Logs & Troubleshooting
When a local stack breaks, the fastest path is usually:
- confirm current runtime state
- check the right logs
- verify site/domain/SSL configuration
Start with state
Section titled “Start with state”Run:
loom currentThis confirms what Envloom thinks is currently active and helps avoid debugging the wrong runtime version.
Logs viewer coverage
Section titled “Logs viewer coverage”Envloom exposes logs for:
- runtime processes
- PHP
- Nginx
- MariaDB
Use the logs viewer first before editing multiple settings at once.
Common issues
Section titled “Common issues”loom command not found
Section titled “loom command not found”- Restart the terminal session.
- Confirm Envloom CLI/shims are installed and available in PATH.
- If the app was just installed or updated, relaunch Envloom.
Local domain does not resolve
Section titled “Local domain does not resolve”- Confirm hosts sync is enabled for the site.
- Confirm the hosts file contains the expected local domain entry.
- Check whether another tool reverted the hosts file.
HTTPS certificate warning in browser
Section titled “HTTPS certificate warning in browser”- Confirm local CA trust completed successfully.
- Confirm the certificate matches the exact local domain.
- Reapply SSL for the site if the certificate was removed or changed.
Nginx starts, but app returns errors
Section titled “Nginx starts, but app returns errors”- Check PHP logs and site PHP version configuration.
- Confirm the site root/path points to the correct project directory.
- Reload services after site config changes.
MariaDB does not start
Section titled “MariaDB does not start”- Check MariaDB logs for port conflicts or permission errors.
- Confirm no other database server is occupying the same port.
Troubleshooting workflow that scales
Section titled “Troubleshooting workflow that scales”Use this order to avoid chasing noise:
loom current- Relevant service logs (Nginx/PHP/MariaDB)
- Site config (domain, path, SSL, per-site PHP)
- Reload services
- Retry request/command
If a change improves the issue, capture it in your team onboarding docs so others avoid the same failure mode.