User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:33.0) Gecko/20100101 Firefox/33.0 Build ID: 20141106120505 Steps to reproduce: Please watch the short video. Easier than trying to explain the bug. Please note that I was pressing F5 multiple times to refresh the pages: https://dl.dropboxusercontent.com/u/63231818/firefox_bug.mp4 Actual results: When the Internet connection is slow, sometimes while a page is still loading, I'll press F5 to reload it, and Firefox forgets the URL and shows "New Tab" on the tab that was just refreshes. Also occurs sometimes when you close Firefox and re-open it, tabs that once had URLs no longer do. Expected results: The URL should persist in the address bar if refreshed or not. It's very frustrating to close Firefox and most of the URLs don't load or when you refresh a page because the page is not loading, then the URL disappears and you have no idea what the URL was.

Tim, can we use session store and/or (its) tab attributes in some way to make this situation work, ie have a "desired" URL rather than the browser's currentURI? I expect the same issue is plaguing the session restore situation described in comment #0 - if we close while the tab is still on about:blank, we presumably don't store the fact that it was really trying to load foo.com instead...

(In reply to :Gijs Kruitbosch from comment #1) > Tim, can we use session store and/or (its) tab attributes in some way to > make this situation work, ie have a "desired" URL rather than the browser's > currentURI? I expect the same issue is plaguing the session restore > situation described in comment #0 - if we close while the tab is still on > about:blank, we presumably don't store the fact that it was really trying to > load foo.com instead... IIRC we actually use browser.userTypedValue and .userTypedClear to support the latter case described here. Maybe we should do the same when refreshing a blank page is attempted? Sounds like something we'd handle in browser.js then?

(In reply to Tim Taubert [:ttaubert] from comment #2) > (In reply to :Gijs Kruitbosch from comment #1) > > Tim, can we use session store and/or (its) tab attributes in some way to > > make this situation work, ie have a "desired" URL rather than the browser's > > currentURI? I expect the same issue is plaguing the session restore > > situation described in comment #0 - if we close while the tab is still on > > about:blank, we presumably don't store the fact that it was really trying to > > load foo.com instead... > > IIRC we actually use browser.userTypedValue and .userTypedClear to support > the latter case described here. Maybe we should do the same when refreshing > a blank page is attempted? Sounds like something we'd handle in browser.js > then? Yeah, sounds like it. Let's backlog this, at least.

this happens to me too. also because my internet is slow then when I am downloading and browsing and then open a tab (say from a free reader website) firefox doesn't load URL in address bar and in the bottom it says "connecting to "+website-address .but when it cant (which with my bad connection can happen many times) it loads an empty page with no address so I cant even refresh it or remember what website is was. This make firefox inferior to even IE. why isn't this assigned already? Isn't there a workaround so I can use that in the meantime until a fix is provided?

(In reply to rezad from comment #5) > this happens to me too. > also because my internet is slow then when I am downloading and browsing and > then open a tab (say from a free reader website) firefox doesn't load URL in > address bar and in the bottom it says "connecting to "+website-address .but > when it cant (which with my bad connection can happen many times) it loads > an empty page with no address so I cant even refresh it or remember what > website is was. > > This make firefox inferior to even IE. > > why isn't this assigned already? Because this isn't a show-stopping bug that breaks a lot of things for a lot of people, and because it isn't a regression compared to earlier versions of Firefox, and because when looking at some of the other (tens of thousands of) bugs in our bugtracker it's hard/impossible to keep track of and prioritize things outside of the main projects teams are working on, considering the sheer volume of bugs reported. > Isn't there a workaround so I can use that in the meantime until a fix is > provided? I've not tried this, but I wonder if closing the tab followed by "undo close tab" would fix this. It may be helpful to know if that does/doesn't... help.

first let me say this is IMHO somehow worse that showstopper bugs. I closed too many blank tabs in my more that 40 open tab that were open at the same time , thinking that they were for file download empty open tabs or tabs I open with new tab command and forgot about them. this makes it so that I have to double check any link I open to see if they are correctly opened and not lost forever then close its parent tab ( or my feed reader in this case) . I will check your suggestion if I remember when it happens for me. btw , I live in Iran and this also happens sometimes when I try to open websites that the government had filtered at IP level( they are usually https sites and they don't even give a BLOCK page that is shown for http filtered websites , like Facebook). but I don't think this error is confined to just this case.

I tried opening a like to facebook without use of vpn and it would be stuck in loading phase. then I closed the tab and tried to undo cloce tab and it didn't work. (it opened the tab that was closed before the facebook tab)