URL hijacking

I’ve been using the online resources provided by the university library a lot lately.  Mostly I’ve been accessing online journals and downloading pdfs of articles that are pertinent to my work.  Generally, access to these journals seems to be afforded to people on campus by routing the request from the library website through a proxy server that the library maintains.

Unfortunately, I’ve found that basically any URL ending with mit.edu gets hijacked by the proxy server or my browser somehow routes the request through the proxy server if I’m using Firefox.  This isn’t a problem for sites that don’t require authentication but it’s produced a number of problems for sites that do require authentication.  Basically, I can no longer access the online purchasing portal or even the student services portal using Firefox.  This might not seem like a huge problem since I should just be able to use a different browser.  However, MIT doesn’t provide a means for installing personal certificates in Chrome or IE 8 (which are the only other browsers I have installed), so I can’t use these browsers to access sites requiring certificate-based authentication using alternate browsers either.  I ended up having to use a different computer altogether in order to pay my student account and register for the Spring semester.

I’ve been in contact with the help desk at the library.  They’ve offered some suggestions but, so far, nothing has worked.  First, they recommended that I check the proxy settings in Firefox to make sure that I didn’t have the library proxy set as my means of connecting to the internet.  I checked my settings and, sure enough, the option selected was “No Proxy”.

They suggested clearing my browsing history in Firefox.  So, I opened Firefox and chose Tools -> Options -> Privacy -> “clear your recent history”.  I told it to remove everything and I selected all the various types of data available to delete – browsing and download history, form and search history, cookies, cache, active logins, and site preferences.  After clearing them I attempted to go to the MIT purchasing portal but the request was redirected again through the library proxy.  I tried typing a few characters of the URL and I found that the address bar still offered a bunch of choices from sites I had visited in the past.  What?  I thought this stuff was supposed to be cleared!

So, I downloaded CCleaner, a donation-ware program that can search your system and clear out temporary files, cached files, and various things that browsers tend to accumulate.  So, I ran that and cleared out everything including my history and cache in IE8 and Chrome.  I went back to Firefox but, although everything was gone, if I chose View -> Sidebar -> History (Ctrl H) or History -> Show All History (Ctrl + Shift + H), if I type a few letters in the address bar, it still suggests to me sites that I’ve visited in the past.  How do I get rid of the stuff in the address bar?

It turns out that the sites that still come up in the address bar are sites I’ve bookmarked.  I had forgotten that the address bar in Firefox also searches your bookmarks as well as your browsing history.  So, my history is clear after all and using CCleaner was unnecessary.  That said, it did clear out like 2GB of temporary files, which is a nice little bonus.

I was at a loss about what to do when I had a flash of insight – maybe the problem is related to the DNS cache.  So, I followed these instructions and flushed my DNS cache.  I was certain this would work.  But, it didn’t.

As a last resort, I decided to check my Firefox add-ons.  My first thought was that the problem could be related to NoScript since it’s the most invasive of the add-ons that I use.  However, the problem persisted even when I turned NoScript off.  Then, looking through the list of add-ons I use, I spotted Zotero.  This add-on is for managing journal article and book references.  I have used this add-on extensively during my library research, so I decided to look under the hood.  What I found was this:

I took a screenshot of this for future reference and then removed the entry.  Voila!  I’m now able to access the purchasing portal, the student services site, etc.

So, if you’re having weird problems with certain sites being redirected through library proxies and you have Zotero installed, Zotero is probably the culprit.  I suggest you start by checking on Zotero’s proxy settings.  If there isn’t anything in the list or if removing items on the list doesn’t fix your problem, try some of the other things I tried.  In particular, I recommend you check Firefox’s proxy settings, followed by flushing your DNS cache.

I hope that someone will find this useful!

2 Responses to “URL hijacking”

  1. 1 Amanda Giermann
    January 26, 2010 at 5:34 pm

    The suspense was killing me reading this, since I certainly looked up a lot of journal articles via the library but never had the proxy problem. If you haven’t already, you should pass this info on to the library in case someone else calls with the same problem :o)

  2. 2 Colin
    January 26, 2010 at 5:39 pm

    Just be glad that I figured it out as quickly as I did – it could have gone on for paragraphs and paragraphs! Anyway, yeah, I let the people at the libraries know. Hopefully, if they get questions from other people they’ll be able to get to the bottom of it quickly.

    That type of thing is kind of the purpose of this blog. As I figure out things that had been bugging me, I try to post my solution so other people don’t have to waste a bunch of time on the same thing. That said, my blog probably comes up on page 100 of any Google search but at least it’s there.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


January 2010

Recent Twitterings

Follow Me on Twitter

RSS That to which I am listening

  • An error has occurred; the feed is probably down. Try again later.

%d bloggers like this: