Observability
Platform metrics for your instances, volumes and network interfaces, read every 30 seconds and queryable from the GraphQL API.
- Five platform metrics
- Read every 30 seconds
- Queryable from the GraphQL API
oc_compute_instance_cpu_seconds_total
Features
See what your infrastructure is doing.
Platform metrics
CPU time, bytes received and sent per network interface, and bytes read and written per volume.
Rates and increases
Ask for a RATE or an INCREASE over a period. A stop and start never makes a data point negative.
Priced per use
Metric API requests, ingested samples and custom metrics have their own lines in the public price list.
Log aggregationComing soon
Centralized log collection and search.
AlertingComing soon
Threshold and event alerting on platform metrics.
Use cases
What it’s for.
Infrastructure monitoring
Watch CPU, network and volume throughput for every instance you run.
Application performance
Line up a slow request window against instance CPU and volume reads to tell a code problem from an infrastructure one.
Incident responseComing soon
Threshold alerts on platform metrics.
Platform metrics are live today, in open beta.
The Observability docs cover querying metrics and every metric name, with GraphQL API and CLI examples, readable without an account.