On-call toolkit
Troubleshooting runbook
These are the same steps VaultScope engineers use during incidents. Start at the top, gather evidence and loop in support with full context if you need a hand.
1. Server fails to start
- Check the console output for stack traces or port binding errors.
- Verify the main jar or binary exists and the start command references the correct filename.
- Ensure RAM allocation matches the JVM flags—out-of-memory kills appear as exit code 137.
- Run a manual backup before retrying so you can roll back if the disk becomes corrupted.
Still stuck? Zip the latest log file and attach it to your support ticket for faster analysis.
2. Players cannot connect
- Confirm the server is online in the panel and not in a suspended or stopping state.
- Verify the correct allocation is assigned under the Network tab.
- Check firewall rules or proxies for recent changes.
- Test connectivity using
mtrorpingfrom a remote machine. - Rotate subdomains or DNS records if propagation issues linger.
3. Panel or file manager errors
- Hard refresh the browser or try a different profile to rule out cached assets.
- Check status.vaultscope.dev for active incidents.
- Collect console logs from the browser DevTools if a specific request returns 500.
- Attempt the action via SFTP or the API to confirm whether the issue is UI-specific.
Include request IDs or timestamps when contacting support—we can trace them in our application logs.
4. Performance degradation
- Review resource graphs in the panel for CPU or memory exhaustion.
- Capture a profiler snapshot (spark, WarmRoast or VisualVM).
- Check the Schedules tab for overlapping automated tasks causing spikes.
- Compare current plugins or mods against a known-good backup. Revert recently added modules to isolate the issue.
Share profiler output with performance@vaultscope.dev if you need tuning suggestions.
5. Database problems
- Confirm the database exists and credentials are valid in the Databases tab.
- Check that your IP is whitelisted if connecting externally.
- Inspect server logs for deadlocks or slow query warnings.
- Restore from the most recent backup if corruption is detected.
The database guide covers common maintenance tasks in more depth.
6. Escalating to VaultScope support
When opening a ticket or emailing support@vaultscope.dev, include:
- Server ID, region and the exact timestamp the issue started.
- Steps to reproduce and whether the problem persists after a restart or backup restore.
- Relevant logs, profiler output or screenshots.
- Impact assessment (e.g., player count affected, revenue, tournaments).
For critical outages, submit the ticket first, then ping the on-call role in the VaultScope Discord. Include the ticket number so the engineer can pick it up immediately.
Post-incident checklist
- Document the root cause, impact and resolution steps in your runbook.
- Create or update schedules and alerts to prevent recurrence.
- Share learnings in #operations so the wider team stays informed.
- Celebrate the fix—operational excellence is a team sport.