Struggeling with the terminology

[lsmith77]

My goal is to comply with the rate limits on:
https://developers.hubspot.com/docs/api/usage-details#rate-limits

i.e. I want 100 requests per 10 seconds max.
If more requests come in, they need to wait until hopefully the queue has emptied out.

Now trying to implement this with this package I am stumbling over the terminology.

What is clear ->allow(100) should limit it to 100 and ->everySeconds(10) to 10 seconds for this amount.

Now where I am unsure is how to configure how the package should handle additional jobs. Again my goal would be that such jobs 1) do not lead to failures and instead get delayed 2) if they do lead to 429 responses, that they are not considered as failed but simply get delayed (until maybe some timeout, let's say 1h).

First up retryUntil() in the documentation it speaks of timeout. This is the maximum amount of time the job would get delayed until it is considered failed? So I would use return now()->addHour(1); ?

I also do not understand the term release. Does this mean the amount of time a job is delayed by? Why do I need to configure this? Isn't this implicitly clear from the allow() and every...() calls?

[lsmith77]

Alright retryUntil is a core feature documented here https://laravel.com/docs/9.x/queues#time-based-attempts

[lsmith77]

Ok releaseAfterBackoff() seems to make sense if there is no specific rate limit but when you want to prevent a server from getting overloaded.

so I implemented it now as follows:

        $rateLimitedMiddleware
            ->allow(config('hubspot.rate.limit'))
            ->everySeconds(config('hubspot.rate.interval_seconds'));