Hi there. A Cloudflare-protected API I am calling from my background worker is rate-limiting my app on any calls, but only when the calls are made from Render. On speaking with the team, they believe that CF is rate-limiting my calls as there are other calls coming from the same IP blocks that Render uses, and CF rate limits are based on IP.
Is there any way to pay extra to add a dedicated outbound IP to a service? If not, I’ll have to move to another provider.
We don’t have a way to provide dedicated outbound IPs but I’m intrigued what the API you’re interface with is here? Feel free to respond to firstname.lastname@example.org if you don’t want to share it here,
OpenSea is the API - they use CloudFlare sitting in front of their API servers and while the calls work from my local dev env, AWS, and a few other places, all calls from my Render service (using curl in the Render console and via Axios/Node), fail with a CF 429 error.
an interesting experiment to perform if you’re able would to be deploy your service (or a dummy service that you can curl from) to one of our different regions to see if you’re rate limited from there.
Yes, works from the SG region, so moving there for now. Hopefully there aren’t other services nuking the rate limits.
Would pay a fee every month for a dedicated IP BTW!
Right, but that might be a temporary fix - just be mindful of that.
It’s interesting that the Opensea docs only talk about rate limits per API key as opposed to an IP rate limit.
Yes, understood on the temporary, which will require us to close out our Render account should another user start to abuse the rates.
On the rate limits, it seems they have both in place unfortunately.