KBI 312176 Node Monitoring Is Skipped Because It Cannot Be Pinged
Version
Argent Omega 2.2A-2404-A and later
Date
Thursday, 13 June 2024
Summary
Performance-monitoring Relator uses the option ‘Do NOT Apply Rules In This Relator To An Inaccessible Node’ for better performance — reading Windows performance metrics is quite expensive
This is especially true when a machine is offline, because the Rule has to wait to timeout and the long timeout can cause tasks to be cancelled and other issues.
The Relator might use a large Monitoring Group to cover hundreds of Windows machines.
However, some machines cannot be pinged because of firewall settings.
(A cheap ping is executed before the actual expensive performance Rule.)
The failure of the ping cause the Relator to skip for these machines.
One workaround is to create a separate Relator NOT using ‘Do NOT Apply Rules In This Relator To An Inaccessible Node’. But it is a cumbersome and odious approach.
The CMDB-X has been enhanced to have property ‘Ping Blocked’, The default value is No.
When the property is set to Yes, the Relator does not execute the ping and assume the node is up.
Technical Background
N/A
Resolution
Upgrade to Argent Omega 2.2A-2404-A or later