Alerts

Refer to the severity definition and SLA to determine the alert severity.

Alerts refer often to a specific cluster (mainnet/sepolia) and a SuperChain Networks (Base,Mode,…):op, mode, …. other chain networks) Alerts should already come with a specific log filtering and dashboard that should already be helpful for the investigation.

Alerts

Note

An alert contains the name of the network to which the alert refers. For example:

Fault Detector Stalled (OP MAINNET A1S2 PROD)

• OP refers to the optimism network. Other example values could be MODE, BASE, … (for a specific superchain network).
• A1S2 PROD: is the specific deployment type on which the alert was triggered. For the same network, we may have multiple deployments of the monitoring to have multiple sources of truth in terms of which node we use for our checks.

Escalation

For Superchain networks, escalation should be handled through the communication team (@Mattie Fairchild).

References

Other useful links for the operator to refer to.

1. two-step-withdraw-forgery-detection-error-unhandled:
   1. Error unhandled refer to an error fired at some point. Some of those errors are normal, or may happen under normal circumstances. In order to evaluate the issue, directly from the dashboard is possible to review the logs and the rate at which the error occurred. Example below show an error that should resolve by itself: “block is out of range”