Thread Info
  • State Not Answered
  • Locked Locked
  • Replies 4 replies
  • Subscribers 0 subscribers
  • Views 11028 views
  • Users 0 members are here
Options
Suggested
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Time spent on Facebook and Pinterest.

nkadd

We have just put in place our first Sophos Web Appliance and have a couple of test machines (mine and another IT guys) using it as a proxy.  Unfortunately, we are getting unintended consquences when visiting valid sites.  For instance, when you visit a valid news site that has a Facebook commenting feature, it is counting site visits, bandwidth...etc., towards the news site AND Facebook.   It is also doing the same for Pinterest and I'm not even sure how.

Another example is that I have the Dropbox client running on my machine and it is my top visited site!  Keep in mind that I hardly ever visit Facebook and never visit Pinterest or Dropbox.com.

I'm sure this isn't just a problem for Sophos products, but how do you overcome these obstacles if I truly want to know how much staff time is being spent on Facebook or other non-work related sites?

Thanks in advance!

-Bryan

:45481


This thread was automatically locked due to age.
  • 0

    We ran into the same issue when we first implemented the Sophos web filter. There isn't really a fix for it because there really is a web request going out to those sites. Facebook, Pintrest and Twitter were our top three.

    We found that if we looked at the amount of hits the user had and the amount of data transfered we could usually weed out the likely offenders. Typically if it was a hit caused by being packaged with another site there my be a large about of hits but a rather small amount of data. If we found a user with a decent number of both hits and data then we dug in further and looked at the actual url for each hit. It was easy then to tell what was actual Facebook usage and what was just extra baggage from a news site.

    :45753
  • 0

    Thanks for your response Eby!  That makes sense.  

    Does anyone else have any other suggestions?

    Thanks,

    -Bryan

    :45897
  • 0

    Yes, this is a common occurance.   Many websites have the Facebook "like" button and other social media icons on their pages.  The icons and scripts that support this functionality are downloaded directly from their respected sites and thus the requests appear in the reporting.   Unfortunately all the requests don't even use a refferrer  header, so we can't easily exclude them form reporting as we can't verify that they are part of a different website.

    Hope this helps!

    Petr.

    :46581
Unfiltered HTML