Actions26
- Delete Collection
- Get Collection
- Check Collection Exists
- Create Collection
- Update Collection
- Retrieve Point
- Retrieve Points
- Upsert Points
- Delete Points
- Update Vectors
- Delete Vectors
- Set Payload
- Overwrite Payload
- Delete Payload
- Clear Payload
- Batch Update Points
- Scroll Points
- Count Points
- Query Points
- Query Batch Points
- Query Points Groups
- Matrix Pairs
- Matrix Offsets
- Payload Facets
- Create Payload Index
- Delete Payload Index
Overview
The node provides an interface to update vectors in a Qdrant collection. It allows users to modify existing points by updating their vector representations within a specified collection. This operation is useful in scenarios where vector data needs refinement or correction, such as improving search relevance, updating embeddings after retraining models, or correcting erroneous vector entries.
Practical examples include:
- Updating user profile embeddings in a recommendation system after new behavior data is collected.
- Refining image feature vectors in a multimedia search engine when improved feature extraction algorithms are deployed.
- Correcting or augmenting vector data in a knowledge base to enhance semantic search accuracy.
Properties
| Name | Meaning |
|---|---|
| Collection Name | The name of the collection in which the vectors will be updated. |
| Points | A JSON array listing points to update. Each point must have an id and one or more vector arrays representing the new vector values. Example: [{"id":0,"vector":[0.9,0.1,0.1]}, {"id":1,"vector":[0.1,0.9,0.1]}]. |
| Shard Key | (Optional) Specifies which shards to look for the points. If not provided, all shards are searched. |
| Wait | Boolean flag indicating whether to wait for the changes to be fully applied before continuing. |
| Ordering | Defines ordering guarantees for the update operation. Options are: Weak, Medium, Strong. |
Output
The node outputs JSON data reflecting the result of the update operation. Typically, this includes confirmation of successful updates or error details if the operation failed. The exact structure depends on the Qdrant API response but generally confirms which points were updated.
If binary data output is supported, it would represent raw responses or related binary payloads from the Qdrant service; however, this node primarily deals with JSON vector data.
Dependencies
- Requires connection to a Qdrant vector database instance.
- Needs an API key credential configured in n8n for authenticating requests to the Qdrant service.
- The base URL of the Qdrant instance must be set in the credentials.
- The node uses HTTP requests with JSON payloads to interact with the Qdrant API.
Troubleshooting
Common Issues:
- Incorrect collection name: Ensure the collection exists in Qdrant before attempting updates.
- Malformed points JSON: Validate that each point has a valid
idand properly formatted vector arrays. - Shard key misconfiguration: If specifying shard keys, verify they correspond to existing shards.
- Network or authentication errors: Confirm API credentials and network connectivity to the Qdrant instance.
Error Messages:
- "Collection not found": The specified collection does not exist. Create it first or correct the name.
- "Invalid vector format": Vectors must be arrays of numbers matching the collection's vector size.
- "Unauthorized" or "Authentication failed": Check API key validity and permissions.
- Timeout or wait-related errors: If
Waitis true, ensure the Qdrant service is responsive and can process updates timely.
Resolving these typically involves verifying input parameters, credentials, and Qdrant server status.