Troubleshooting Backup Monitoring KM Issues

Unable to find xxx_load.kml

The KM files have not been properly installed.

Check the Load KM browser is looking for *.kml files under PATROL_HOME/lib/knowledge folder.

XXX_SETUP icon does not appear after loading the KM

The managed system is not supported by the KM or is not properly configured.

The KM version installed on the PATROL Console and the PATROL Agent is different.

The PATROL Agent tuning variable, /AgentSetup/AgentTuning/pslInstructionMax has not been increased as suggested in the KM documentation.

The PATROL Agent user does not have the privileges required to read and write to the Agent Configuration Database.

Look for errors displayed in the SOW to better identify the source of the issue.

Make sure your configuration complies with the requirements listed in the KM documentation.

Check in the Agent’s Access Control List (/AgentSetup/accessControlList) whether the PATROL Agent user has the privileges required to read and write to the Agent Configuration Database.

Parameter XXXLoginStatus is in suspicious (warning) status (F)

A command executed by the KM failed.

The Operating System User does not have the privileges required to run commands and access the files of the managed system.

Look for failing commands in the information provided by the last annotation point of the XXXLoginStatus parameter or check the KM Status Report (Right-click the KM Main Icon > KM Commands > KM Status).

Right-click the KM main icon > KM Commands > Configuration > Login to modify the OS Login account.

KM Objects disappeared from the Console

The KM version installed on the PATROL Console and the PATROL Agent is different.

The password has been changed on the system.

The KM has been uninstalled.

The KM instance limits have been exceeded.

The main instance is in OFFLINE state.

Install the same KM version on both PATROL Console and PATROL Agent.

Right-click the KM main icon > KM Commands > Configuration > Instance Limits to modify the KM instance limits for the affected objects.

(F) If you’re using KMs in “Multi-node Mode”, check whether the KM has failed over to another managed node. In Multi-node Mode, when the KM fails over, the shared instances created on the previous active node will be destroyed by default. This behavior can be changed by setting the pconfig variable, /XXX/DestroyInstances, to false:
%PSL pconfig("REPLACE","/XXX/DestroyInstances", false);

KM Behavior Unchanged after Upgrade

The installation is not complete.

Check which version is actually installed (Right-click the KM main icon > KM Commands > InfoBox and look for the Product Version).

Make sure the PATROL Console and the PATROL Agent are uninstalled and installed correctly during the KM upgrade.

KM takes a long time to refresh all the parameters and application classes on the PATROL Console

Data collectors run according to their scheduling intervals (polling cycle) defined in the KM. These intervals are defined for a standard environment with minimal resource impact.

Modify the scheduling intervals from the PATROL Developer Console to suit your environment. Refer to the PATROL Console User Guide for more details.

Infinite Loop Errors

The tuning variable /AgentSetup/AgentTuning/pslInstructionMax is not properly configured.

Increase the tuning variable /AgentSetup/AgentTuning/pslInstructionMax to an optimum value. The default value of 500,000 should be increased to at least 5,000,000 on a standard environment.

Disable this functionality by setting the value of the tuning variable to zero (0).

Some Application Instances do not appear

The KM instance limits have been exceeded.

(F) If the KM is running in “Multi-node Mode”, the passive node monitors the components which are not shared. The KM only operates in full monitoring mode from the node where the KM is in active mode.

The main instance of the KM is in offline state. It will remain in offline state until the initial discovery is complete.

Right-click the KM main icon > KM Commands > Configuration > Instance Limits to increase the KM instance limits for the affected objects. You can refer to the SOW or to the KM Status Report to obtain information for the recommended limit.

Change the monitoring mode to single-node by right-clicking the KM main icon > KM Commands > Configuration > Monitoring Mode. Single-node mode allows the KM to perform full monitoring on the selected system

Performance Issues

The KM is installed on a heavily used system.

Too many components are being monitored.

(F) Right-click the KM main icon > KM Commands > Configuration > [Components] to disable the monitoring of unnecessary application classes

Increase the scheduling interval (polling cycle) for data collectors

Deactivate unnecessary data collectors and associated application classes

(F) Deactivate unnecessary data collectors during selected time intervals where there is no activity. You can for example use the following PSL command to disable the monitoring between 9am and 4 pm every day except weekends:

%PSL pconfig("REPLACE","/<KMCode>/<collector name>",
"0|32400|57600|0|0|||||||||||0|0"); Where

0 disables data collection

32400 corresponds to the default start/ends times in seconds since midnight (i.e. 9am)

57600 corresponds to the default start /end times in seconds since midnight (i.e. 4pm)

Purge unnecessary information in index databases, catalog databases or log files.

Set the instance limits of the relevant component to 0 to disable their monitoring (Right-click the KM main icon > KM Commands > Configuration > Instance Limits)

Refer to “Infinite Loop Errors” for a possible PATROL internal scheduling delay which may impact the performance of the KM.

(F) If you’re using Backup KMs, reduce the number of days to monitor backup jobs.

Old Active Save Groups or Old Active Jobs are not removed from the Console (F)

By default, all active save groups and all active jobs are monitored, and they are exempted from ageing.

If you are using the Backup KMs, right-click the Save Groups/Jobs icon > KM Configuration > Jobs and check the Include Active Jobs box.

Old acknowledged save groups are still kept in the pconfig database (F)

By default the KM stores all acknowledged save groups.

Run the following PSL command to keep only the last <Number> save groups/jobs:

%PSL pconfig("REPLACE",
"/XXX/XXX_JOB/JobAcknowledgementCapacity",<Number>);

First try to activate the Debug Mode by right-clicking the KM main icon > KM Commands > Configuration > Debug.

Then, turn the debug on by setting the PATROL Agent Configuration variable with a timestamp value. This timestamp value determines when the debug should be turned off. For example, to turn on the debug for 60 minutes from now, run the following PSL through PATROL Console:

%PSL pconfig("REPLACE","/<KM>/<Component>Debug",
time()+3600); Where
<KM> corresponds to the KM code (NSR, NBU, VCS, VVM, etc.)
<Component> is either the server or the collector name (e.g.: Server, DaemonCollector, JobCollector)

This problem can be resolved by changing the group details to read “<Any>” in the HP Data Protector GUI.