[Satellite6] Tomcat/candlepin consumes much CPU and memory due to misconfigured virt-who

Environment

• Satellite6
• virt-who

Issue

• tomcat process on Satellite6 consumes much CPU and memory resources
• optionally, candlepin logs errors on sending messages (see Diagnostic steps here for details)
• some guests running under a hypervisor report their subscriptions to Satellite via virt-who

Resolution

Ensure every virt-who instance has appropriately high VIRTWHO_INTERVAL. It is recommended to stay with the default value of 3600 seconds. In case of decreasing the interval, it is not worth using values lower than 1800 seconds until explicitly recommended.

Ensure there is no rogue instance of virt-who running within your network, reporting subscriptions updates to Satellite frequently.

The interval can be configured in these 2 locations depending on the version of virt-who you are using:

1. VIRTWHO_INTERVAL in /etc/sysconfig/virt-who
2. interval in /etc/virt-who.conf

Root Cause

Some mis-configured virt-who instance sends subscriptions status report too frequently to Satellite6, causing overload of candlepin or congestion of gutterball - both running under tomcat.

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.