ReletRelet Docs

Monitoring

Operate queue, callback, and runtime monitoring with clear escalation paths.

Monitoring is your early-warning system. Use it to detect issues before customers feel impact.

Monitoring goals

These goals keep monitoring actionable and not noisy.

  1. Critical failures surface quickly with clear ownership.
  2. Repeated warning signals trigger preventative action.
  3. Incident evidence is complete enough for fast escalation.

Monitoring workflow

Use this routine during operational checks.

Review active queue backlogs and failed job trends.
Review callback health and non-success response spikes.
Review subscription and payment-related warning patterns.
Escalate recurring anomalies with timestamps and impact scope.

Alert response priorities

Prioritize using operational impact first.

  1. Critical: data loss risk, billing lockouts, or broad workflow outage.
  2. High: repeated callback failures or sustained queue backlog growth.
  3. Medium: intermittent failures with safe retry and no customer impact.
  4. Low: isolated anomalies with no trend signal.

Operational best practices

Use these practices to keep monitoring effective.

Release window pairing

Always increase monitoring depth during release windows, billing-cycle events, and known high-throughput periods.

Next steps

  1. Verify provider dependencies in Payment providers.
  2. Verify callback handling in Webhook callbacks.
  3. Use Troubleshooting for incident response.

Last updated on

On this page