Additional information about ten_secondly_rolling limit

[LF-integrations]

Hello,

Recently I have encouraged ten_secondly_rolling limit error.

Can you please tell me how you calculate this?

Is is a kind of "window" rate limit:
- first requests hits at 10:01:23;
- I can send 99 more requests between 10:01:23 and 10:01:33;
- the limit resets at 10:01:33 and I can send up to 100 requests before 10:01:43.

or maybe somethig different, like:

- 100 requests between 00:00:00 and 00:00:10;
- 100 requests between 00:00:10 and 00:00:20.

Thank you in advance,

Przemyslaw

[Willson · Accepted solution]

Hi @LF-integrations 

You're correct with the following outline:

- 100 requests between 00:00:00 and 00:00:10;
- 100 requests between 00:00:10 and 00:00:20.

The ten secondly limit allows for a maximum of 10 calls per second thus 100 calls per 10 seconds as you've outlined above. 

For more information, be sure to check out our documentation around working with HubSpot API Limits.

I hope this helps!

[Sachavskyi]

Hey @Willson! 
We are trying to minimize 429 responses from HubSpot API and have noticed that your ten_secondly_rolling limits don't recover exactly at 00:00:00, 00:00:10, and so on...
E.g.: 
11:10:40 - we started sending requests
11:10:46 - we stopped sending requests (95 requests were sent)
11:10:51 - we started sending requests again, but we got x-hubspot-ratelimit-remaining: 89. Why not 100? ten_secondly_rolling limits had to be fully recovered at 11:10:50.
Could you provide an algorithm or explanation on how the HubSpot API recovers ten_secondly_rolling limits?

[LF-integrations]

Thank you @Willson, it explains a lot.

Have a great day.

Cheers,
Przemysław