← Back to Documentation
Operations
Debug Logs
Collect useful logs and request context for faster support and incident triage.
Key Steps
- Capture timestamps, request IDs, user impact, and affected provider or model names.
- Enable debug logging only long enough to reproduce the issue.
- Redact secrets, provider keys, prompts, and personal data before sharing logs.
Article Scope
This guide focuses on the operational steps needed to run Orchestris with team access, provider control, and predictable usage.
Visual Reference
What to collect
- Server logs around the failing request time.
- Client version, platform, and server URL.
- Provider, model alias, request ID, and error code if available.
Safe handling
- Do not paste provider keys, deployment secrets, or raw tokens into tickets.
- Redact prompt content unless the support path explicitly requires it and the data is approved for sharing.
- Keep logs in your normal secure support or incident system.
Common signals
- Repeated 401 or 403 responses usually point to auth, organization, or role issues.
- Repeated 429 responses usually point to quota, rate limit, or automation behavior.
- Provider 5xx errors may require provider fallback or later retry.
Related Topics
Operations
Usage Analytics
Track requests, tokens, latency, and estimated cost across users, providers, and models.
Operations
Conversation Management
Help users organize work while keeping shared policies and retention expectations clear.
Operations
Multi-device Sync
Keep the same workspace, conversations, and model access available across client platforms.