> I suspect it is not 'max_uid_ccaches' as this is the total number of unique credentials caches per user, perhaps 'max_ccache_size' is the one being triggered.


I set 'debug_level' to 8 and did a bunch of logins/logouts. Every login creates a new cred cache, and after a while I no longer get a new credential on login. Error message:

(2020-11-24  6:54:17): [kcm] [kcm_op_set_default_done] (0x0040): Cannot set default ccache 1432158287: The maximum number of stored secrets has been reached


So it is indeed 'max_uid_ccaches' thats the culprit, if I increase it the error goes away and I once again get new credentials on login. But since they are never removed this is not a solution, only delaying the inevitable error. 

//Adam



From: externaly-forwarded@smhi.se <externaly-forwarded@smhi.se> on behalf of Winberg Adam <Adam.Winberg@smhi.se>
Sent: 24 November 2020 07:36
To: End-user discussions about the System Security Services Daemon
Subject: [SSSD-users] Re: kcm, gssproxy and klist
 

There are no error log messages in the kcm log file at all, only

[sssd[kcm]] [orderly_shutdown] (0x0010): SIGTERM: killing children


I have not set a '[kcm]' entry in my sssd.conf, whats the default loglevel? Should at least log errors I guess.


The bugzilla is for Fedora, I cloned it for RHEL8 and described my use case there:

https://bugzilla.redhat.com/show_bug.cgi?id=1900973


//Adam



From: Justin Stephenson <jstephen@redhat.com>
Sent: 23 November 2020 22:47
To: End-user discussions about the System Security Services Daemon
Subject: [SSSD-users] Re: kcm, gssproxy and klist
 
You can read more about this in the following BZ, but this should not prevent a user from acquiring new credentials.


If you are hitting a KCM quota, a failure message will be logged stating the affected quota in the sssd_kcm.log file. I suspect it is not 'max_uid_ccaches' as this is the total number of unique credentials caches per user, perhaps 'max_ccache_size' is the one being triggered.

Thanks,
-Justin


On Mon, Nov 23, 2020 at 6:46 AM Winberg Adam <Adam.Winberg@smhi.se> wrote:

With KCM and gssproxy we often see a long list of credentials when doing  a 'klist':



[user.u@lxserv2114 ~]$ klist
Ticket cache: KCM:17098:66803
Default principal: user.u@AD

Valid starting       Expires              Service principal
01/01/1970 00:00:00  01/01/1970 00:00:00  Encrypted/Credentials/v1@X-GSSPROXY:
01/01/1970 00:00:00  01/01/1970 00:00:00  Encrypted/Credentials/v1@X-GSSPROXY:
01/01/1970 00:00:00  01/01/1970 00:00:00  Encrypted/Credentials/v1@X-GSSPROXY:
01/01/1970 00:00:00  01/01/1970 00:00:00  Encrypted/Credentials/v1@X-GSSPROXY:
01/01/1970 00:00:00  01/01/1970 00:00:00  Encrypted/Credentials/v1@X-GSSPROXY:
01/01/1970 00:00:00  01/01/1970 00:00:00  Encrypted/Credentials/v1@X-GSSPROXY:
01/01/1970 00:00:00  01/01/1970 00:00:00  Encrypted/Credentials/v1@X-GSSPROXY:
and so on...

The actual gssproxy credentials at /var/lib/gssproxy/clients/ does not correspond with this output, it only contains what could be expected - a TGT and maybe some service tickets. 


The ever growing 'klist' list of credentials is a problem, after a while the user can no longer get any new credentials and therefore has no access to its NFS homedir (sec=krb5). I'm guessing it's the 'max_uid_ccaches' option in sssd-kcm that prevents this. 


What is going on here - have we configured gssproxy/kcm wrong or is this a bug?


Regards

Adam 



_______________________________________________
sssd-users mailing list -- sssd-users@lists.fedorahosted.org
To unsubscribe send an email to sssd-users-leave@lists.fedorahosted.org
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedorahosted.org/archives/list/sssd-users@lists.fedorahosted.org