I've been fighting the Google canonical page algorithm bug for quite some time. One advice I received is to set GSC URL Parameters to "Every URL", since we use a page generation script with a "page" parameter (example here), and do not use "Let Googlebot decide". When setting that, if I click on "Show example URLs", GSC shows something like this for recently crawled URLs:
Also I've attached a screen shot: For sure none of these pages exist on our web server. As far as I can tell, our GSC account has not been hacked, at least I don't see any evidence of anyone submitting indexing requests other than me. Entering any of these parameters causes our site to return a hard 404. Why would Google be crawling with random page parameter values ? And a corollary question, could this affect Google's canonical page selection ?