Operations

Running the platform, and what to do at three in the morning.

During an incident

Start here, in this order.

1. Runbook — the default operator path.
2. On-call — open the service, SLO, and alert dashboards first.
3. Incident response — severities and response targets.
4. Incident timeline template — record it while it happens, not afterwards.
5. Postmortem template — once it is over.

Running it

Page	Read it when
Infra and VPS operations	You are deploying: the production contract for a small VPS, created by OpenTofu.
Backup and restore	You need the backup cadence, or you need to restore.
Observability	You are following one request from browser to gateway to service.
Capacity and cost	You are changing a default. Measure first.
Maintenance template	You are announcing planned work.
Public status	You want the published status snapshot.

Generated references

These two are written by scripts/generate-docs.mjs from the compose files, .env.example files, and the Go source. Edit the source, never the page.

• Published ports
• Environment variables

---

Table of contents

• Infra and VPS operations
• Public status
• Runbook
• On-call
• Incident response
• Incident timeline template
• Postmortem template
• Maintenance template
• Backup and restore
• Capacity and cost
• Observability
• Published ports
• Environment variables