I installed Trigger Happy on a development box in an effort to determine how it affects the server if we log edits everywhere. To set the baseline, I had no user or agent edits and just let TH run for a day. My idea is to build one agent in one db and execute that agent for all edits in all dbs (less admin overhead, easier to set up and maintain).
OK, so I discovered that the server is a busy busy gal…dbdirman.nsf (aka Domino Directory Cache) does a ton of updates. In less than 24 hours, TH logged 3/4 million edits and the log db I'm using grew to about 700 megs.
I'm now cleaning out my log db and revising the trigger to exclude the Cache db.
Several other system dbs also had some activity; Catalog, Monitoring Results (StatRep) and I'm guessing others depending on your server setup.
One thing I need to figure out if I go with the one agent run everywhere option - I want to be able to figure out from the logged data how to know which doc was edited. Using the docID is a non-starter because it's not human friendly. The problem is that to build a human friendly id is going to be different in each db; one might use an instruction and page number, another a user name, etc. Not sure how I'm going to manage that.
The option is to build unique logging agents for each template but that increases the admin/design workload. I'll also need to test server loading if I have unique agents in each db based on a template - does TH require more server resources in either situation or is it a wash?
If anybody has any clever ideas, that would be great.
Still lovin' this app.