Troubleshooting

Troubleshooting pages are organised by who is hitting the problem, not by which service is at fault. Pick the page that matches your seat, then scan the symptom tables.

• For users — test calls, microphone, audio quality, missing conversations, login.
• For team admins — flows not publishing, tools/variables, API keys, invites, billing.
• For platform operators — container, network, secret, and migration symptoms with pointers into service-specific runbooks.

Each page follows the same shape:

Symptom (what you see)	Likely cause	What to check

When a fix touches production, links go through to Recovery recipes.