Reporting a slow application or server
Describe the user-visible symptom before assuming that more CPU or RAM is the solution. Slow requests may originate in the application, database, storage, external API or network. Useful measurements make the next investigation much more focused.
Collect a short evidence set
- Affected URL or operation, approximate response time, expected behaviour and when the problem began.
- Whether all users or only certain networks are affected, plus relevant times with timezone.
- Recent deployments or configuration changes and representative CPU, RAM, disk-latency and network graphs.
- Redacted application errors or request identifiers that help correlate logs with the incident.
Avoid restarting every component before collecting evidence unless immediate recovery requires it. Do not run uncontrolled benchmarks or load tests on production. Include the service reference and business impact so support can prioritise the investigation within the agreed support arrangements.
