Implement qdrnt. dify 0.6.12

[koryakovda]

Hello, I am trying to change my VECTOR_STORE from Weaviate to Qdrant on my Windows.

1)In the previous version, it worked fine when I changed the VECTOR_STORE parameter in docker-compose.yaml. However, in the new version, the VECTOR_STORE parameter has been moved to the .env file.

I am using the 'docker compose up -d' command to start Docker with Dify.

Before running docker compose up -d, I changed the VECTOR_STORE in the .env file (line 244) to VECTOR_STORE=qdrant, but I can't access the Qdrant UI at http://localhost:6333/dashboard.

2)I also tried running Qdrant in a separate Docker container:

fceedcc16930 qdrant/qdrant "./entrypoint.sh" 3 hours ago Up About an hour 0.0.0.0:6333-6334->6333-6334/tcp elastic_driscoll

and changed (line 251) in the .env file:
The Qdrant endpoint URL. Only available when VECTOR_STORE is qdrant.
QDRANT_URL=http://localhost:6333/ and http://host.docker.internal:6333

This didn't help.

3)I also tried running Qdrant in the cloud and changed:

(line 251) # The Qdrant endpoint URL. Only available when VECTOR_STORE is qdrant.
QDRANT_URL=https://1722***fe3.us-east4-0.gcp.cloud.qdrant.io:6333
(line 254) # The Qdrant API key.
QDRANT_API_KEY=dgeq*****Fev

This didn't help either.

What do I need to change to make it work for all of these methods?

Thank you!

[dosubot[bot]]

Hello, @koryakovda! I'm here to help you with your Dify issue. I can assist with bug fixes, answer questions, and guide you through becoming a contributor. Let's tackle this problem together!

To successfully change your VECTOR_STORE from Weaviate to Qdrant on your Windows machine using Dify version 0.6.12, follow these steps:

1. Modify the .env file:
   Update the environment variable in the .env file to use Qdrant:

   VECTOR_STORE=qdrant
   

2. Modify the docker-compose.yaml file:
   Update the environment variable in the docker-compose.yaml file for both the api and worker services:

   environment:
     VECTOR_STORE: qdrant

3. Configure Qdrant-specific settings:
   Ensure that the Qdrant-specific settings are correctly configured in the docker-compose.yaml file:

   qdrant:
     image: langgenius/qdrant:v1.7.3
     restart: always
     volumes:
       - ./volumes/qdrant:/qdrant/storage
     environment:
       QDRANT_API_KEY: 'difyai123456'
     ports:
       - "6333:6333"
       - "6334:6334"

4. Update Qdrant settings in the .env file:
   Ensure that the Qdrant-specific environment variables are set in the .env file:

   QDRANT_URL=http://qdrant:6333
   QDRANT_API_KEY=difyai123456
   QDRANT_CLIENT_TIMEOUT=20
   QDRANT_GRPC_ENABLED=false
   QDRANT_GRPC_PORT=6334
   

5. Run the migration command:
   Execute the following command in your terminal or Docker container to migrate the data:

   flask vdb-migrate

   Or, if using Docker:

   docker exec -it docker-api-1 flask vdb-migrate

Additionally, ensure that the necessary ports (6333 and 6334) are not blocked by your Windows firewall or any other network security settings. You may need to create firewall rules to allow traffic on these ports ^{[1][2][3][4][5]}.

To continue talking to Dosu, mention @dosu.

[koryakovda]

Hello,

1)I have modified the .env file:

2)I have modified the docker-compose.yaml file (it has changed in the new version):

3) Configured Qdrant-specific settings (they have changed in the new version):

4) Updated Qdrant settings in the .env file:

5) After all these steps, I started Docker with the command docker compose up -d.

However, I still couldn't get access to the Qdrant UI at http://localhost:6333/dashboard.

[dosubot[bot]]

To successfully access the Qdrant UI at http://localhost:6333/dashboard after following the steps you mentioned, ensure the following:

1. Verify Qdrant Container Status:
   Ensure that the Qdrant container is running without issues. You can check the status using:

   docker ps

   Look for the Qdrant container and ensure it is up and running.

