As discussed in a <long thread> on the <#C06T99N9G...
# fusion-reactor
d
As discussed in a long thread on the #cfml-general channel, one of our CF servers (CF 2018) has an intermittent problem, where after some days, it starts throwing a lot of nonsense errors. In several recent specific cases, nothing was promoted up to production in the past week, so zero code changes, and mid-day we got a blast of these:
Copy code
Element EXECUTIONMODE is undefined in THISTAG
Could not find the included template
The missing template error refers to a template that DOES exist and is used on lots of pages, and is also the one that throws the EXECUTIONMODE error. These are the same errors we've seen before. Clearing the template cache has fixed it, every time, but typically it comes back some days later. In the #cfml-general discussion, one hypothesis is that there's some sort of race condition or conflict between the background processes that compile, cache and execute CFML code. I wondered if FusionReactor has any visibility into those processes, or more generally, if you have any thoughts on what FR can tell us about the cause of these problems, or potential remediations. (As it stands, our standard error handling catches these errors, logs them, and sends email to devs, so they don't show up in FR at all. That's not what I'm asking about. I'm looking for info that can help us understand the compilation and caching mechanisms in CF, and determine if somewhere in there is what's breaking.) Any thoughts are welcome.