Helicone Community Page

Updated 3 months ago

Seeing request timeouts

Hi there, we're seeing an increased volume of these errors in our app:

Plain Text

Request timed out: HTTPSConnectionPool(host='oai.hconeai.com', port=443): Read timed out. (read timeout=600)

Is there an uptime checker we can use to track the health of the API?

13 comments

JJustin

there is status.helicone.ai

JJustin

Can you share what your request looks like?

aabe

Thank you -- it seems like our timeouts are not showing up on this status page

aabe

I'm not sure what the request looks like, this is a production system where I'm just observing the logs. My guess is its an embedding call to OpenAI

JJustin

hmmm

JJustin

Are you seeing the request at all in Helicone?

JJustin

I would also like to rule out that the timeouts are not coming from OpenAI if possible

aabe

We have a really large volume of requests, so its difficult for me to identify which one it is in the helicone dashboard

VVinay Varma

We are facing the same issue since an hour. It is happening intermittently and especially with models like gpt-3.5-turbo-1106. We noticed that it wasn't happening with gpt-4

CCole

Hi , sorry you're experiencing this. Are you seeing the requests in Helicone at all?

CCole

Could you add me to your organization so I can take a look? My email is [email protected]. Thanks!

VVinay Varma

hey no, I could not see the requests that were hanging up.

Add a reply