Troubleshooting
Troubleshooting and operational playbooks for common Curio issues.
Last updated
Troubleshooting and operational playbooks for common Curio issues.
This section is built from real operator issues seen in #fil-curio-help, with notes tied back to the actual Curio code paths.
Start here if you are:
stuck during setup/migration,
seeing task failures in the UI,
seeing repeated errors/warnings in logs,
unsure what information to collect before asking for help.
Most Curio incidents fall into one of these buckets:
Chain / Lotus connectivity
Curio can’t talk to the full node, or the full node is unhealthy/out of sync.
Control plane (DB + scheduler)
Yugabyte is slow/unhealthy, schema upgrades failed, or tasks are not being scheduled/claimed.
Data plane (storage + retrieval)
the bytes aren’t where the system thinks they are (paths, indexes, URLs, permissions).
If you don’t know where to start: check DB health first. A sick DB creates “random” symptoms everywhere.
Collect debug info (copy/paste checklist)
Common errors (error string → meaning → next step)
Last updated