No text cursor for login on CitiCard website

[Greg]

Binary package hint: firefox

The cursor disappears on the first login line of the login form of the CitiCard website after a second or two and is unrecoverable. While it is present, it is inactive. This is not a problem with Google Chrome, nor with any other website using Firefox in my experience. It is annoying to have to use two web browsers in order to get my work done and so would appreciate your remedying this issue.

Greg Morgansen

ProblemType: Bug
DistroRelease: Ubuntu 10.04
Package: firefox 3.6.3+nobinonly-0ubuntu4
ProcVersionSignature: Ubuntu 2.6.32-21.32-generic 2.6.32.11+drm33.2
Uname: Linux 2.6.32-21-generic i686
NonfreeKernelModules: nvidia wl
Architecture: i386
Date: Wed May 5 09:57:36 2010
FirefoxPackages:
 firefox 3.6.3+nobinonly-0ubuntu4
 firefox-gnome-support 3.6.3+nobinonly-0ubuntu4
 firefox-branding 3.6.3+nobinonly-0ubuntu4
 abroswer N/A
 abrowser-branding N/A
InstallationMedia: Ubuntu 10.04 LTS "Lucid Lynx" - Release i386 (20100429)
ProcEnviron:
 LANG=en_US.utf8
 SHELL=/bin/bash
SourcePackage: firefox

[Davemgarrett (davemgarrett)]

I get a couple of these in the error console:

Error: obj is null
Source File: https://www.citicards.com/cards/wv/js/homeBott2Min.js
Line: 23

Easy to reproduce under Linux and works fine under Windows XP.

I highly suspect they're doing something stupid in there somewhere and this is either a TE bug or a Flash bug, but I'm not sure. Confirming, nonetheless.

[Matt Lavin (matt-lavin)]

If you have a flash blocker installed you can work around this problem by disabling flash on their main page. I guess the flash content is somehow blocking the login text entry from getting focus.

[Greg (gsmorgansen)]

Binary package hint: firefox

The cursor disappears on the first login line of the login form of the CitiCard website after a second or two and is unrecoverable. While it is present, it is inactive. This is not a problem with Google Chrome, nor with any other website using Firefox in my experience. It is annoying to have to use two web browsers in order to get my work done and so would appreciate your remedying this issue.

Greg Morgansen

ProblemType: Bug
DistroRelease: Ubuntu 10.04
Package: firefox 3.6.3+nobinonly-0ubuntu4
ProcVersionSignature: Ubuntu 2.6.32-21.32-generic 2.6.32.11+drm33.2
Uname: Linux 2.6.32-21-generic i686
NonfreeKernelModules: nvidia wl
Architecture: i386
Date: Wed May 5 09:57:36 2010
FirefoxPackages:
 firefox 3.6.3+nobinonly-0ubuntu4
 firefox-gnome-support 3.6.3+nobinonly-0ubuntu4
 firefox-branding 3.6.3+nobinonly-0ubuntu4
 abroswer N/A
 abrowser-branding N/A
InstallationMedia: Ubuntu 10.04 LTS "Lucid Lynx" - Release i386 (20100429)
ProcEnviron:
 LANG=en_US.utf8
 SHELL=/bin/bash
SourcePackage: firefox

[Leo Arias (elopio)]

could you provide the URL of the website?

[Greg (gsmorgansen)]

The url for CitiCard website is: www.citibank.com/us/cards/index.jsp

Greg Morgansen

[Mozilla-bugs-micahscomputing (mozilla-bugs-micahscomputing)]

If you change the User Agent to IE, it works fine in Firefox. If you try in Fennec 1.0 or Conkeror which is a xul based browser, it also works fine.

[Mozilla-bugs-micahscomputing (mozilla-bugs-micahscomputing)]

*** Bug 399202 has been marked as a duplicate of this bug. ***

[Mozilla-bugs-micahscomputing (mozilla-bugs-micahscomputing)]

Another thing is that the cursor is automatically put in the test field, so if you type your username, hit tab, and type the password, it seems to work. Tested on:
Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.3a5pre) Gecko/20100505 Ubuntu/10.04 (lucid) Minefield/3.7a5pre
and
Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.4) Gecko/20100417 Ubuntu/10.04 (lucid) Firefox/3.6.4 <-- FIREFOX_3_6_4_BUILD3

[Draycen DeCator (ddecator)]

I tested it on my system, and while my cursor does not disappear I can confirm that the site is unresponsive to clicks. I found that this is due to a flash animation that is supposed to play automatically but fails to. Right-clicking and then selecting "Play" caused a Flash-based overlay to appear. After closing that, the site functioned normally. Could you please test whether this is true for you as well? Thanks in advance!

[Greg (gsmorgansen)]

Yes. The process you describe causes the login to function normally.

What is the fix?

Greg Morgansen

[Micah Gersten (micahg)]

Since there's an upstream bug, I'll mark this triaged for now until upstream decides if they want to or can do anything about it. Please report any other issues you may find.

