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

Sophos ID - login problems when using Firefox

Hi folks

Sorry, not a UTM-related question, but I was just wondering if anyone else had encountered any problems when using Firefox to log into the Sophos site (and thus this forum)?

Normally I use Debian, but due to the need to run an x86 package, I've been booting into Windows 10 for the past couple of weeks. As a result of that, I have [successfully] been using the Windows version of Firefox (72.0.2) to access this forum, but over the past 2 days, rather than presenting me with a login prompt, the site has instead been redirecting me to 3.id.sophos.com/.../tcs, which presents me with the below error message:

I do have Sophos in my Noscript whitelist, I have tried clearing cookies and site data (for all Sophos entries) and I've tried lowering the 'enhanced tracking protection' from strict to standard, but none of the above tricks helped.

Doubtless I've missed something obvious, but at the moment, I cannot think what it could be; anyone have any cool ideas?

Kind regards,

Briain

PS Used [original] MS Edge to post this. 



This thread was automatically locked due to age.
  • Hi  

    I'm able to open the Sophos Community website successfully with default Firefox settings. If you're still facing the issue, would you please PM me details of this?

    Regards

    Jaydeep

  • No problem here either, Briain.

    Cheers - Bob

     
    Sophos UTM Community Moderator
    Sophos Certified Architect - UTM
    Sophos Certified Engineer - XG
    Gold Solution Partner since 2005
    MediaSoft, Inc. USA
  • Hi  

    Hi  

    Thank you both for confirming that you can gain access via Firefox. Based on that information, I tried completely disabling the Firefox NoScript add-on and that did the trick. What's mighty odd is that there already were 2 Sophos related sites in the NoScript white-list (and that these worked fine until this issue first arose, earlier this month).

    After again enabling NoScript, I managed to capture the URL prior to the one shown in my first post (I'll append it to the end of this post) and I noted that it began 'https://login.sophos.com/'. Having looked at the white-list, I noted the below 2 sites were already added as being trusted (and I'd have thought that the first one would have covered it)...

    sophos.com

    us-central1-sophos-home.cloudfunctions.net

    ...but just to be sure, I manually added login.sophos.com, but unfortunately, that didn't help.

    It's odd as NoScript does show when there's an additional site (needing to be trusted) by changing its logo, so I wonder if there's yet another URL which appears so briefly that the NoScript add-on doesn't have time to show a change to its logo (before it's redirected to the 3.id.sophos.com/token_proxy URL)?

    It's strange as I've never encountered an issue this on any other sites - I can always add a site via clicking on the NoScript icon and trusting it - so if anyone has any cool thoughts about other URL(s) that I could try manually adding (to the NoScript white-list), please let me know.

    Kind regards

    Briain

    PS Below shows the format of the [edited to save space] 'interim' URL that I captured (i.e. just before it re-directed me to 3.id.sophos.com/token_proxy):

    https:// login.sophos.com/login.sophos.com/B2C_1A_signup_signin/api/CombinedSigninAndSignup/error?code=UX004&diags=%7B%22pageViewId%22%3A%22374904e4-34d3-4ab4-b585-3b4f982b44ab%22%2C%22pageId%22%3A%22CombinedSigninAndSignup%22%2C%22trace%22%3A%5B%7B%22ac%22%3A%22T005%22%2C%22acST%22%3A1582209988%2C%22acD%22%3A5%7D%2C%7B%22ac%22%3A%22T027%22%2C%22acST%22%3A1582209988%2C%22acD%22%3A0%7D%2C%7B%22ac%22%3A%22T021%22%2C%22acST%22%3A1582209988%2C%22acD%22%3A58%7D%5D%2C%22code%22%3A%22UX004%22%7D&csrf_token= ---cut to save space--- ==&tx=StateProperties=---cut to save space---gifQ&p=B2C_1A_signup_signin&desc=https%3A---cut to save space---%2Fhtml%2Fsign-in.html

  • Hi 

    I haven't been here for ages due to the above login problem (having to completely disable NoScript before I can get to the login page) so today I used Wireshark to capture what was going on. There weren't many clues, but after using nslookup on some of the IP addresses, I found a few IPs were related to cloudfront.net servers, then after creating an exception (in NoScript) for cloudfront.net, it all worked as it should. Of course, I'm not really wanting to let everything from cloudfront.net to be able to run scripts in my browser, so I was wondering if you (or anyone else) has any ideas on how to restrict my exception to something more sensible? The interesting thing about all this is that without that global cloudfront.net exception in the list, a NoScript flag informing me that something is being blocked isn't showing (which is why I couldn't fix it in the normal way; clicking on the NoScript icon and electing to temporarily or permanently permit the blocked foo.cloudfront.net site). It's also mildly interesting in that this is the first instance of there being an issue like this (with NoScript) and I've been running with NoScript active for many, many years.

    I'm delighted that I'm now back in action (though I suspect many others here will be dismayed; I can almost hear the collective thought 'Oh no, tedious Briain has returned!') but it would be most interesting to know why this site is doing something so different (to all others that I've ever visited) that NoScript couldn't identify there as being a problem, and of course, also how to narrow down my exception (which I suspect might not even be possible); any tips from anyone would be gratefully received.

    Kind regards to all,

    Briain :-)