organization · inferred from evidence
timvisee
A maintainer who confirmed the fix.
- There is no the possibility to configure the initial cluster topology from the qdrant/config/config.yaml file
When deploying a multi-node Qdrant cluster with docker-compose / swarm / kubernetes, the initial cluster topology (bootstrap, peer URIs) could only be passed as CLI flags (--bootstrap, --uri), not configured in config.yaml, making container
- Timeout error on query if timeout=1 in a distributed deployment
On a sharded (multi-node) collection, recommend/query requests fail with a timeout error when timeout=1 is passed, even though the actual execution takes less than a second. Without timeout, or with timeout=2, the same request succeeds.
- Very high server load average after upgrading Qdrant to v1.14.0 with batch search
The maintainers reproduced and confirmed this as a regression in v1.14.0 related to batched search (suspected PR #6326). The fix shipped in the v1.14.1 patch release; maintainer timvisee confirmed on the thread that 1.14.1 includes the patc