Overview
This check monitors CRI-O.
Setup
Installation
The integration relies on the --enable-metrics option of CRI-O that is disabled by default, when enabled metrics are exposed at 127.0.0.1:9090/metrics.
Configuration
Edit the crio.d/conf.yaml file, in the conf.d/ folder at the root of your Agent’s configuration directory to start collecting your CRI-O performance data. See the sample crio.d/conf.yaml for all available configuration options.
Restart the Agent.
Validation
Run the Agent’s status subcommand and look for crio under the Checks section.
Data Collected
CRI-O collects metrics about the count and latency of operations that are done by the runtime.
The Datadog-CRI-O integration collects CPU and memory usage of the CRI-O golang binary itself.
Metrics
crio.cpu.time
(gauge) | Total user and system CPU time spent in seconds
Shown as second |
crio.mem.resident
(gauge) | Resident memory size in bytes
Shown as byte |
crio.mem.virtual
(gauge) | Virtual memory size in bytes
Shown as byte |
crio.operations.count
(count) | Counter of CRI-O operations |
crio.operations.latency.count
(gauge) | Count of CRI-O operations latency |
crio.operations.latency.quantile
(gauge) | Quantiles of CRI-O operations latency |
crio.operations.latency.sum
(gauge) | Sum of CRI-O operations latency
Shown as microsecond |
Service Checks
crio.prometheus.health
Returns CRITICAL if the check can’t access the metrics endpoint.
Statuses: ok, critical
Troubleshooting
Need help? Contact Datadog support.
