CubeAPM

What it covers in an RCA

CubeAPM is the primary application-telemetry integration: use it to inventory services, find error traces, inspect a trace waterfall, correlate logs, measure error and latency signals, and map dependency propagation.

Credentials

The preferred environment wiring is:

CUBEAPM_SERVER=cube.example.com
CUBEAPM_EMAIL=user@example.com
CUBEAPM_PASSWORD=replace_me

CUBEAPM_HOST=https://cube.example.com is an optional full-URL alias; .envrc maps it to CUBEAPM_SERVER when that variable is not already set. Deployments that differ from the defaults may also set CUBEAPM_QUERY_PORT, CUBEAPM_INGEST_PORT, and CUBEAPM_ADMIN_PORT.

The CLI honors the repo-local XDG_CONFIG_HOME. Running cubeapm login stores the fallback profile at .config/<env>/cubeapm-cli/config.yaml.

Read-only contract

This workspace uses CubeAPM for telemetry reads. Keep the integration to the query commands below, start broad, narrow by service and time range, and use -o json when an agent will parse the output.

Key commands

Use metric label values, not traces services, as the canonical service inventory:

cubeapm metrics label-values service -o json
cubeapm traces search --service <svc> --status error --last 1h -o json
cubeapm traces get <trace-id> -o json
cubeapm logs query 'error' --service <svc> --last 1h -o json
cubeapm logs hits --query 'level:error' --last 6h --step 15m -o json
cubeapm metrics query 'rate(http_requests_total{status=~"5.."}[5m])' -o json
cubeapm metrics query 'histogram_quantile(0.99, sum by (le) (rate(http_duration_seconds_bucket[5m])))' -o json
cubeapm traces dependencies --last 1h -o json

Links

• CubeAPM documentation
• cubeapm-cli source