← Back to Staff Overview

Operator

Operators own live-server stability and safe operations. You deploy updates, watch the consoles, and keep GoPlay online, fast, and recoverable.

Access: ✅ Full (Prod + Test)
Status: Operations
Review Cycle: On-call & Change Board

What You’ll Do

  • Deploy releases & hotfixes; manage restarts and maintenance windows.
  • Monitor console/logs, TPS, latency, and error rates; triage incidents.
  • Own backups, rollbacks, and recovery runbooks.
  • Coordinate with Dev on config/plugin changes and with Admin on comms.
  • Track SLOs and publish postmortems for major incidents.
Live Server Access: ✅ Full (Owner-approved keys)
Releases Hotfixes Backups Rollbacks Runbooks Observability Incident Triage Change Windows
Personality Match:
Calm Under Pressure Clear Communicator Systems Thinker Decisive & Safe Accountable Status-Update Friendly

Role Objectives

  • Minimize downtime; predictable change windows.
  • No data loss; verified backups + restores.
  • Fast incident MTTA/MTTR with clear comms.
  • Stable performance under peak load.

Success = reliable uptime, safe changes, and quick recoveries.

Tools & Access

  • Prod/Test consoles, log viewers, metrics dashboards.
  • Backup system (snapshots + restore validation).
  • Release/rollback scripts, incident runbooks.
  • Change board & maintenance calendar.

All production changes follow a checklist and second-pair review.

Policies & Safety

  • Two-person rule for irreversible actions and data changes.
  • Pre/post-change announcements in #staff-logs and status page.
  • Document incidents with timeline, impact, and remediation.
  • Security hygiene: key rotation, least privilege, audit trails.
Targets: MTTA < 5m • MTTR < 30m (P1) • Verified backups daily

Works Alongside

Click a pill to toggle its Discord role color. Right-click to reset to neutral.

Quick Start (First Week)

  • Say hi in #staff-hq; review on-call expectations.
  • Walk through runbooks: backup/restore, release, rollback, incident.
  • Shadow a maintenance window; run the next one with a mentor.
  • Trigger a test restore in a sandbox and capture timings.
  • Publish a brief ops report (findings, risks, next steps).