Monitoring HTTP redirects from Firefox
Does anyone know of a way from within Firefox that I can capture a sequence of redirects and work out what's happening? I can use wget and see a sequence of 301/302 redirects that way, but I need to do this within a session so I really need to follow it in Firefox. Or Opera/IE I suppose (all on Ubuntu 8.04). Something obscure is happening on one of my sites when I follow an HTTPS link which is causing it to redirect back to HTTP, which is probably something to do with something in the session that I'm not getting chance to investigate. Therefore the ability not just to capture the sequence but also to step through it would be helpful. -- Mark Rogers // More Solutions Ltd (Peterborough Office) // 0845 45 89 555 Registered in England (0456 0902) at 13 Clarke Rd, Milton Keynes, MK1 1LG
Hi Mark,
Does anyone know of a way from within Firefox that I can capture a sequence of redirects and work out what's happening?
I've used this in the past to diagnose a similar sort of problem. It seems to work. http://livehttpheaders.mozdev.org/ Good luck, Rob.
Rob Page wrote:
I've used this in the past to diagnose a similar sort of problem. It seems to work.
I installed this last night, but although it showed me the HTTP conversation for the current page I couldn't see how to get the details of the trail of redirects that got me there. That said I didn't try very hard, and I can't check right now as I'm on site checking my email via VPN, so it may be that I'm missing something obvious. Thanks for the pointer, I'll investigate further when back at my desktop. -- Mark Rogers // More Solutions Ltd (Peterborough Office) // 0845 45 89 555 Registered in England (0456 0902) at 13 Clarke Rd, Milton Keynes, MK1 1LG
Mark Rogers <mark@quarella.co.uk> wrote: [...]
I can use wget and see a sequence of 301/302 redirects that way, but I need to do this within a session so I really need to follow it in Firefox. [...]
Can't you use wget's cookie-saving options to follow it? Regards, -- MJ Ray http://mjr.towers.org.uk/email.html tel:+44-844-4437-237 - Webmaster-developer, statistician, sysadmin, online shop builder, consumer and workers co-operative member http://www.ttllp.co.uk/ - Writing on koha, debian, sat TV, Kewstoke http://mjr.towers.org.uk/
MJ Ray wrote:
Can't you use wget's cookie-saving options to follow it?
Probably yes, but the process needed to get to the right state for the tests (in this case creating an account, logging in, entering delivery details, etc) would mean that I really needed a GUI for wget that allowed me to fill in the right form fields etc. Writing a GUI for wget is probably harder than writing a logging tool for Firefox :-) The problem is that the site should be switching to HTTPS when it comes to finalising an order process, and the relevant steps in the process aren't available until you've created the order. What appears to be happening is that the site is correctly linking to an HTTPS page which is incorrectly then redirecting back to an HTTP version of itself. In reality it's not a single redirect, but a sequence of three redirects none of which I can see any reason for (not my code!). -- Mark Rogers // More Solutions Ltd (Peterborough Office) // 0845 45 89 555 Registered in England (0456 0902) at 13 Clarke Rd, Milton Keynes, MK1 1LG
Just a thought and I know it means switching to a lower level tool but couldn't you just stick wireshark in the mix somewhere (even on the target machine) and then use the filtering and follow stream options to see what is going on ? It should be possible to get an output from wireshark that is simply a transcript of client requests and server responses etc.
On 24/01/2008, Wayne Stallwood <ALUGlist@digimatic.co.uk> wrote:
Just a thought and I know it means switching to a lower level tool but couldn't you just stick wireshark in the mix somewhere (even on the target machine) and then use the filtering and follow stream options to see what is going on ?
It should be possible to get an output from wireshark that is simply a transcript of client requests and server responses etc.
Wasn't he talking about https, which would be encrypted? Peter.
On Thu, 2008-01-24 at 13:36 +0000, samwise wrote:
On 24/01/2008, Wayne Stallwood <ALUGlist@digimatic.co.uk> wrote:
Just a thought and I know it means switching to a lower level tool but couldn't you just stick wireshark in the mix somewhere (even on the target machine) and then use the filtering and follow stream options to see what is going on ?
It should be possible to get an output from wireshark that is simply a transcript of client requests and server responses etc.
