Isolated URL - only found via a canonical
This means that the URL in question is isolated, as it can only be found via a canonical link.
Why is this important?
This means that the URL has no incoming internal links. And if a URL has no incoming internal links, this means it is not really part of the overall website structure, in the sense that users could not navigate to it.
Crawlers are able to access the page - but only by following a link rel="canonical" tag:
Since none of the other pages in the website link to Page C, this means that Page C ends up isolated from the link graph:
Similarly, it is not being assigned link equity/value in the traditional sense.
Canonical tags exist to differentiate duplicate content, they allow website owners to say to search engines, 'hey I know that Page B and Page C are duplicate, and I would like you to include Page C in the index - not Page B.'
The thing is, Google do not always agree with the assessment of site owners, and often end up ignoring the canonical completely. This is an example where that might happen. The canonical tag effectively identifies Page C as more important than Page B. But the internal link signals do not support this, as they are pointing at Page B instead of Page C, which Google will likely find contradictory and thus ignore the canonical assignment.
In most cases, if this Hint triggers, it's because of some sort of unintentional mis-configuration on the website.
What does the Hint check?
This Hint will trigger for any internal URL that has no incoming internal links, and is listed as the canonical URL on another internal URL.
Examples that trigger this Hint:
Consider the URL: https://example.com/page-b
This URL is canonicalized to another internal page,
where this canonical URL https://example/com/page-c has no incoming links from any internal URLs.
The Hint would trigger for this URL: https://example/com/page-c
How do you resolve this issue?
Since it is a canonical URL, search engines are getting the message that it is a 'preferred URL for indexing'. If you want a URL to be indexed, you presumably also want it to rank well, so you would want to be directing at least some internal links at it.
The logical assumption in this scenario is that the URL in question has been accidentally left out of the main site architecture. As such, the solution to this issue is to arrange for the URL to be included in the site architecture, so that it has internal links from at least 1 other internal URL.
Issues like this will typically need to be resolved on a case by case basis.