2. Check Qdrant Logs:
   Inspect the logs of the Qdrant container to identify any potential issues:

   docker logs <qdrant_container_id>

   Replace <qdrant_container_id> with the actual container ID of the Qdrant service.

3. Firewall and Network Settings:
   Ensure that your Windows firewall or any other network security software is not blocking port 6333. You can follow these steps to allow traffic on port 6333:

   • Open Windows Firewall settings (Win + R, type firewall.cpl, and press Enter).
   • Click on "Advanced settings" and create a new inbound rule to allow TCP traffic on port 6333.

4. Health Check Endpoint:
   Use the Qdrant health check endpoint to verify that the service is running correctly. You can do this by running the following Python script:

   from qdrant_client import QdrantClient
   
   client = QdrantClient(url="http://localhost:6333")
   
   try:
       collections_response = client.get_collections()
       print("Qdrant service is running correctly.")
   except Exception as e:
       print(f"Qdrant service health check failed: {e}")

5. Docker Network Configuration:
   Ensure that the Docker network configuration allows communication between containers and the host. You can inspect the network settings using:

   docker network inspect bridge

By following these steps, you should be able to diagnose and resolve the issue preventing access to the Qdrant UI ^[1][2].

To continue talking to Dosu, mention @dosu.

[koryakovda]

1. Verify Qdrant Container Status:
   is ok, its up
   81f03f619e6e langgenius/qdrant:v1.7.3 "./entrypoint.sh" About an hour ago Up About an hour 6333-6334/tcp docker-qdrant-1
2. Check Qdrant Logs:
   Seems ok.

According to logs I am obtaining 2 POST reguests to qdrnt database when testing RAG chatbot with Rerank model:

-2024-07-02T14:06:35.595389Z INFO actix_web::middleware::logger: 172.25.0.8 "POST /collections/Vector_index_a6a954ed_8089_4b93_a51e_7bad75a9208a_Node/points/scroll HTTP/1.1" 200 6558 "-" "python-httpx/0.27.0" 0.034362
-2024-07-02T14:06:36.706689Z INFO actix_web::middleware::logger: 172.25.0.8 "POST /collections/Vector_index_a6a954ed_8089_4b93_a51e_7bad75a9208a_Node/points/search HTTP/1.1" 200 18724 "-" "python-httpx/0.27.0" 0.019699

and one POST request when deliting knowlage:
-2024-07-02T14:09:59.912764Z INFO actix_web::middleware::logger: 172.25.0.7 "POST /collections/Vector_index_a6a954ed_8089_4b93_a51e_7bad75a9208a_Node/points/delete?wait=true HTTP/1.1" 200 92 "-" "python-httpx/0.27.0" 0.229652

3. Firewall and Network Settings:
   Ok. Not blocking port 6333

4)Health Check Endpoint:
Faild.

"Qdrant service health check failed: [WinError 10061] No connection could be made because the target machine actively refused it"

5. Docker Network Configuration:
   The Qdrant container is in the docker_default network as it is supposed to be when you use the docker compose up -d command.

@dosu doesent helps

[koryakovda]

How can I connect my https://cloud.qdrant.io/ cluster to Dify, which I run using the docker-compose up -d command?

[dosubot[bot]]

To connect your Qdrant Cloud cluster to Dify, which you run using the docker-compose up -d command, follow these steps:

1. Update the .env File:
   Ensure that the .env file contains the correct Qdrant Cloud endpoint and API key:

   VECTOR_STORE=qdrant
   QDRANT_URL=https://your-qdrant-cluster-url.qdrant.io
   QDRANT_API_KEY=your-qdrant-api-key
   QDRANT_CLIENT_TIMEOUT=20
   QDRANT_GRPC_ENABLED=false
   QDRANT_GRPC_PORT=6334
   

