Independent, self-managed infrastructure Read the production requirements

Migration runbook

Migrating BigBlueButton from another provider

The safe unit of migration is the service, not the server. Inventory every API consumer, room frontend, recording path and DNS dependency before the first file is copied.

01 Inventory Versions, APIs, identity, data and traffic
02 Build Clean supported destination and monitoring
03 Rehearse Recordings, integrations and representative classes
04 Cut over DNS/API change with rollback window
Build and prove the destination before changing the public path.

Executive brief

What matters

  1. 01

    Prefer a fresh supported installation and migrate data; do not clone hidden problems into production.

  2. 02

    Recording migration, frontend data and BigBlueButton API consumers are separate workstreams.

  3. 03

    Lower DNS TTL early, freeze changes, define success metrics and keep the source intact through verification.

01

Inventory the whole service

Record BigBlueButton, Greenlight, Scalelite and operating-system versions; hostnames and certificates; API URL/secret consumers; SSO and SMTP settings; firewall and TURN configuration; recordings by state; storage mounts; backups; monitoring; and scheduled automation. Capture current load and a representative health baseline.

02

Build a clean destination

Install the supported BigBlueButton version on a clean dedicated host meeting official requirements. Apply configuration through persistent override mechanisms, establish monitoring and test media from external networks. If the move is also a major-version upgrade, follow the project’s clean-install guidance rather than an in-place shortcut.

03

Move each data class deliberately

Migrate Greenlight/PostgreSQL using application-aware backup and restore. Transfer published and required raw recordings using the BigBlueButton procedures appropriate to your versions, then rebuild or resynchronise metadata where required. Verify sample playback, captions and visibility—not only file counts.

04

Cut over with a rollback condition

Lower TTL in advance, freeze room and configuration changes, take final backups and switch the API hostname or frontend configuration. Run create/join/end/record tests and monitor real sessions. Define the time or error threshold that triggers rollback, and do not destroy the source until retention and stakeholder sign-off are complete.

Evidence base

Sources and further reading

We prefer project documentation and first-party product guidance. Community links are included where they reveal recurring operational questions rather than establish product guarantees.

  1. BigBlueButton installation and upgrade guidance (opens in a new tab)
  2. BigBlueButton customisation and recording transfer (opens in a new tab)
  3. BigBlueButton command-line diagnostics (opens in a new tab)
  4. Community recording resynchronisation discussions (opens in a new tab)

Practical answers

Questions teams ask

Can I copy a BigBlueButton VM image to the new provider?

A fresh supported installation is generally safer because networking, kernel, certificates and hidden state change between environments. Migrate configuration and data intentionally.

Will existing recording links keep working?

Only if hostname, metadata and recording paths remain compatible or redirects are designed. Test real historical links before cutover.

Should the API secret change?

Plan for secret rotation, but coordinate every consumer. A staged endpoint or controlled cutover reduces the risk of silently breaking LMS integrations.