Search Support

Avoid support scams. We will never ask you to call or text a phone number or share personal information. Please report suspicious activity using the “Report Abuse” option.

Learn More

Back-button stops working

  • 22 replies
  • 33 have this problem
  • 239 views
  • Last reply by hallvord

more options

Hello experts.

I have a curious problem where the back-button stops working - and other threads regarding this issue has proven unsuccessful in my case, like changing the browser.backspace_action value, resetting Firefox and disable hardware acceleration.

Alright, try this link: http://htmlticker.nmd.dk/udenadform/index.php?page=1 When clicking the number buttons, the page changes - and when clicking the back-button, it returns to the previous pages. So far, so good.

Now, try this: http://htmlticker.nmd.dk/medadform/index.php?page=1 Try clicking some numbers, for instance 1 to 5 and click back. Nothing happens, right? The page reloads, but the browser doesn't return to the previous page.

The difference between the two pages are that the upper link's sourcecode points directly to the ticker HTML, while the lower link's sourcecode points through Adform advertisement engine - but still, it should rewind as usual.

Any bright ideas are much appreciated - at least a reason, so I can inform our customer.

Hello experts. I have a curious problem where the back-button stops working - and other threads regarding this issue has proven unsuccessful in my case, like changing the browser.backspace_action value, resetting Firefox and disable hardware acceleration. Alright, try this link: http://htmlticker.nmd.dk/udenadform/index.php?page=1 When clicking the number buttons, the page changes - and when clicking the back-button, it returns to the previous pages. So far, so good. Now, try this: http://htmlticker.nmd.dk/medadform/index.php?page=1 Try clicking some numbers, for instance 1 to 5 and click back. Nothing happens, right? The page reloads, but the browser doesn't return to the previous page. The difference between the two pages are that the upper link's sourcecode points directly to the ticker HTML, while the lower link's sourcecode points through Adform advertisement engine - but still, it should rewind as usual. Any bright ideas are much appreciated - at least a reason, so I can inform our customer.

Chosen solution

Hi, to understand this problem it's important to keep in mind what the Back button is supposed to do (even though it may seem obvious :-)). Back essentially means "revert the last navigation action that took place on this page". On pages that use frames/iframes, the browser must take care to ensure Back only affects the frame that was navigated most recently - or it will in most cases fail to work as the user expects.

What I believe (based on similar cases I've analysed) is happening here, is that some JavaScript on this page navigates an IFRAME in a way Firefox registers as a "reversible" navigation. A single click on the back button will thus reverse the navigation in the (possibly invisible) IFRAME and the user will believe nothing happened. Double-clicking the back button will often successfully go back when you are in this state.

If it works in other browsers, it means Firefox is using slightly different logic to register what navigations are "reversible" - and I would personally consider that a bug in Firefox, though it's entirely possible that we would break back functionality on some other site trying to fix this. -Hallvord

Read this answer in context 👍 1

All Replies (2)

more options

Hello Hallvord.

Thank you for your reply.

To everybody having this problem; Yesterday, one of our clients found a way around, which suits our situation: http://www.bold.dk/test/backbuttontest.php?page=4&version=0 - not working http://www.bold.dk/test/backbuttontest.php?page=4&version=4 working

If you try to view the source code, you'll see that it's solved by jQuery.

All the best.

more options

Wow.. that's a sort of ugly-but-clever hack. Congrats to whoever came up with it!

Now, it shouldn't be necessary to start with, and I'll try to keep pushing for a bug fix here ;-)

  1. 1
  2. 2