Skip to content

Monitoring

VIBE monitoring, focusing on Claude Code observability.

VIBE monitoring mainly uses Claude Code OpenTelemetry data. Code-Server and JupyterLab do not expose Prometheus metrics. Use systemd and logs to check their health.

Claude Code Observability

VIBE writes default OpenTelemetry env vars into ~/.claude/settings.json:

{
  "env": {
    "CLAUDE_CODE_ENABLE_TELEMETRY": "1",
    "OTEL_METRICS_EXPORTER": "otlp",
    "OTEL_LOGS_EXPORTER": "otlp",
    "OTEL_EXPORTER_OTLP_METRICS_PROTOCOL": "http/protobuf",
    "OTEL_EXPORTER_OTLP_LOGS_PROTOCOL": "http/protobuf",
    "OTEL_EXPORTER_OTLP_METRICS_ENDPOINT": "http://127.0.0.1:8428/opentelemetry/v1/metrics",
    "OTEL_EXPORTER_OTLP_LOGS_ENDPOINT": "http://127.0.0.1:9428/insert/opentelemetry/v1/logs",
    "OTEL_RESOURCE_ATTRIBUTES": "ip=<host>,job=claude"
  }
}

VIBE merges claude_env with the default settings. Use it to specify API keys or custom endpoints.

Grafana Dashboard

Grafana includes claude-code dashboard by default:

  • Portal: https://<domain>/ui/d/claude-code
  • Direct: http://<ip>:3000/d/claude-code

Runtime Checks

systemctl status code-server
systemctl status jupyter
journalctl -u code-server -f
journalctl -u jupyter -f

Port checks:

ss -tlnp | grep 8443
ss -tlnp | grep 8888

Claude Logs Query

Via VictoriaLogs:

curl -G 'http://127.0.0.1:9428/select/logsql/query' \
  --data-urlencode 'query=job:claude'

Was this page helpful?

Last Modified 2026-10-12 · update extensions (4d4b1d96)