280143 - citibank.com.sg - unable to log in, applet loading error

Status: RESOLVED WORKSFORME

(Plugins Graveyard :: Java (Java Embedding Plugin), defect)

Description

[Richard]

User-Agent:       Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.7.5) Gecko/20041107 Firefox/1.0
Build Identifier: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.7.5) Gecko/20041107 Firefox/1.0

"An applet required to access you bank online services did not load correctly in
your browser..."

I followed the instructions to enable Java and Java Script. Does not work. 

Reproducible: Always

Steps to Reproduce:
1. Login 
2. Click ok as per message
3. Follow instructions to enable Java.

Actual Results:  
Same thing

Expected Results:  
Logged into my account.

Comment 1

[Daniel Veditz [:dveditz]]

No evidence of a security issue, clearing flag.

I'm surprised they'd rely on Java since many people don't have it. Without an
account it's hard to say, but my initial guess is that they're using some
feature not supported by the Apple JVM, or worse, only supported by Microsoft's
JVM. What version of Java do you have? What version does Citibank require?

If you have access to a windows machine does it work from Firefox there?

If it's a Mac JVM issue Tech Evangelism may not be appropriate, but if there's
no newer JVM from apple then it's still a site problem.

Comment 2

[Felix Miata]

Conforming summary to TFM item 10 at 
http://www.mozilla.org/projects/tech-evangelism/site/procedures.html#file-new

Comment 3

[Richard]

I have tried installing again the latest Java:

java version "1.4.2_05"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.4.2_05-141.3)
Java HotSpot(TM) Client VM (build 1.4.2-38, mixed mode)

I don't know if it is a jave issue. I tried Firefox on a PC and it can't load
the page either. I have not tried re-installing java 1.4.2. 

Am wondering if this a Firefox bug. 

Comment 4

[Richard]

I have tried the Mac browser, Safari, after re-installing Java and it works!
However, Firefox still can't log into banking websites for online transactions.
The same error message appears.

The same thing happened on a PC I tried with the new Java VM. 

Suspect it is a Firefox bug.

Comment 5

[Hiro]

Java1.4.x cannot be used by the standard yet.
See bug197813

Comment 6

[Steven Michaud [:smichaud] (Retired)]

As best I can tell without an account, this URL works fine with the
Java Embedding Plugin:

http://javaplugin.sourceforge.net/

By the way, the reason this site doesn't work with the default plugin
(Apple's "Java Applet.plugin") is that this plugin doesn't support
LiveConnect.  The Java Embedding Plugin does support it (via its "MRJ
Plugin JEP").

Comment 7

[Hiro]

The problem of live connect is bug232652.
If it is possible to use it with MacIE, it is likely to be able to use it also
with Mozilla by fix of this bug. 

If it is N.G, the Java1.4 support is also necessary.

Comment 8

[Steven Michaud [:smichaud] (Retired)]

Interesting!

I guess you're saying that if this URL works in Mac IE (which doesn't
support LiveConnect), then this problem can't be a LiveConnect
problem.

It _does_ work in Mac IE (as far as I can tell without having an
account).  And I've confirmed that it _doesn't_ work in any of the
Mozilla-family of browsers without the Java Embedding Plugin (or
specifically the MRJ Plugin JEP that comes with it).

When the Java implementation is able to do LiveConnect (as the JEP
can), citibank.sg does use JavaScript-to-Java LiveConnect.  But
LiveConnect doesn't seem to be required, since Mac IE also works
(which uses Apple's "Java Applet.plugin" that doesn't support
LiveConnect).

Why does the URL work with Mac IE?  I don't know, but I suspect that
citibank.sg is behaving differently according to which agent string it
sees.  I don't know how to test this.  Can someone else give it a try?

In any case, though, the easiest way to get this site to work on the
Mozilla-family is to install the JEP.

Comment 9

[Steven Michaud [:smichaud] (Retired)]

> I don't know, but I suspect that citibank.sg is behaving differently
> according to which agent string it sees.  I don't know how to test
> this.  Can someone else give it a try?

I've now tried this, and it didn't make any difference -- even when
you set Firefox's (1.0) or Mozilla's (1.7.5) user-agent string to
match Mac IE's, this URL still fails with an error message about an
applet not being loaded when you're not using the Java Embedding
Plugin (i.e. when you're using Apple's "Java Applet.plugin").

Puzzling.

(By the way, to make Firefox or Mozilla use a different user-agent
string, use about:config to create the string value
"general.useragent.override" and set it to whatever you want.  The
string for Mac IE is "Mozilla/4.0 (compatible; MSIE 5.22;
Mac_PowerPC)".  To restore the original user-agent string, right-click
(or control-click) on "general.useragent.override" and choose
"reset".)

Comment 10

[Richard]

I followed Steven's suggestion and downloaded the java pluging at 

http://javaplugin.sourceforge.net/

and followed the instructions carefully. Now I am able to log into citibank with
Firefox! 

