This insight appears when communication delays between your cluster nodes exceed 10ms. This metric is crucial for maintaining cluster synchronization, cache coherency, and collaborative features.

Read more in interpreting infrastructure metrics in-product diagonstics

How does high internode latency affect performance?

When node communication slows down, users experience:

• Content updates take longer to appear across nodes

• Users might see outdated information due to cache issues

• Collaborative editing becomes sluggish

• Cluster stability suffers during high-traffic periods

What's the recommendation?

Start with these key actions to improve node communication:

Verify network connectivity

Network issues are the most common cause of latency:

• Basic testing:

  ping -c 10 <other_node_ip> netcat -zv <other_node_ip> <port>

   

This confirms if nodes can communicate through the required ports. Find your node IPs and ports in Administration > Clustering.

Important considerations:

• Ensure all nodes are in the same geographic region (especially important for AWS deployments)

• Regular tests may not catch intermittent issues - set up continuous monitoring

• If using AWS, verify that nodes are in the same region to prevent cross-region latency

• Work with your network team if you see consistent latency or connectivity issues

Learn how to set up a Confluence Data Center cluster

Monitor system resources

Resource constraints can manifest as increased latency:

Resource tracking

1. CPU monitoring:

   • Track utilization patterns across nodes, review thread states during high-load periods, and monitor process priorities to ensure Confluence processes get adequate CPU time.

2. Memory management:

   • Verify heap settings, monitor garbage collection patterns, and track available memory across all nodes to keep your instance healthy. Also, watch for memory pressure indicators that might affect node communication.

Performance analysis

1. System metrics:

   • Monitor all cluster nodes

   • Track resource trends

   • Document baseline performance

   • Set up alerting thresholds

2. Regular review:

   • Analyze performance patterns

   • Identify bottlenecks

   • Update monitoring thresholds

   • Schedule maintenance windows

Explore system monitoring

Review cluster configuration

Proper cluster configuration is essential for minimal latency:

Hazelcast settings

1. Review configuration:

   • Go to Administration > Clustering

   • Check timeout configurations

   • Verify heartbeat intervals

   • Monitor cluster stability

2. Configuration best practices:

   • Optimize join timeouts

   • Configure proper heartbeat intervals

   • Set appropriate backup counts

   • Review operation timeouts

Node management

Place nodes with geographic proximity in mind and maintain consistent network paths between them. Create a regular schedule for:

• Node health monitoring

• Performance tracking

• Maintenance windows

• Configuration updates

Preventive measures

Implement these ongoing practices to maintain optimal internode communication:

• Set up alerts for latency spikes above 10ms

• Conduct weekly network performance tests

• Document your baseline performance metrics

• Schedule monthly infrastructure reviews

Contact Atlassian Support

1. Gather a Support Zip from all cluster nodes.

2. Include results from ping and netcat tests.

3. Contact Atlassian Support for further assistance.