Hi Don664,

Then the first thing I would do is to change them back and see if the problem goes away.  If it does then I would move forward again making just one  change at a time, until it comes back. 

Sometimes changes are not divisible so maybe you have to change two things at once.

 

I’m (reasonably) sure that if an App pool recycles then there is no notification to the applications – it’s supposed to be transparent to them.  It seems a reasonable bet that there would be something in the server’s Event Log.  That’s where I’d look first for it.

---

Best wishes,
- Richard
Agile Development Consultant, Practitioner, and Trainer
www.dynamisys.co.uk

Problem solved!

Turns out that it was the DNN News Feeds module which was getting headlines from a feed that had over 2300 items in it. This feed basically had news items appended to it daily with no purging of older items. So the file size was well over 400kb.

We had the cache time set to 5 minutes and that correlates to the reason why every 7 minutes, or so, the login would grind to a halt. It was because the feed cache was being emptied and then re-populated from an XML file that was huge. As a result of the homepage being the default page after login it explains why it seemed that logins were the problem.

With that module having been on the site since launch it was always going to be the last suspect... so a huge thank you must go to Joe Craig from DNN Creative (http://www.dnncreative.com/) who tracked it down for me!