PCS/PPS can be deployed as a cluster pair in Active/Passive and Active/Active Mode.
In Active/Passive mode, one PCS/PPS actively serves user requests, while the other PCS/PPS runs passively in the background to synchronize state data which includes system state, user profile, and log messages. Each node in a cluster is in fairly constant communication with one another, to ensure they are all active and there is data synchronization as well; so both nodes know each other's status and each node has the latest configuration. User requests to the cluster VIP (virtual IP address) are passed to the active PCS/PPS. If the active PCS/PPS goes off-line, the standby PCS/PPS automatically starts servicing user requests.
In Active/Active mode, all the machines in the cluster actively handle user requests sent by an external load balancer.
When there are issues with the cluster, use the following steps to enable and collect the required logs:
- Enable the Nodemonitoring option on both the PCS/PPS nodes.
This option is available in PCS/PPS admin GUI under Maintenance > Troubleshooting > Monitoring > Nodemonitoring
Log Settings (PCS):
Select the Nodemonitoring enabled checkbox and Save changes
Log Settings (PPS):
Select the Nodemonitoring enabled checkbox with LogSize=10 MB and Monitoring Interval= 10 Seconds
Turn on the Debug log with the settings as mentioned below
- Enable the Group Communication option in both the PCS/PPS nodes.
The option is available in PCS/PPS admin GUI under Maintenance > Troubleshooting > Monitoring > Clustering > Group Communication
Enable the Access Group Communication option and configure the log settings as:
Maximum statistics log size =5 MB
Monitoring Interval=10 Seconds
- Enable the Debuglog option on both the PCS/PPS nodes.
This option is available in PCS/PPS admin GUI under: Maintenance > Troubleshooting > Monitoring > Debuglog
Log Settings (PPS):
Select the Enable Debuglog checkbox with Log Level=20, Log Size=50 MB, Event Code(s):agentman,gateman
Log Settings (PCS):
Select the Enable Debuglog checkbox.
Issue: Cluster Formation or Transitioning issues, Cluster Hanging. Event Codes: -DSUtil,-DSLog,-DSConfig , Log Level: 15 , Log size: 50
Issue: Cluster VIP Failover Issues. Event Codes: dsnetd::ipat,dsnetd::garpsweep , Log Level: 15 , Log size: 50
- Start the TCP Dump from the Internal port of both the PCS/PPS by clicking the Start Sniffing option.
This option is available in PCS/PPS admin GUI under:
Maintenance > Troubleshooting > Tools > TCP Dump
- After replicating the issue, disable the TCP Dump on both of the nodes by clicking the Stop Sniffing option.
- Enable the Access System snapshot option
Available in PCS/PPS admin GUI under:
Access System snapshot option
Enable the following system snapshot options - Include:
system config Include debug log
- After enabling the System Snapshot options, click Take Snapshot. This will generate snapshot with the name: Admin generated snapshot.
Important: Take snapshots from both the PCS/PPS nodes of the Cluster by following the above procedure . Make sure you have logged on to the PCS/PPS using the Physical IP of the respective nodes to enable logging and capture logs.
- After the System Snapshot is taken, disable the Monitoring options enabled in Step 1 and 2. (This includes debug log monitoring and node monitoring).
- Collect the following:
- System Snapshot taken from both the nodes (step 6).
- TCP dump captured from both the nodes (step 4).
- Event/User Access/ Admin logs by using the save all logs options available In PCS/PPS admin GUI under System > Log/Monitoring.
- Screenshot of the clustering Status page. This option is available under System > Clustering.
- Screenshot of the System Status page. This option is available under System > Status.
- Time stamp of the Issue.
- After capturing data, upload the logs to the case via pulse secure web case manager.