Intermittent DatabaseReplicationMonitor Connectivity Failure

Products

• CA API Gateway

Issue/Introduction

Intermittent warnings from DatabaseReplicationMonitor on our CA API Gateway cluster when checking replication status for the secondary database.

2025-12-13T20:01:00.752+0530 WARNING com.l7tech.server.cluster.DatabaseReplicationMonitor:
Error accessing host/database xx.xx.xx.xx/ssg (attempt to get secondary failover replication status: sequence):
Failed to obtain JDBC Connection; nested exception is com.mysql.cj.jdbc.exceptions.CommunicationsException: Communications link failure
2025-12-13T20:01:08.751+0530 INFO Replication recovered for host/database xx.xx.xx.xx/ssg (secondary failover)

Observations:

• The alert lasted ~8 seconds and then recovered automatically.
• Replication remained healthy before and after the event.
• No failover or data inconsistency occurred.
• This pattern has occurred multiple times this year (approx. 2000 alerts), typically self-clearing.

 

 

Environment

API Gateway 11.1

Resolution

Work-around to address the issue.

1. Add or increase db.replicationDelayCheckInterval (a hidden cwp) from default 10s.

2. Increase the connection timeout 

• • Add the following line to Gateway\node\default\etc\conf\hibernate.properties (create the file if it's not already there). Default is 2000ms (2s), maybe try doubling to 4 seconds.

  • l7.mysql.url.parameter.connectTimeout = 4000

-------------------------------------

Refined Replication Monitoring feature is scheduled for a future release targeting Gateway 11.1.4 or later (ETA mid-2026; subject to change).