Alert manager fails to trigger Alerts for any events after Leader Node Failover

Products

• VMware Avi Load Balancer

Issue/Introduction

• Following a leader node failover, the system generates critical events, specifically SE_UP (Service Engine Up) and CONTROLLER_NODE_JOINED. However, the alert_manager fails to generate and trigger the corresponding alerts for these events. 
• Any transient events during failover fail to generate related alerts. 

Environment

Any version below 22.1.7-2p7

Cause

• The cause of this issue is a timing mismatch between the alert generating component (alert_manager) and the log collection service (log_mgr) after a node failover. The log_mgr requires over five minutes to initialize, while the alert_manager only maintains a maximum 30-second search window for past events. By the time the log_mgr is fully operational, the alert_manager's search window has already advanced, causing it to miss the initial, critical failover events.
• This is a known issue on 22.1.7-2p7 or below.

Resolution

This issue is fixed in Avi version 22.1.7-2p8. Refer to AV-241070 in release notes  https://techdocs.broadcom.com/us/en/vmware-security-load-balancing/avi-load-balancer/avi-load-balancer/22-1/vmware-avi-load-balancer-release-notes/release-notes-22-1-7.html