denis256 notes and projects

ingestr v1.0.39 was published on June 23 2026. The release adds MongoDB change data capture support, a Twilio source, and streaming queue broker sources. It also adds an ingestr load timestamp column and updates the server UI to match CLI capabilities.

The full release notes and downloads are on the GitHub release page.

MongoDB change data capture source

The release includes a new MongoDB CDC source. Change data capture lets pipelines read inserts, updates, and deletes from MongoDB collections as they occur. This approach avoids repeated full collection scans that load both the source and the network. Downstream systems receive only the changed records and can apply them directly.

Catalog changes accompany the source. MongoDB CDC schemes were added to the server connector catalog. A dedicated catalog entry is now used for MongoDB CDC.

Benchmark ingestion paths for MongoDB were optimized in this release. Operators running load tests or large syncs may see better performance on the CDC paths.

Twilio source

A Twilio source was added. Teams that use Twilio for messaging or communications can now pull relevant records through ingestr using the standard source configuration.

The source entry was added to the connector catalog. Table validation logic for Twilio was updated to accept granularity modifiers.

Some of the supporting changes aligned the Twilio work with patterns already used for the sendgrid source. This makes it possible to bring Twilio activity data into the same warehouse or lake used for other business data.

Streaming queue broker sources

Streaming queue broker sources were added to ingestr. This gives users a way to consume data that arrives through queue brokers rather than only through batch oriented connectors.

Queue based streaming fits use cases where events are published continuously and the destination needs to stay close to real time. The change sits alongside existing source types. The queue support was one of several source expansions included in the release.

Load timestamp column and DuckDB staging

An ingestr load timestamp column was added. When data lands in the target, a column records the time at which ingestr performed the load. This column helps with auditing, incremental processing, and freshness checks in downstream queries. The column is named in a way that makes its purpose clear when inspecting loaded tables.

For DuckDB destinations the replace staging logic now uses the target schema. Previously, replace operations could place staging tables in the default schema even when a target schema was configured. The fix ensures consistent behavior when a target schema is set on the destination.

Server UI parity and container fixes

The server UI gained CLI parity. Capabilities that exist on the command line are now available through the server interface as well.

Curl was restored to the OCI image. Health check workflows and container setups that call curl for readiness or liveness probes will function without extra workarounds.

A fix addressed schema evolution arrow ownership. Streaming transform test coverage was also restored during this cycle. These changes improve stability for data flows that rely on arrow structures and streaming transforms.

Where to get it

• The full release notes and downloads are on the GitHub release page.
• The project repository is at https://github.com/bruin-data/ingestr
• The release tag is v1.0.39