Wasn't he talking about https, which would be encrypted?
Peter.
A valid point I may have misunderstood the original question and thought that all the "problems" where happening on http pages.
Wayne Stallwood wrote:
On Thu, 2008-01-24 at 13:36 +0000, samwise wrote:
On 24/01/2008, Wayne Stallwood <ALUGlist@digimatic.co.uk> wrote:
It should be possible to get an output from wireshark that is simply a transcript of client requests and server responses etc.
Wasn't he talking about https, which would be encrypted?
A valid point I may have misunderstood the original question and thought that all the "problems" where happening on http pages.
Peter is right; I am talking about https (or at least they're in the mix). Wireshark is of-course an excellent package which I hadn't thought about on this occasion, though, so thanks for reminding me! -- Mark Rogers // More Solutions Ltd (Peterborough Office) // 0845 45 89 555 Registered in England (0456 0902) at 13 Clarke Rd, Milton Keynes, MK1 1LG
My favourite tool for debugging HTTP requests is Charles Proxy: http://www.xk72.com/charles/ - Runs on Windows, Linux, OS X Gives you a true view of what's going on as it sits between your web browser and the web server, intercepting all of the traffic flowing between them and letting you view every element of the request and the response. If you're using Windows then a free tool called Fiddler does nearly as much: http://www.fiddlertool.com/fiddler/ hth Gareth On Jan 21, 2008, at 10:43 AM, Mark Rogers wrote:
Does anyone know of a way from within Firefox that I can capture a sequence of redirects and work out what's happening?
I can use wget and see a sequence of 301/302 redirects that way, but I need to do this within a session so I really need to follow it in Firefox. Or Opera/IE I suppose (all on Ubuntu 8.04).
Something obscure is happening on one of my sites when I follow an HTTPS link which is causing it to redirect back to HTTP, which is probably something to do with something in the session that I'm not getting chance to investigate. Therefore the ability not just to capture the sequence but also to step through it would be helpful.
-- Mark Rogers // More Solutions Ltd (Peterborough Office) // 0845 45 89 555 Registered in England (0456 0902) at 13 Clarke Rd, Milton Keynes, MK1 1LG
_______________________________________________ main@lists.alug.org.uk http://www.alug.org.uk/ http://lists.alug.org.uk/mailman/listinfo/main Unsubscribe? See message headers or the web site above!
Gareth Watts wrote:
My favourite tool for debugging HTTP requests is Charles Proxy: http://www.xk72.com/charles/ - Runs on Windows, Linux, OS X
Gives you a true view of what's going on as it sits between your web browser and the web server, intercepting all of the traffic flowing between them and letting you view every element of the request and the response.
Thanks for this; I've installed it to play with. It's Java based and doesn't seem to like anything other than Sun's jvm, but I got it working in the end! It overcomplicates things by treating http://www.mysite and https://www.mysite as different sites making it hard to track the sequence between them, which seems odd. It also is showing me 301 (permanent redirect) responses but unless I'm missing something seems to not be telling me where the redirect destination is. My suspicion is that there's something wrong in the response that's not giving a complete answer and therefore getting the browser to "guess", and the browser is guessing wrong. It would just be nice to see the actual response to be sure! Nonetheless a useful tool even if it won't solve this problem. I could have used this hundreds of times in the past. [Later:] I just noticed that in preferences I can tell it not to merge headers with content, and that now shows me the full response headers. -- Mark Rogers // More Solutions Ltd (Peterborough Office) // 0845 45 89 555 Registered in England (0456 0902) at 13 Clarke Rd, Milton Keynes, MK1 1LG
On 21/01/2008, Mark Rogers <mark@quarella.co.uk> wrote:
Does anyone know of a way from within Firefox that I can capture a sequence of redirects and work out what's happening?
I know it's not Linux, but I've found Fiddler brilliant for this in the past - http://www.fiddlertool.com/fiddler/ It acts as a proxy, and logs everything. Also good for working out which bits of a complex page are taking too long to load. Greg
participants (7)
-
Gareth Watts -
Greg Thomas -
Mark Rogers -
MJ Ray -
Rob Page -
samwise -
Wayne Stallwood