Some bots just traverse links from my main navigation not fetching any other links or resources. I want to log those crawls. Could do it in PHP using PiwikTracker but that would lead to duplicate pageviews from normal users. Javascript call and PHP call would be invoked one after another.
Any tips on accomplishing this?
Thanks!
For solution I was thinking about generating custom variable vith identification value, that would point to those duplicated calls. Registering two websites in Piwik UI – “JS-calls” and second “PHP-calls”. Later on by some means query the database for pageviews with this ID in “PHP-calls” that is not present in “JS-calls”. Pageviews would be still duplicated but in different websites, and my db query would return visits I was describing.
Looking for some simplier way.
My thoughts (Trying to keep it as simple as possible) would be to set up a separate instance of Piwik pointing at a different database. That would keep all calls separate and the dashboard for each would show only the data you wanted.
Hi Valdhor, thanks for input!!!
Mmm that would add some freedom. But still, how that would yield pageviews that hit PHP but didnt hit JS nor noscript tag. Target knowledge demand is how on the timeline such a funky robot is used on my website (if robot at all).
I’m not gonna lie, such requests are rare, and probably they don’t come from “valuable” visitor Piwik is, I suppose, ment to track.
I would check the user agent sent. If it’s one of the ones from http://user-agents.org/ (They have an XML feed) then log it as a bot. If not, then log it as a normal user.
Unfortunately this robot introduces itself as generic Mozilla. Coded this sistem as described in my first post and it works well.
PS. As Piwik isnt any audit tool I think my use case is somehow abusive towards Piwik.