[knobberhead (knobberhead)]

a few more notes on what was originally reported as bug 577276 (which was marked as a duplicate of this bug)...

The virtual account number page can be reached after logging into the citicards site (I didn't have any problem with the cursor disappearing at the initial login screen). Anyway, after logging into the virtual account number page, it is not responsive to clicks. I right clicked on this page and it pops up with the options "Settings" and "About Adobe Flash Player 10". Selecting "Settings" presents a pop up window that is also not clickable. There is no "Play" option. Unlike the original poster, this problem happens with the Firefox, Seamonkey, Chromium, Konqueror and Epiphany browsers that can be installed from the Synaptic Package Manager in Ubuntu 10.04. So, five browsers...none of them working - I guess it is a Flash problem?

[Micah Gersten (micahg)]

@knobberhead

You are most likely also experiencing bug 410407.

On 05/08/2010 11:00 PM, knobberhead wrote:
> a few more notes on what was originally reported as bug 577276 (which
> was marked as a duplicate of this bug)...
>
> The virtual account number page can be reached after logging into the
> citicards site (I didn't have any problem with the cursor disappearing
> at the initial login screen). Anyway, after logging into the virtual
> account number page, it is not responsive to clicks. I right clicked on
> this page and it pops up with the options "Settings" and "About Adobe
> Flash Player 10". Selecting "Settings" presents a pop up window that is
> also not clickable. There is no "Play" option. Unlike the original
> poster, this problem happens with the Firefox, Seamonkey, Chromium,
> Konqueror and Epiphany browsers that can be installed from the Synaptic
> Package Manager in Ubuntu 10.04. So, five browsers...none of them
> working - I guess it is a Flash problem?
>

[Bugzilla-tf (bugzilla-tf)]

Is this still an issue ?
The site seems to be changed..

[Parag Joshi (pmj005)]

Just tested and it seems to work now.

[Bugzilla-tf (bugzilla-tf)]

thank your for the response

[Ralph Navarro (ralph-navarrocomputing)]

Two and a half years later and now I can use Firefox 18.0.2 running on Ubuntu 12.04.2 LTS to login to CitiCards!

This bug had prevented anyone from using Linux Firefox to access a large financial institution's site. This bug should have been given a higher priority by Mozilla at the time I opened it. The lack of response from Mozilla had caused me to give up on using Firefox entirely.

I guess the only positive thing about this bug is that Mozilla kept it on the books. Maybe I'll try using Firefox again. Hopefully, Mozilla will now consider compatibility issues with major financial institutions' sites important enough to fix for users.

[Ralph Navarro (ralph-navarrocomputing)]

Oops. Make that three and a half years.

[Bugzilla-tf (bugzilla-tf)]

(In reply to Ralph Navarro from comment #9)
 > I guess the only positive thing about this bug is that Mozilla kept it on
> the books. Maybe I'll try using Firefox again. Hopefully, Mozilla will now
> consider compatibility issues with major financial institutions' sites
> important enough to fix for users.

This was a bug in the page according to comment#3
We want to be compatible to the web standards as every other browser vendor but we are not doing something to workaround bugs in pages.

[Ralph Navarro (ralph-navarrocomputing)]

Did you notice that the problem worked fine in other Linux browsers but not in Firefox? What does that do for Firefox's market share? The current web standards have flaws which allow for interpretation. Isn't it also important to make a browser compatible to the de facto standards as interpreted by most other browsers?

Your attitude is what drew me away from Firefox. I was almost willing to give Firefox another try. However, if Mozilla has attitudes like yours, I don't want to waste my time.

Ralph

[Bugzilla-tf (bugzilla-tf)]

>Did you notice that the problem worked fine in other Linux browsers but not in Firefox?

comment#3 says that just changing the user Agent makes it work.
In case that you don't know what the UA is i will post my current UA:
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:21.0) Gecko/20100101 Firefox/21.0

It's just a name that every browser sends to the webserver. If the behavior of the page changes with a UA change this is a 100% bug in the page that is detecting the user Agent string and does something different depending on the UA string. That means that you can use your "other browser" and change the User Agent of this browser and it should also fail to work.

>The current web standards have flaws which allow for interpretation
Such flaws will be discussed in the standard working groups if such a flaw is found. The standard will be changed based on that discussion.
You can't simple change the browser to make it work because other pages will break with this change.

>Isn't it also important to make a browser compatible to the de facto standards as interpreted by most other browsers?
No, that would bring as back to the IE6 days with a complete broken web.

>Your attitude is what drew me away from Firefox.
I'm just a user like you but I try to tell people the truth.
It seems that most people want to hear the usual marketing speak, full of lies :-(

[Ralph Navarro (ralph-navarrocomputing)]

> The standard will be changed based on that discussion.
Poppycock! While, yes, the standard can get changed, the timing becomes too slow to be practical. The standard won't get changed for years after the initial conversation and after many discussions/revisions. Meanwhile, solutions have to be created and implemented by developers. The new standard looses its effectiveness until major browsers catch up.

