[Question]: How to import and export the data for ragflow system

[XiaodeZhuang]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (Language Policy).
• Non-english title submitions will be closed directly ( 非英文标题的提交将会被直接关闭 ) (Language Policy).
• Please do not modify this template :) and fill in all the required fields.

Describe your problem

I want to move the ragflow system from old PC to new one, I don't know how to export the data from the old PC and import them into the new PC, could you help to provide the guidelines for it ?
Include the following:

1. what kind of data should I export and import ?
2. How to do for each part ?
3. Is there a function or script to do this easily ?

[dosubot]

To move your RAGFlow system from one PC to another, you need to export and import data from three main components: MinIO (file storage), MySQL (database), and Elasticsearch (knowledge graph chunks). Here’s how to handle each part:

1. What kind of data should you export/import?

• MinIO: Stores your uploaded files and knowledge base data.
• MySQL: Contains all database records, including knowledge base metadata and agent configurations.
• Elasticsearch: Stores the knowledge graph chunks and related search data.
• Configuration files: Especially /ragflow/conf/service_conf.yaml, docker/.env, and docker/service_conf.yaml.template if you have custom settings.

2. How to export and import each part?

MinIO (file storage):

• Method 1 (requires root):
  Copy the MinIO data directory:

  cp -a /var/lib/docker/volumes/docker_minio_data/_data ./_data

• Method 2 (using MinIO client):

  docker exec ragflow-minio mc config host add backup http://127.0.0.1:9000 rag_flow infini_rag_flow
  docker exec ragflow-minio mkdir /root/backup
  docker exec ragflow-minio mc mirror --overwrite backup /root/backup
  docker cp ragflow-minio:/root/backup /tmp/minio-backup

  On the new PC, reverse the process to restore the data
  [source].

MySQL (database):

• Export:

  docker exec ragflow-mysql mysqldump -uroot -pinfini_rag_flow rag_flow > ./ragflow-mysql.sql

• Import on the new PC:

  docker exec -i ragflow-mysql mysql -uroot -pinfini_rag_flow rag_flow < ./ragflow-mysql.sql

  [source]

Elasticsearch (knowledge graph):

• Install elasticdump:

  npm install elasticdump -g

• Export (repeat for each index, e.g., ragflow_documents):

  elasticdump --input=http://elastic:infini_rag_flow@localhost:1200/ragflow_<your index name> --output=./es-analyzer.json --type=analyzer
  elasticdump --input=http://elastic:infini_rag_flow@localhost:1200/ragflow_<your index name> --output=./es-mapping.json --type=mapping
  elasticdump --input=http://elastic:infini_rag_flow@localhost:1200/ragflow_<your index name> --output=./es-data.json --type=data

• Import on the new PC (reverse the above with --input and --output swapped)
  [source]

Configuration files:
Copy /ragflow/conf/service_conf.yaml, docker/.env, and docker/service_conf.yaml.template to the new PC to preserve your settings
[source].

3. Is there a function or script to do this easily?
There is no official one-click backup/restore script for all components; each must be handled separately [source]. However, for advanced users, the Infinity Python API’s RemoteTable class provides export_data() and import_data() methods for exporting/importing knowledge base data, but this requires a good understanding of RAGFlow’s business logic and is not recommended unless you are comfortable with scripting and troubleshooting [source].

Important:

• Always back up your data before making any changes.
• Avoid using docker compose -f docker/docker-compose.yml down -v as it will remove Docker volumes and cause data loss [source].

If you need more details or run into issues with a specific step, let me know!

To reply, just mention my name @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}   

[XiaodeZhuang]

elasticdump --input=http://elastic:infini_rag_flow@localhost:1200/ragflow_ --output=./es-analyzer.json --type=analyzer
elasticdump --input=http://elastic:infini_rag_flow@localhost:1200/ragflow_ --output=./es-mapping.json --type=mapping
elasticdump --input=http://elastic:infini_rag_flow@localhost:1200/ragflow_ --output=./es-data.json --type=data

How to get this value _ ?

[dosubot]

In RAGFlow, Elasticsearch index names follow the pattern ragflow_<unique_id>. The unique_id is typically related to your tenant or knowledge base. To find the actual index names in your Elasticsearch instance, you can list all indices using the Elasticsearch API:

curl -u elastic:infini_rag_flow 'http://localhost:1200/_cat/indices?v'

Look for index names that start with ragflow_. For example, you might see something like ragflow_bd8362ad287f11f09a8728c5d23000bb. Use this full name in your elasticdump command:

elasticdump --input=http://elastic:infini_rag_flow@localhost:1200/ragflow_bd8362ad287f11f09a8728c5d23000bb --output=./es-analyzer.json --type=analyzer

If you have multiple tenants or knowledge bases, there may be several ragflow_... indices. Choose the one(s) relevant to your data export. The naming pattern and how it's constructed in code is described here, and you can see an example of this pattern in use here.

If you need to match a specific knowledge base or tenant, you may need to check your RAGFlow configuration or database for the corresponding unique_id.

To reply, just mention my name @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}   

[XiaodeZhuang]

I am using docker based deployment for ragflow, so if I backup all the mapping folders , I could backup the whole system and restore to another PC ?

[dosubot]

Backing up only the "mapping folders" is not enough to fully back up and restore your RAGFlow system on Docker. RAGFlow stores its data across several components: MinIO (file storage), MySQL (database), and Elasticsearch (search/index data), each using separate Docker volumes or directories. You need to back up all persistent data directories/volumes for these services, plus your configuration files (like service_conf.yaml, .env, and service_conf.yaml.template) to ensure a complete and restorable backup. Each component requires its own backup and restore process—there’s no one-click solution yet. For details on how to back up each part, see the step-by-step instructions here: source, source.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}