2. Modify the docker-compose.yaml File:
   Ensure that the docker-compose.yaml file for both the api and worker services includes the necessary environment variables:

   services:
     api:
       environment:
         VECTOR_STORE: qdrant
         QDRANT_URL: ${QDRANT_URL}
         QDRANT_API_KEY: ${QDRANT_API_KEY}
         QDRANT_CLIENT_TIMEOUT: ${QDRANT_CLIENT_TIMEOUT}
         QDRANT_GRPC_ENABLED: ${QDRANT_GRPC_ENABLED}
         QDRANT_GRPC_PORT: ${QDRANT_GRPC_PORT}
     worker:
       environment:
         VECTOR_STORE: qdrant
         QDRANT_URL: ${QDRANT_URL}
         QDRANT_API_KEY: ${QDRANT_API_KEY}
         QDRANT_CLIENT_TIMEOUT: ${QDRANT_CLIENT_TIMEOUT}
         QDRANT_GRPC_ENABLED: ${QDRANT_GRPC_ENABLED}
         QDRANT_GRPC_PORT: ${QDRANT_GRPC_PORT}

3. Restart Docker Services:
   Restart the Docker services to apply the changes:

   docker-compose down
   docker-compose up -d

4. Verify Connectivity:
   Use a Python script to verify the connection to your Qdrant Cloud cluster:

   from qdrant_client import QdrantClient
   
   client = QdrantClient(url="https://your-qdrant-cluster-url.qdrant.io", api_key="your-qdrant-api-key")
   
   try:
       collections_response = client.get_collections()
       print("Qdrant service is running correctly.")
   except Exception as e:
       print(f"Qdrant service health check failed: {e}")

By following these steps, you should be able to connect your Qdrant Cloud cluster to Dify successfully ^[1][2][3].

To continue talking to Dosu, mention @dosu.

[koryakovda]

1 Done
2 Done
3 Done
4 Done. Got 'Qdrant service is running correctly.'
But it still doesn't work. I've tried to create 'knowledge' but got an error.

[dosubot[bot]]

To resolve the "403 Forbidden" error when creating knowledge in Dify after connecting your Qdrant Cloud cluster, ensure that the QDRANT_API_KEY environment variable is correctly set and has the necessary permissions to operate on all knowledge bases. Here are the steps to verify and correct this:

1. Verify QDRANT_API_KEY Permissions:
   Ensure that the API key you are using has the necessary permissions to interact with Dify's knowledge creation feature. The API key must have permissions to operate on all knowledge bases.

2. Check docker-compose.yaml Configuration:
   Ensure that the docker-compose.yaml file includes the correct environment variables for both the api and worker services:

   services:
     api:
       environment:
         VECTOR_STORE: qdrant
         QDRANT_URL: ${QDRANT_URL}
         QDRANT_API_KEY: ${QDRANT_API_KEY}
         QDRANT_CLIENT_TIMEOUT: ${QDRANT_CLIENT_TIMEOUT}
         QDRANT_GRPC_ENABLED: ${QDRANT_GRPC_ENABLED}
         QDRANT_GRPC_PORT: ${QDRANT_GRPC_PORT}
     worker:
       environment:
         VECTOR_STORE: qdrant
         QDRANT_URL: ${QDRANT_URL}
         QDRANT_API_KEY: ${QDRANT_API_KEY}
         QDRANT_CLIENT_TIMEOUT: ${QDRANT_CLIENT_TIMEOUT}
         QDRANT_GRPC_ENABLED: ${QDRANT_GRPC_ENABLED}
         QDRANT_GRPC_PORT: ${QDRANT_GRPC_PORT}

3. Enable Debug Mode:
   Enable detailed logging and debugging information to get more insights into the error. Update your .env file with the following settings:

   LOG_LEVEL=DEBUG
   DEBUG=true
   FLASK_DEBUG=true
   

4. Restart Docker Services:
   Restart the Docker services to apply the changes:

   docker-compose down
   docker-compose up -d

5. Check Logs for Detailed Error Information:
   After enabling debug mode, check the logs for more detailed information about the "403 Forbidden" error:

   docker logs <api_container_id>

By ensuring the API key has the correct permissions and enabling detailed logging, you should be able to diagnose and resolve the "403 Forbidden" error ^[1][2][3].

To continue talking to Dosu, mention @dosu.

[StephenLiz]

Log into the qdrant account and check if it still works.it works for me