Comment #3 was posted 9 months afterwards. It has been so long, I don't remember if I tried Comment #3's suggestion or not. What I do remember is the frustration that I had with Firefox on Linux while this 'bug' on the site's page was able to work with Firefox on WinXP.

Why is Mozilla treating Linux with a lower priority than MS? Shouldn't us users be able to expect that browsers have the same behavior across platforms? When the browser behaviors start to drift, is it unreasonable to expect that the browser with the worst user experience be the one to get fixed? We all know that MS Windows is a different enough beast than Linux which might make common behavior hard to achieve; but what is the harm in trying?

[Bugzilla-tf (bugzilla-tf)]

(In reply to Ralph Navarro from comment #14)
> > The standard will be changed based on that discussion.
> Poppycock! While, yes, the standard can get changed, the timing becomes too
> slow to be practical. The standard won't get changed for years after the
> initial conversation and after many discussions/revisions. Meanwhile,
> solutions have to be created and implemented by developers. The new
> standard looses its effectiveness until major browsers catch up.

That is just wrong. Asking it the standards Working Group for something that isn't clear in the standards is pretty fast. No single browser vendor will change their browser to follow a different browser if they think they are correctly following the standard and it makes sense to do it in that way.
What I don't understand is that your website screwed up and you are their customer.
I would just move to a different bank. I did that once 8 years ago with an IE only bank webpage. Why should they fix their broken page if the don't have to ?

> Comment #3 was posted 9 months afterwards. It has been so long, I don't
> remember if I tried Comment #3's suggestion or not. What I do remember is
> the frustration that I had with Firefox on Linux while this 'bug' on the
> site's page was able to work with Firefox on WinXP.

It can't be a Firefox bug if it works on Windows and doesn't work on Linux.
There is no difference in the html/JS/imagelib etc. between the platforms !
The main differences are only in the graphic output and system integration like the default browser, Themes but that doesn't affect the content of webpages.

> Why is Mozilla treating Linux with a lower priority than MS? Shouldn't us
> users be able to expect that browsers have the same behavior across
> platforms? When the browser behaviors start to drift, is it unreasonable to
> expect that the browser with the worst user experience be the one to get
> fixed? We all know that MS Windows is a different enough beast than Linux
> which might make common behavior hard to achieve; but what is the harm in
> trying?

Again, the whole rendering code is cross-platform and the same source code is used on windows, linux, OS/2, BSD, AIX
The only difference in your case is that Firefox tells the webpage that it's a Firefox browser that runs on linux and not on Windows.

[Davemgarrett (davemgarrett)]

The problem here, which happens on sites from time to time, is that they use lazy poorly written UA sniffing to check if they support a browser. I've seen plenty of sites, frequently banks, that have a list of UAs they support and have their site automatically say they don't support any others. Their whitelist gets the working site and everyone else gets an error message or broken code. It breaks for Linux not because of anything having to do with Firefox or Linux, but because in their lazy world Linux doesn't even exist. It looks for the versions of Firefox it supports on Windows and Mac, then assumes the rest are incompatible. It's irritating, but it happens (though less frequently these days, I think). In these instances it's not that Mozilla is treating Linux as anything other than a Tier 1 platform, it's the site doing it.

Aside from hacking around it with UA spoofing, as mentioned above, the only recourse for things like this is to tell the site to fix it. After I confirmed this bug it needed a little more investigation, and one of use probably should've done so and moved this to the Technical Evangelism component to have someone ask them to fix it, though sadly, that often goes nowhere too. Sorry. Glad it's fixed now, though.

[Davemgarrett (davemgarrett)]

That being said, comment 0 did say people had been complaining about it since 2003, so sadly they just appeared to not care to fix it due to Linux's small marked share.

[Ralph Navarro (ralph-navarrocomputing)]

Thank you for taking the time to explain. I better understand what happened. However, we are mostly guessing here as to what the real problem was. If I run into a similar problem again, I will do a diff of the pages received by both browsers to see if the web site is actually doing something different.

Ultimately, though I suspect you are both right about this problem. The financial institution should be supporting Linux better. If I run into this again, I will try to escalate the problem with the bank so it gets fixed. Otherwise, I will move to another bank.

Sorry for the rant earlier.

[Paul White (paulw2u)]

We are sorry that we do not always have the capacity to review all reported bugs in a timely manner. You reported this bug some time ago and there have been many changes in Ubuntu since that time.

Presumably this problem was a one time occurrence and/or is no longer an issue? Can the bug report now be closed? If we do not hear from you the bug report will close itself in approximately 60 days time.

Thank you again for helping make Ubuntu better.

Paul White
[Ubuntu Bug Squad]

[Paul White (paulw2u)]

No reply from reporter but upstream report closed "RESOLVED WORKSFORME".
Site appears to have changed since report raised so Firefox now works. Closing by marking "Invalid" as no action appears to have been taken by Mozilla developers