Monitoring & Alerting
Watch every contract and service, and act before issues become incidents.
What it does
A platform holding other people’s assets cannot find out about a problem from its customers. This module is designed to watch the things that matter — contract state and events, failed transactions, stale oracles and valuations, queue depth, service health — and to raise an alert with enough context to act on, before a degradation becomes an outage. It is the difference between operating a platform and hoping it is fine. In development.
How it works
Monitoring & Alerting is the operations layer. It watches what Keeper & Automation runs, and what it observes ends up in the Audit Trail.
Select a step to replay it. Hover to pause.
Step 1 of 3: Watch. Contracts, services, oracles and queues are observed continuously.
Key capabilities
Contract and service health
On-chain and off-chain components are watched together, because an outage does not respect the boundary.
Stale data detection
A valuation or feed that has stopped updating is treated as a fault, not as current.
Actionable alerts
An alert carries the context to act, rather than only reporting that something is wrong.
Escalation paths
Someone is responsible for each class of alert, and the path is defined in advance.