Troubleshooting

When something goes wrong with your site or server, TurboStack gives you everything you need to find the cause and fix it: a live Health view, fleet-wide Monitoring, a full change History, and the TurboStack CLI (tscli) for acting on the running server. This page explains a repeatable method; the pages below cover specific symptoms.

A method that works

1. Reproduce & scope. What exactly happens, on which application/domain, and since when? Does it affect one site or the whole host?

2. Observe. Open the host's Health tab (CPU, RAM, swap, disk and per-service checks) and the Monitoring dashboard (active alerts).

3. Correlate with changes. Check History - Revisions, Deploys and Cloning. A recent change is the most common trigger.

4. Act on the running server. Use the TurboStack CLI (tscli) over SSH to reload services or clear caches - see each symptom page for the exact command.

5. Fix permanently. Make lasting changes in the TurboStack Platform - the GUI or Source (YAML) - and publish. tscli acts on the live server now, but a future publish re-applies your saved configuration.

6. Verify. Confirm the metric or error is gone in Health/Monitoring before you close out.

First places to look

Where	What it tells you
Health	Live resource metrics and per-service status for one host, with Top Issues.
Monitoring	Active alerts across all your hosts, by priority (P1/P2/P3).
History	Recent configuration changes and deploys - and the rollback control.
Logs (over SSH)	Web server, PHP, database and mail logs for the details.
TurboStack CLI	Inspect and act: reload services, clear caches, check the firewall.

Symptom guides

• Site problems: Why is my site slow?, 502, 503 and 504 errors, 403, 413 and 429 errors

• Resources: Out of memory, High CPU and load, Disk full

• Services: Database problems, TLS certificate problems, Email deliverability

• Security: OpenSSH vulnerability alert in a security scan

• Application-specific: see the Troubleshooting page under each app in Applications.

What TurboStack manages for you

Several layers are operated by Hosted Power, so they are rarely the thing you need to fix yourself: the Firewall and TurboShield, Backups, and OS security updates. If you suspect one of these, or an issue needs a managed change, contact Support.

When to contact support

If you have worked through the relevant symptom guide and the problem persists - or it involves a managed layer - reach out via Support. Include the host name and domain, what changed and when it started, and any error message or log output.

Related

• TurboStack CLI
• Health
• Monitoring
• Publishing changes
• Performance tuning
• Glossary - what the terms and abbreviations mean
• Support