation makes this very clear.
(633) redundant { rlm_ldap (ldap_personel): Closing connection (10): Hit idle_timeout, was idle for 53546 seconds rlm_ldap (ldap_personel): Closing connection (12): Hit idle_timeout, was idle for 53529 seconds rlm_ldap (ldap_personel): Closing connection (11): Hit idle_timeout, was idle for 53522 seconds rlm_ldap (ldap_personel): You probably need to lower "min" rlm_ldap (ldap_personel): Closing connection (13): Hit idle_timeout, was idle for 53521 seconds
Why do you have so many connections idle for ~10 hours?
rlm_ldap (ldap_personel): You probably need to lower "min"
If the connections aren't needed, perhaps pay attention to this message, and set "min=0".
Any idea why "redundant" behaves different in the inner-tunnel or what am I missing?
It's behaving differently because the inputs to the LDAP module are different. But until you post the *full* debug output, it's impossible for us to say what those differences are. Alan DeKok.