Monitoring and Alarms for Hub

For Tetra Hub, the following infrastructure-level notifications are sent to TetraScience by AWS automatically:

• Disk Space Alarm: Indicates when a Hub is using more than 90% of its available disk space.
• Heartbeat Metric: Validates the connection (heartbeat) between the TDP and a Hub every minute. If the TDP doesn’t receive a heartbeat message for more than 20 minutes or if the metric stops reporting, the platform assumes that the Hub is offline.
  • Offline Alarm: Indicates when a Hub is offline by using the Heartbeat Metric. For information about disaster recovery options, see Disaster Recovery Options for Hub.
• Proxy Heartbeat Metric: Validates the connection (heartbeat) between the TDP and each Hub’s proxies every minute. If the TDP doesn’t receive a proxy heartbeat message for more than 20 minutes or if the metric stops reporting, the platform assumes that the Hub’s proxy is offline.
  • Proxy Offline Alarm: Indicates when a Hub’s proxy is offline by using the Proxy Heartbeat Metric.

For more information, see Logs, Metrics, and Alerts.

📘

NOTE

These infrastructure-level alerts contain no sensitive information. They indicate which Hub failed and how only. This information helps the TetraScience team provide timely and effective support.

Disaster Recovery Options for Hub

Tetra Integrations that use a Tetra Hub to connect to the Tetra Data Platform (TDP) store their own states (locally for Agents and in the cloud for Connectors). Integrations automatically use these stored states to retry operations that fail during a temporary Hub outage.

If a Hub's host server is irretrievably lost, no Agent or Connector data is lost, because the data is always retained by the integration outside of the Hub. However, you must fail over the integration to a new Hub if this happens.

📘

NOTE

It's recommended that customers who prefer to have redundancy in their Hub setups provision a second Hub host server as a standby.

Failover Options

You can fail over any Tetra Agent or Pluggable Connector to another Hub or the cloud if your network requirements permit it.

To change an Agent’s Hub, select a new Hub from the Install Agent page in the TDP. Then, copy the Hub's Connection URL and enter it in the Connection Url field in the Agent's local management console. For more information, see the Setup procedure for Hub Connections.

To change a Pluggable Connector's Hub, see Change a Pluggable Connector's Host.