Troubleshooting AI Autodocumentation

Common issues

This page allows to solve common issues encountered during work with AI Autodocumentation feature. If you don't see your issue here, please contact our Support team.

Rate limit reached in organization

Error text

Rate limit reached for gpt-4 in organization org-XXXXXXXXXXXXXXXXXXXX on tokens per min (TPM): Limit 10000, Used 7803, Requested 5731. Please try again in 21.204s. Visit https://platform.openai.com/account/rate-limits to learn more.

Potential causes

The situation typically occurs during creating descriptions for tables/views with a lot of columns. It is caused by the limits imposed by OpenAI. You can check your limits on the page https://platform.openai.com/account/rate-limits.

Solutions

• Wait a moment (the time is specified in the error message) and press the buttons Rerun generation under each object to generate missing descriptions.
• If you use the GPT-4 engine, try to switch to the GPT-3.5 engine which has bigger limits.
• You can also lower the Max tokens value in AI configuration (please note that the generated descriptions will be shorter). See the article Configuring AI Autodocumentation for GPT for further information.

SSL connection could not be established

Error text

The SSL connection could not be established, see inner exception.

Potential causes

The AI Autodocumentation requires access to AI provider's API. If the Dataedo Portal is installed without access to the internet, the connection to AI provider's API could not be established.

Solutions

• Enable connection to the internet.
• To use the feature without enabling access to the internet, add the address of AI provider's API (https://api.openai.com/v1/chat/completions in the case of OpenAI) to the firewall exception list or another solution used to block internet access.