What is the canonical way to do clock skew corrections in v4? #3794
Replies: 1 comment 3 replies
-
I assume in your first example calling Either version do not make an HTTP request to resolve the endpoint. It runs the request through a generated rules engine to determine the endpoint based on the parameters set on the request and the service client config. For most services there is a single endpoint for a region but there are a handful of services like S3 that use request level parameters to determine the endpoint. Again an HTTP request is never made and I'm not sure a cache mechanism would speed up much here since you would need to do an equality check on the entire request object. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
It seems there is little guidance available what the canonical way of doing clock skew correction is. One of of dying it seem to be something like
another way is to use a previous request like
It seems though that this always call the service and there is no caching underneath. While it seems obvious the URL can be cached, it might still be beneficial to know officially whether that is indeed the case or not. Also, having some updated documentation pointing those things out would be nice because keyword search often returns a very outdated blog post from 2025.
Beta Was this translation helpful? Give feedback.
All reactions