Curious issue affecting users in Japan

[Philip Gladstone]

As part of running PSKReporter, I monitor javascript errors coming from the map page. There are a surprising number of weird errors, but they don't affect many people and it just seems to be random.

As a result of this, when I get a burst of errors (as I did today), I take a look. What I find is that users running old versions of the map page are very often in Japan. I have no idea why. The common error today (which was triggered by somebody reporting a NULL mode), triggered a failure in the web page that hasn't been available for over a month (I report the version number of the web page with every error). 

Having said that, the same issue tripped up a user in the US who was running a version of the page from January 2025. The pages have a cache timeout of 4 hours.

Anybody have any ideas? If anybody wants to reach out off-list privately to explain what they are doing, then I'd be grateful.

The TL;DR is that I fix bugs in the map page code fairly regularly and it is frustrating to see people running buggy versions for months after I've fixed the issue.

Philip

Philip Gladstone

[Tom M0LTE]

Look into ‘cache-busting’ as a technique to make sure that your js and html doesn’t get stuck in intermediate caches, which may or may not be in the control of your users.

--
You received this message because you are subscribed to the Google Groups "PSK Reporter" group.
To unsubscribe from this group and stop receiving emails from it, send an email to psk-reporter...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/psk-reporter/CAMd_wWr_g--bONgn9MW1wkgQwgrypckWQiQThHrfjst9U1SDyg%40mail.gmail.com.

[Mike Karliner]

A little googling found this tool  to show the age of cached items on a page.