Joel Taqueban wrote:
>
> I have a problem which my boss is asking me to answer. It was a
> Sunday (actually yesterday), and during Sundays, from 0300 hrs to 1800
> hrs., this is one of the days that I usually disallow my users to
> access the web. Problem is, my boss was in the office 0930 hrs. and
> was able to access the web! Specific site he was accessing was
> www.cnn.com
>
> That made him furious since he was expecting he would be disallowed
> web access (probably he was testing). I tried to simulate what
> happened, move my system clock to that day and time (Sunday and 0930
> hrs), restarted squid and tried to access the web. I then tried to
> access cnn.com and my Netscape Navigator readily returned an 'access
> denied' message!
>
> I told my boss what had happened, that I was disallowed web access
> during that Sunday. But still he wasn't convinced. He said, I must
> do more investigation, like could it be that the cnn pages are
> already in his cache or on the proxy's cache? I would like to know
> the ff:
>
> * I tried to see the contents of my boss' disk cache but all I saw
> were .jpg and .gif files. How can I tell if my boss is accessing
> his disk cache from these .jpg and .gif files?
> * If it is not from his disk cache, how would I know that the pages
> he fetched were from the proxy?
Look at your proxy logs.
> * And a question I couldn't answer, how would I know the contents
> of the pages on my proxy such that I know cnn.com or other pages
> are there?
> * Or could it be that the proxy was already fetching pages from
> those popular websites before 0930 hrs. Sunday, and stored them
> on the proxy's cache? How would I know what time those pages
> were cached on the proxy? What were the pages cached before 0930
> hrs.?
Do you limit miss_access as well as http_access? If not, people can
probably still fetch cached pages during your browsing ban.
> * How do I tell squid to fetch pages from news pages automatically
> every 4 hrs. for example so that the user is ensured he always
> gets a fresh copy?
Use a refresh_pattern for the site.
>
> Please help. I'm a newbie and honestly, I'm not good at this. I just
> wish someone would be willing to help.
>
> I'm eager to learn and would certainly appreciate someone's answers.
>
> Joel
Received on Sun Jul 25 1999 - 21:40:43 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:47:33 MST