API rate limiting in 2023

[FlyingFrames]

Meraki APIs are set to 10 API calls every second but that equates to 864,000 API calls in a day per organization.

https://developer.cisco.com/meraki/api/rate-limit/

Other vendors stop short of one tenth of that e.g. 5000 per hour. Just confirming if the above are true & someone has experienced crossing the above limits!

[RaphaelL]

I'm hiting the rate limit every single day. It just depends on the size of your Org , number of admins doing API calls and if you are using asyncio or not. 

[Oren · Accepted solution]

That is true, the current API rate limit is 10 calls per second, resulting in 864,000 call per day.

It is possible to reach the rate limit, when several apps/scripts run at once.

[PhilipDAth]

I frequently hit the rate limit.  Especially in orgs with multiple integrations.

I use the Python SDK, and it's retry logic often works.  I frequently have to reduce the number of concurrent outstanding API requests with:

	async with meraki.aio.AsyncDashboardAPI(
		output_log=False,
		print_console=False,		
#		maximum_retries=100,
		maximum_concurrent_requests=5
	) as dashboard:

In busier orgs this is not enough, and I have to also use a throttler.

import throttler

	async with meraki.aio.AsyncDashboardAPI(
		output_log=False,
		print_console=False,
		maximum_retries=100,
		wait_on_rate_limit=True
	) as dashboard:
		dashboard._session._concurrent_requests_semaphore = throttler.Throttler(rate_limit=4, period=1.0)
		await ...

[FlyingFrames]

Is there a specific error code that Meraki will give back? e.g. error 429 once rate limit is exceeded?

[RaphaelL]

Yes , Meraki will return HTTP 429 codes.

https://developer.cisco.com/meraki/api-v1/rate-limit/#per-source-ip-address