langgenius/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-07-21 03:55:25 -04:00
Code Issues 775 Packages Projects Releases 10 Wiki Activity

Error in Embedding of Local Knowledge Base, base model is deepseek v3, oracle cloud #7578

New Issue
Closed
opened 2026-02-21 18:21:18 -05:00 by yindo · 3 comments
yindo commented 2026-02-21 18:21:18 -05:00
Owner
Copy Link
Copy Source

Originally created by @lukefan on GitHub (Jan 8, 2025).

Self Checks

  • This is only for bug report, if you would like to ask a question, please head to Discussions.
  • I have searched for existing issues search for existing issues, including closed ones.
  • I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
  • [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
  • Please do not modify this template :) and fill in all the required fields.

Dify version

0.15.0

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

I deployed the application on Oracle Cloud. I'm using pgvector as the vector database and the bge-m3 model on SiliconFlow as the embedding model.
When I try to store articles, it shows a successful operation, but the segmented information inside is empty. Moreover, when I attempt to perform retrieval and recall, it fails. This problem has been present since version 0.14.2.
Even after switching to the DeepSeek V3 base model, the same issue persists. The system still shows a successful storage, but the segmented information remains empty.

✔️ Expected Behavior

I hope that if the segmented information is found to be empty, an error can be reported. This way, we can retry the operation.
3411736313249_ pic

❌ Actual Behavior

Currently, it's marked as successful, preventing any further processing.

Originally created by @lukefan on GitHub (Jan 8, 2025). ### Self Checks - [X] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general). - [X] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones. - [X] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)). - [X] [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:） - [X] Please do not modify this template :) and fill in all the required fields. ### Dify version 0.15.0 ### Cloud or Self Hosted Self Hosted (Docker) ### Steps to reproduce I deployed the application on Oracle Cloud. I'm using pgvector as the vector database and the bge-m3 model on SiliconFlow as the embedding model. When I try to store articles, it shows a successful operation, but the segmented information inside is empty. Moreover, when I attempt to perform retrieval and recall, it fails. This problem has been present since version 0.14.2. Even after switching to the DeepSeek V3 base model, the same issue persists. The system still shows a successful storage, but the segmented information remains empty. ### ✔️ Expected Behavior I hope that if the segmented information is found to be empty, an error can be reported. This way, we can retry the operation. ![3411736313249_ pic](https://github.com/user-attachments/assets/3594cb1b-2584-4dbd-b2b8-d5cc93b8c3c5) ### ❌ Actual Behavior Currently, it's marked as successful, preventing any further processing.
yindo added the 👻 feat:rag label 2026-02-21 18:21:18 -05:00
yindo closed this issue 2026-02-21 18:21:18 -05:00
yindo commented 2026-02-21 18:21:19 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Jan 8, 2025):

The issue you're experiencing with the segmented information being empty, despite the operation being marked as successful, could be related to several factors in the Dify platform. Here are some potential causes and solutions:

  1. Document Content: If the uploaded document contains only images, the extractor might not be able to extract text, resulting in empty segmentation. Ensure that the document contains text that can be processed.

  2. ETL Configuration: Check the ETL (Extract, Transform, Load) configuration. If the Unstructured.IO service is not properly invoked through Dify, it might result in empty text extraction. Ensure that the service is correctly configured and operational.

  3. Segmentation Rules: Incorrect configuration of segmentation rules, such as improper segment identifiers or maximum segment length settings, can lead to empty or improperly segmented information. Review and adjust these settings as needed.

  4. Cleaning Strategies: Dify's built-in cleaning strategies remove meaningless characters or empty lines. If these strategies are not applied correctly, it could result in empty segmentation. Verify that the cleaning strategies are correctly implemented.

  5. Error Handling: Currently, the system marks the operation as successful even when segmentation fails. Implementing a mechanism to report errors when segmented information is empty could help in identifying and retrying failed operations. Dify includes a "Retry on Failure" feature that might be useful in this context [1][2][3][4].

By addressing these areas, you may be able to resolve the issue with empty segmented information.

To continue talking to Dosu, mention @dosu.

Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other

@dosubot[bot] commented on GitHub (Jan 8, 2025): <!-- Answer --> The issue you're experiencing with the segmented information being empty, despite the operation being marked as successful, could be related to several factors in the Dify platform. Here are some potential causes and solutions: 1. **Document Content**: If the uploaded document contains only images, the extractor might not be able to extract text, resulting in empty segmentation. Ensure that the document contains text that can be processed. 2. **ETL Configuration**: Check the ETL (Extract, Transform, Load) configuration. If the Unstructured.IO service is not properly invoked through Dify, it might result in empty text extraction. Ensure that the service is correctly configured and operational. 3. **Segmentation Rules**: Incorrect configuration of segmentation rules, such as improper segment identifiers or maximum segment length settings, can lead to empty or improperly segmented information. Review and adjust these settings as needed. 4. **Cleaning Strategies**: Dify's built-in cleaning strategies remove meaningless characters or empty lines. If these strategies are not applied correctly, it could result in empty segmentation. Verify that the cleaning strategies are correctly implemented. 5. **Error Handling**: Currently, the system marks the operation as successful even when segmentation fails. Implementing a mechanism to report errors when segmented information is empty could help in identifying and retrying failed operations. Dify includes a "Retry on Failure" feature that might be useful in this context <sup>[[1]](https://github.com/langgenius/dify/issues/11649)</sup><sup>[[2]](https://github.com/langgenius/dify-docs/blob/main/jp/guides/knowledge-base/create-knowledge-and-upload-documents/chunking-and-cleaning-text.md)</sup><sup>[[3]](https://github.com/langgenius/dify-docs/blob/main/zh_CN/guides/knowledge-base/create_knowledge_and_upload_documents.md)</sup><sup>[[4]](https://github.com/langgenius/dify-docs/blob/main/zh_CN/guides/knowledge-base/create-knowledge-and-upload-documents/chunking-and-cleaning-text.md)</sup>. By addressing these areas, you may be able to resolve the issue with empty segmented information. <!-- Dosu Comment Footer --> *To continue talking to [Dosu](https://dosu.dev), mention @dosu.* --- <sup>Help Dosu learn! Give it feedback: [Great Response](https://app.dosu.dev/response-feedback/2ac6b7ff-0471-4f43-aa88-62aa01fb0349?feedback_type=great_response) | [Irrelevant Answer](https://app.dosu.dev/response-feedback/2ac6b7ff-0471-4f43-aa88-62aa01fb0349?feedback_type=irrelevant_answer) | [Incorrect Sources](https://app.dosu.dev/response-feedback/2ac6b7ff-0471-4f43-aa88-62aa01fb0349?feedback_type=incorrect_sources) | [Too Verbose](https://app.dosu.dev/response-feedback/2ac6b7ff-0471-4f43-aa88-62aa01fb0349?feedback_type=too_verbose) | [Hallucination](https://app.dosu.dev/response-feedback/2ac6b7ff-0471-4f43-aa88-62aa01fb0349?feedback_type=hallucination) | [Other](https://app.dosu.dev/response-feedback/2ac6b7ff-0471-4f43-aa88-62aa01fb0349?feedback_type=other)</sup>
yindo commented 2026-02-21 18:21:19 -05:00
Author
Owner
Copy Link
Copy Source

@lukefan commented on GitHub (Jan 8, 2025):

Thank you for reporting this issue. After further investigation, we've found some additional details. The same article can be processed successfully when operating locally, but fails when deployed on Oracle Cloud. Strangely, if we delete the article and re-add it, it then succeeds.

@lukefan commented on GitHub (Jan 8, 2025): Thank you for reporting this issue. After further investigation, we've found some additional details. The same article can be processed successfully when operating locally, but fails when deployed on Oracle Cloud. Strangely, if we delete the article and re-add it, it then succeeds.
yindo commented 2026-02-21 18:21:19 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Feb 8, 2025):

Hi, @lukefan. I'm Dosu, and I'm helping the Dify team manage their backlog. I'm marking this issue as stale.

Issue Summary:

  • Bug with DeepSeek V3 model on Oracle Cloud: segmented information remains empty.
  • Successful storage operations but retrieval failures since version 0.14.2.
  • Suggested error reporting for empty segmentation.
  • Issue specific to Oracle Cloud; local processing works fine, and re-adding resolves it.

Next Steps:

  • Is this issue still relevant to the latest version of the Dify repository? If so, please comment to keep the discussion open.
  • Otherwise, this issue will be automatically closed in 15 days.

Thank you for your understanding and contribution!

@dosubot[bot] commented on GitHub (Feb 8, 2025): Hi, @lukefan. I'm [Dosu](https://dosu.dev), and I'm helping the Dify team manage their backlog. I'm marking this issue as stale. **Issue Summary:** - Bug with DeepSeek V3 model on Oracle Cloud: segmented information remains empty. - Successful storage operations but retrieval failures since version 0.14.2. - Suggested error reporting for empty segmentation. - Issue specific to Oracle Cloud; local processing works fine, and re-adding resolves it. **Next Steps:** - Is this issue still relevant to the latest version of the Dify repository? If so, please comment to keep the discussion open. - Otherwise, this issue will be automatically closed in 15 days. Thank you for your understanding and contribution!
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Knowledge Pipeline
Proposal
WIP
cloud
dependencies
feat:webapp
good first issue
outdated
performance
pull-request
Mirrored from GitHub Pull Request
 python:uv
refactor
size:L
stale
status: accepting prs
status: blocked
status: in discussion
stt & tts
vulnerablity
web
wont-fix
🌊 feat:workflow
🌚 invalid
🐞 bug
👻 feat:rag
💪 enhancement
📚 documentation
🔨 feat:tools
🙋‍♂️ question
🙏 help wanted
🤔 cant-reproduce
🤖 feat:agent
No Label 👻 feat:rag
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify#7578
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?