Skip to content

Migrate into Meilisearch

Tested with: Meilisearch 1.54 on Databasezy · zb CLI 0.1

Sources: Self-hosted Meilisearch or Meilisearch Cloud (connection string), another Databasezy instance. Algolia import is tracked separately (coming later).

Per index: create it with the source primary key, copy the settings (ranking rules, synonyms, stop words, filterable and sortable attributes, embedders), page the documents (with _vectors, so nothing is re-embedded) into the target and wait for the tasks to finish. Put the master or an admin API key in the URL password slot.

shell
# --dry-run prints the preflight checklist and the exact plan without changing anything
zb migrate --from "https://:[email protected]" --engine meilisearch --to inst_abc
zb migrate --from "https://:[email protected]" --engine meilisearch --include products --to inst_abc

Follow it with zb migrate status <mig_id>, or in the portal under the instance’s Migrate in tab.

Document count per index (/stats) and a hash of the index settings. Mismatches are listed in the result; nothing is switched for you.

  • Embedder API keys are masked by the source API; re-enter them in the target settings.
  • API keys and the task history are not copied.
  • Documents written during the copy may be missed (no sync).

How migrations work covers preflight, verification and cutover; engine conversions covers moves between engines.