To ensure timely and responsible handling of API issues, we will define clear ownership, establish SLAs, and implement a scalable, configurable alarm system. This will support both proactive detection and reactive incident response.
### Subtasks:
- [ ] Update the API maintainers list for REST APIs
- Document ownership for:
- MediaWiki REST
- Wikimedia REST
- Extension APIs
- API Portal
- Other (e.g., AQS)
- [stretch] Action API modules
- Output: A centralized, maintained ownership list
- [ ] Define MWI SLA for API issue resolution
- Define triage, escalation, and resolution timelines in collaboration with the MWI team
- [ ] Research robust monitoring approaches
- Interview teams with mature monitoring setups
- Capture patterns, tooling, and lessons learned
- [ ] Recommend a notification strategy
- Evaluate Slack, email, dashboards, and integrations
- Propose alert routing strategies to relevant teams
- [ ] Implement PoC alarms for MediaWiki REST APIs
- Trigger alarms based on latency, 5xx rates, etc.
- Ensure they are configurable and actionable
- [ ] Expand alarms to remaining REST endpoints
- Apply alarm strategy toDefine MWI SLA for API issue rest.php endpointsolution
- Route alarms to both MWI- Define triage, escalation, and responsible featureolution timelines in collaboration with the MWI teams
### Acceptance Criteria:
- [] Ownership list is published and linked in internal documentation
- [] SLA is defined and reviewed by the MWI team
- [] Notification channels are documented and aligned with MWI team preferences
- [] PoC alarms are implemented and validated for MediaWiki REST
- [] Alarms are extended to the remaining REST endpointsSLA is defined and reviewed by the MWI team