Steve, if you’re thinking “so there must be a problem with something as simple as CFOUTPUT or CFLOOP” (and you figure your server can’t be the problem), I’ll assert that’s not the case. But it’s a frequent source of confusion.
Instead, the issue is that it’s the tag BEORE that which reports the timeout which took a long time, and therefore took the whole page beyond your set CF timeout (whether the CF Admin timeout, or the CFSETTING timeout). For more on this issue, see a blog entry I did: http://www.carehart.org/blog/client/index.cfm/2010/10/15/Lies_damned_lies_and_CF_timeouts (that’s a long URL, so if it splits over two lines, please just join it back together). I’ve also done a couple of talks on discussing that in the context of some other common surprising challenges, http://carehart.org/presentations/#10plagues, if that may interest you.
So it could be that your CF server was trying to do a CFQUERY to your DB server, and though yes BlogCFC does caching, it could be that this was hanging up on the initial call for a given potentially cachable thing, and therefore it was never cached.
Of course, the problem you had may have been only temporary, so that you may find now that all it working. But if it happens again, hope the info above may help (or may help others.)
/charlie
Hmm. BlogCFC typically caches pretty heavily. Have you checked to ensure you are on the latest version?