I am now wondering shouldn't this plugin a default or embedded into Firefox? 

Thanks, Steven! :-)

Comment 11

[Steven Michaud [:smichaud] (Retired)]

> Now I am able to log into citibank with Firefox!

Glad to hear it.

> I am now wondering shouldn't this plugin a default or embedded into
> Firefox?

When the Java Embedding Plugin is ready, I'd like to see a snapshot
included in the Mozilla/Firefox/Camino distros.  (I want the JEP to
retain its separate existence and license.  But the JEP's license is
compatible with the MPL, and Mozilla.org could treat it the same way
they do libjpeg, which also has a BSD-style license.)

But though it's close, I don't think the JEP is quite ready yet.  And
because the JEP is technically very ambitious (and tricky to get
right), it's probably best to wait longer for its first non-beta
release than if it were an ordinary program.

There's been _lots_ of discussion of this at bug 197813.  For my side
of the argument, see particularly comments 191, 208, 213 and 218.

> Thanks, Steven! :-)

You're most welcome.

Comment 12

[Richard]

Hi Steven,

I very much appreciate the help that you had given and good people like yourself. 

As a final follow through (constructive feedback) I would only like to add that
though the solution works for me it does take almost a minute before the applet
is fully ready for me to log into the web site (access internet banking). Below
is the message from the java console for your reference.

Once again thanks for your dedication. I hope my feedback will aid as a support
to your efforts.

**************************
Java(TM) Plug-in: Version 1.4.2_05
Using JRE version 1.4.2_05 Java HotSpot(TM) Client VM
User home directory = /Users/XXX
Proxy Configuration: Manual Configuration
     Proxy:
http=proxy.XXXX.com:8080,https=proxy.XXXX.com.:8080,ftp=proxy.XXXX.com:8080,gopher=proxy.XXXXX.com:8080
     Proxy Overrides: 

----------------------------------------------------
c:   clear console window
f:   finalize objects on finalization queue
g:   garbage collect
h:   display this help message
l:   dump classloader list
m:   print memory usage
o:   trigger logging
p:   reload proxy configuration
q:   hide console
r:   reload policy configuration
s:   dump system properties
t:   dump thread list
v:   dump thread stack
x:   clear classloader cache
0-5: set trace level to <n>
----------------------------------------------------
MRJ Plugin for Mac OS X v1.0.1
[starting up Java Applet Security @ Sat Feb 12 23:28:53 SGT 2005]

Java(TM) Plug-in: Version 1.4.2_05
Using JRE version 1.4.2_05 Java HotSpot(TM) Client VM
User home directory = /Users/XXXX
----------------------------------------------------
c:   clear console window
f:   finalize objects on finalization queue
g:   garbage collect
h:   display this help message
l:   dump classloader list
m:   print memory usage
o:   trigger logging
p:   reload proxy configuration
q:   hide console
r:   reload policy configuration
s:   dump system properties
t:   dump thread list
v:   dump thread stack
x:   clear classloader cache
0-5: set trace level to <n>
----------------------------------------------------
<<< ProxyClassLoader: defined LiveConnectProxy class. >>>
<<< Here're the permisssions you've got: >>>
<<< java.security.Permissions@b29c9d (
 (java.net.SocketPermission www.citibank.com.sg connect,accept,resolve)
)
 >>>
Stopping applet ...
Joining applet thread ...
Destroying applet ...
Disposing applet ...
Quiting applet ...
Joined applet thread ...
Finding information ...
Releasing classloader: sun.plugin.ClassLoaderInfo@ea7549, refcount=0
Caching classloader: sun.plugin.ClassLoaderInfo@ea7549
Current classloader cache size: 1
Done ...

CONSOLE LOG
2005-02-12 23:23:33.486 ScreenSaverEngine[865] ObscureCursor

/Applications/Firefox.app/Contents/MacOS/firefox-bin: can't map file:
/Library/Internet Plug-Ins/MRJPlugin.plugin ((os/kern) invalid argument)

### MRJPlugin:  getPluginBundle() here. ###

### MRJPlugin:  CFBundleGetBundleWithIdentifier() succeeded. ###

### MRJPlugin:  CFURLGetFSRef() succeeded. ###

2005-02-12 23:29:49.542 firefox-bin[869] JEP creating applet eeapplet
******************************************

Comment 13

[Steven Michaud [:smichaud] (Retired)]

> As a final follow through (constructive feedback) I would only like
> to add that though the solution works for me it does take almost a
> minute before the applet is fully ready for me to log into the web
> site (access internet banking). Below is the message from the java
> console for your reference.

