This works on Linux Apache access_logs and may work on Windows logs with a couple of tweaks, assuming you have awk, grep, sort and uniq installed.
The following single command line extracts all occurrences of the google_adsense_alt_ads.html, filters out just the referral page, sorts them in alpha order, then counts the # of occurrences and displays the results.
grep "google_adsense_alt_ads.html" -i access_log | awk '{print $11}' | sort | uniq -c
grep finds all the occurrences of the .html file in access_log then pipes it using the "|" character to awk, which then outputs just the HTTP referrer (field #11) from the Apache log entry, which is then piped to sort which outputs the file in alpa order to uniq, which counts and removes the duplicates.
The resulting output looks like this:
1 "apage.html"
3 "bpage.html"
1 "cpage.html"
6 "dpage.html"
etc.
If you only see "1", it means there was just a single occurrence, 1 ad unit didn't display, happens, I wouldn't worry.
If you see a big # for lots of occurrences, I'd check the page and see if the ads have stopped running.
