Niklas

Closing, as consumers in the API server have been migrated to Confluent parallel-consumer.

Alternative connection poolers we can experiment with: * [odyssey](https://github.com/yandex/odyssey) * [pgagroal](https://github.com/agroal/pgagroal) * [pgcat](https://github.com/postgresml/pgcat) * [pgpool](https://www.pgpool.net/mediawiki/index.php/Main_Page)

PgBouncer [1.21.0 was just released](https://www.pgbouncer.org/2023/10/pgbouncer-1-21-0) which includes support for prepared statements. The missing PS support could explain the bad performance when I originally tested the setup.

I'm trying to address a few issues in #756, but to make this bullet proof we'll need a proper `vers` implementation takes care of constraint sorting, simplification, and validation for...

Decoupled the centralized tracking of processing status(es) into #664. Deprioritizing this issue.

Project Koala / Transparency Exchange API will not arrive anytime soon, so we need to evaluate alternatives. I don't think we should introduce a generic blob storage for this yet....

Will be solved via implementation of [ADR-002](https://github.com/DependencyTrack/hyades/blob/main/docs/architecture/decisions/002-workflow-orchestration.md) and [ADR-004](https://github.com/DependencyTrack/hyades/blob/main/docs/architecture/decisions/004-file-storage-plugin.md).

Just dawned on me that a simple solution to this could be to add a random jitter to scheduled tasks.

IIRC we do not ingest any affected versions into the DB, if the data comes from sources that we only use for ad-hoc analysis (OSS Index, Snyk). Reason being that...

Few complications: * If new instances of the API server are started later, previous index events may have already been deleted from the Kafka topic; As a consequence, the instance...