I've experienced long delays (10-15 seconds, though not a minute)
every time I connect to this site, no matter which browser I use or
from which platform (Firefox and Safari on Mac OS X 10.3.8 with Java
1.4.2 Update 2 and JEP 0.9.0; Firefox, Mozilla and Konqueror on SuSE
Linux with Sun's Java 1.4.2).  I suspect that it's just that the
Internet pipe is rather narrow between where I am (Chicago, US) and
citibank.sg.

Do you get the same results with different browsers and on different
platforms?

> Stopping applet ...
> Joining applet thread ...
> Destroying applet ...
> Disposing applet ...
> Quiting applet ...
> Joined applet thread ...
> Finding information ...
> Releasing classloader: sun.plugin.ClassLoaderInfo@ea7549, refcount=0
> Caching classloader: sun.plugin.ClassLoaderInfo@ea7549
> Current classloader cache size: 1
> Done ...

This I don't see.  I suspect it only happens after you've successfully
authenticated (which I can't do, since I don't have an account).

> Once again thanks for your dedication. I hope my feedback will aid
> as a support to your efforts.

You're most welcome.  Thanks for your bug report.  As it happens, I
don't think it's a problem with the JEP ... but it never hurts to ask
:-)

Comment 14

[Richard]

Hi Steven,

I am not certain if this new issue I have is related to the script provided
earlier. 

I have noticed that since then I am unable to launch javascripts typically after
the first load. Subsequent javascripts refuse to launch. Any clues? 

p.s. Have enable java and javascripts.

Comment 15

[Steven Michaud [:smichaud] (Retired)]

This doesn't give me enough information to work with.  But (just a
guess) your problem might be related to the following:

http://sourceforge.net/tracker/index.php?func=detail&aid=1086807&group_id=107955&atid=649116

If you can find a way to consistently reproduce the problem, please
open a bug report at:

http://sourceforge.net/tracker/?atid=649116&group_id=107955&func=browse

Comment 16

[Richard]

Hi Steven,

You are right! Same problem. I tested the jepcallback script and the problem is
exactly as described. Guess I should log the bug at sourceforge.net.

Thanks,
richard

Comment 17

[Chris Lawson (gone)]

Steven, it looks to me like this was never really a TE issue in the first place, but rather a problem in the JEP and/or LiveConnect.

The original URL is 404 for me, but there's some sort of Java funkiness going on at

https://www.citibank.com.sg/SGGCB/JSO/signon/DisplayCinSignon.do

If I enter garbage login data, I get a JS alert that says:

"An applet required to access Citibank Online services did not load correctly in your browser. Click the OK button to find out how to set your browser settings to access Citibank Online."

Clicking OK doesn't do anything, of course :-p

I don't have an account there, but perhaps someone who does can verify whether this is still broken if you have a proper login.

Comment 18

[Chris Lawson (gone)]

Oh, it just occurred to me that bug 421855 might have broken this again, but my Camino trunk that I'm testing with should be new enough (2008042901) that it's not a problem. Richard, it would still be helpful if you could test this again with either Firefox 2, Camino 1.6, or a Firefox 3 beta.

Comment 19

[Steven Michaud [:smichaud] (Retired)]

Chris, please test the URL in Safari, and in Firefox on other platforms.

And even if there is still some kind of problem, it's almost certainly unrelated to the original reports ... which are now (of course) very old.

Comment 20

[Chris Lawson (gone)]

I don't have the ability to test in Firefox on any other platforms besides Mac OS, but Safari 3 doesn't give me the alert. It gives me the "invalid login" page, as expected.

I tried it again in Camino trunk, and this time clicking OK in the alert got me this page:

https://www.citibank.com.sg/SGGCB/JSO/signon/AppletError.do

which seems to indicate that it ought to be working.

Firefox 2 works just as Safari did, so there seems to be a trunk bug here. I'll check Minefield in a bit.

Comment 21

[Chris Lawson (gone)]

Yeah, Minefield is equally broken. I'm going to kick this to Core:JEP.

Comment 22

[Chris Lawson (gone)]

Console has

01/05/2008 21:56:50 [0x0-0xb89b89].org.mozilla.firefox[81560] Thu May 01 21:56:50 EDT 2008 JEP creating applet JEapplet (https://www.citibank.com.sg/JCP/encapplet/)

after testing with Minefield, but that's not any help to me, because it showed

01/05/2008 21:47:49 [0x0-0x71f71f].org.mozilla.firefox Thu May 01 21:47:49 EDT 2008 JEP creating applet JEapplet (https://www.citibank.com.sg/JCP/encapplet/) 

after testing with Firefox 2.0.0.14.

cl

Comment 23

[Steven Michaud [:smichaud] (Retired)]

If this happens only on trunk, it can't be a JEP bug.

Comment 24

[Chris Lawson (gone)]

I'm going to close this bug WORKSFORME, since it doesn't seem the original TE issue -- if indeed there ever was one -- still exists. I've filed bug 431869 on comment 17 et seq.