Skip to main content

Lakekeeper Iceberg catalog with OLake Go

Lakekeeper is a self-hosted Iceberg REST (representational state transfer) catalog. It runs in Docker or Kubernetes, on S3, MinIO or Azure storage.

Short answer. To write Iceberg tables to Lakekeeper, create an OLake Go destination with catalog_type: lakekeeper, the Lakekeeper REST endpoint, and OAuth2 client credentials or a token. Trino, Spark, DuckDB, ClickHouse, Presto, Doris and StarRocks read it as a REST catalog. Microsoft Fabric reads tables on Azure storage through OneLake shortcuts.

Key takeaways

  • Lakekeeper is a self-hosted Iceberg REST catalog that runs in Docker or Kubernetes on S3, MinIO or Azure storage.
  • OLake Go writes to it with OAuth2 client credentials or a token, and Trino, Spark, DuckDB and ClickHouse read it.
  • Microsoft Fabric reads Lakekeeper tables on Azure storage through OneLake shortcuts.

Last reviewed October 2026.

How OLake Go writes to Lakekeeper​

OLake Go uses change data capture (CDC) where the source has it. It commits Iceberg tables to Lakekeeper over the Iceberg REST protocol. Lakekeeper runs next to your own storage, including MinIO on premises.

No guide on this site uses Lakekeeper yet. The REST catalog page holds the destination fields, and the Lakekeeper and Trino post is a full walkthrough.

When to use it​

  • You run your own catalog on S3, MinIO or Azure storage.
  • Open-source engines such as Trino, Spark and DuckDB are the readers.

When not to use it​

  • Databricks, BigQuery, Athena or Redshift is the main reader. Their docs do not list Lakekeeper as a catalog they read.
  • You use Starburst Galaxy with S3 storage. Galaxy supports Lakekeeper on Azure storage only.

What OLake Go needs

  • catalog_type: lakekeeper
  • Authentication OLake Go offers: OAuth2 client credentials, Token, None.
  • Full destination reference: Lakekeeper setup.

Engines that read Lakekeeper

EngineReadsHow
Apache DorisYesREST catalog with vended credentials
Apache SparkYestype=rest with OAuth2
ClickHouseYesDataLakeCatalog with catalog_type rest
DuckDBYesATTACH to the REST endpoint
Microsoft FabricYesOneLake shortcut to the table folder on ADLS
PrestoYesREST catalog with OAuth2
SnowflakeYescatalog integration (read-only) (Not yet tested on this site.)
Spark on Amazon EMRYesREST catalog
StarburstPartialGalaxy supports Lakekeeper on Azure storage only
StarRocksYesREST catalog with OAuth2
TrinoYesREST catalog with OAuth2

Each engine row comes from that engine's vendor documentation. The dated sources are on the engine's page, linked in the first column.

Guides that use Lakekeeper

No guide uses Lakekeeper yet.

Frequently asked questions​

How does OLake Go authenticate to Lakekeeper?

With OAuth2 client credentials, a token, or no authentication.

Can Microsoft Fabric read Lakekeeper tables?

Yes, for tables on Azure storage, through a OneLake shortcut. See the Fabric page.

Is there a tested guide on Lakekeeper?

Not yet. The Polaris catalog page lists runs on a comparable self-hosted REST catalog.