PAM Server running out of disk space rapidly
Article ID: 133229
Updated On:
Products
Issue/Introduction
PAM Server is running out of disk space in a matter of days and results in unexpected behavior and affects stability of the service.
Behaviours:
1. Users fail to logon to PAM (Bad User ID or Password)
2. PAM reports it is in "Starting up" state (able to access "https://<pamip>/config/?legacy=1")
3. PAM Cluster state shows either Credential Manager or Session Manger being unavailable
4. Logon to PAM is taking extremely long time
5. Accessing Target device fails reporting Session Recording Mount not being available (misleading message and status)
6. Users who logged on to PAM gets "Unauthorized" error (User session abruptly terminated)
7. etc
Sometimes restarting cluster (and reboot of machines) helps but sometimes it does not.
Environment
Release: PAM 3.x.x (prior to PAM 3.2.5 or PAM 3.3.0)
Component:
Cause
There are 2 log files that are known to be growing fast.
1. tomcat catalina.out file
2. xcd_spfd.log
These logs can take up all available disk space on PAM interrupting operation at the OS level.
Resolution
These issues are resolved on PAM 3.2.5 and 3.3.0
Upgrade is required.
Workaround is to set the Tomcat Loglevel to "SEVERE"
Feedback
