169707 - Mozilla crashes in MRJ plug-in using Oracle JSP application

Status: RESOLVED INCOMPLETE

(Core Graveyard :: Java: OJI, defect)

Description

[Glenn Baxter]

Mozilla 1.1 for MacOS 9.x
Machine ID:604

My company uses Oracle in large measure for many data base items.  The web
interfaces to this use various JSPs.  In all cases, form based execution of
this cause fatal errors to the program.  It dies with a type 2 error.  

Simple example:  We use oracle for our employee look up.  The main page shows
up, but when you click on any of the names, the program dies, and you end up
having to reboot your computer (for safety, not a requirement).  The program
dials home to inform of  the bug, but insufficient information would be present
to identify the bug.  It appears that many Java things do not work right on this
version of Mozilla.

I wish I could offer you the actual JSPs, but that is impossible.  Sorry for no
test cases.

Comment 1

[Phil Schwartau]

This is not going to be a bug in JavaScript Engine. It sounds like
a problem with Java in Oracle apps. That reminds me of bug 164815,
"Oracle Jinitiator is not working in Mozilla 1.1"

Note that bug is assigned to the Plug-ins component instead of OJI,
because Oracle Jinitiator is a modified version of the Java plugin
1.1.7, required to run Oracle Financials.

I don't know if the current bug involves the Jinitiator plug-in
or the Java plug-in, or some other issue.

Glenn:

1. Does anything in bug 164815 sound similar to the current bug?
   I'm guessing "No", since that bug was filed on Windows, and
   this one was filed on Mac. But just in case, notice the test
   at http://bugzilla.mozilla.org/show_bug.cgi?id=164815#c6


2. Are you able to see any error messages in either of these consoles:

           Tools > Web Development > Java Console
           Tools > Web Development > JavaScript Console

     (be sure to clear them of any previous errors first)


3. Do you experience this bug on other machines besides your Mac 9?
   Or is this a Mac-specific bug?


In the meantime, reassigning to Browser-General; cc'ing self and
also bclary, in case he has heard of this issue -

Comment 2

[Greg K.]

Glenn, do you have TalkBack enabled? Does it ever run after these crashes?

Comment 3

[Glenn Baxter]

Hi Greg 
TalkBalk is enabled, has been sent in.  Mozilla dies with a type-2 error, and
"gracefully" quits with the save work and restart standard message.  Regardless
of if I reboot or not, Mozilla will die in the exact same way every time.  
The Java Console doesn't help as the crash is too sudden.  I do have MacsBug, and 
might be able to find more data if you are interested.  Currently, I have MacsBug
disabled.  Let me know if I can provide you or Phil with more data.  I did email
Phil yesterday on this privately.  Cheers, G

Comment 4

[Phil Schwartau]

Here is the info Glenn provided me yesterday (thanks!):

------------------------------------------------------------------------
This does not appear to be related to 164815 as there is no Jinit
for Oracle for Mac that I am aware of.  This is not supposed to
require such either.  I am a manager and use our HR side of oracle,
which does require the Jinit to work at all, and that only works on my PC.
(I'm still using netscape 4.5 on PC, as it is the only one that is stable).
My Mac was using Netscape 4.7, but it was dying 5-6 times a day and
was thus unusable.


To anwer your specific questions:

1.  No this bug is not similar

2.  The machine kaks so there is no way to get to the Java Console.

3.  This problem only appears to occur on Macs that I am aware of.
    However, there are other intranet sites that seem to have similar
    kinds of problems on both platforms. One of my employees has seen
    GUI errors (not crashes) when accessing some particular Java applet.

Unfortunately, there is little more I can tell you, as there is no way to
gain access to  the dump.  Is there some reference material I can give
you from the "dial home" bug function?
------------------------------------------------------------------------


I will try to look up Glenn's Talkback reports now - 

Comment 5

[Phil Schwartau]

Attachment: Talkback stack traces (2)

Comment 6

[Phil Schwartau]

The stack traces both show crashes in the MRJ plug-in.
Reassigning to Patrick and OJI component -

Comment 7

[saari (gone)]

->bnesse, beard is moving away from Java ownership

Comment 8

[James Nurthen]

Have you tried reproducing this issue with IE 5.1? It would really help to know
whether it is an MRJ issue or a Mozilla issue. If it does reproduce with IE and
MRJ then file a TAR with Oracle.

Comment 9

[Frankie]

Does Oracle work with Mozilla Mach-O?

Mozilla CFM build is dead.

Comment 10

[Simon Paquet [:sipaq]]

This bug is targeted at a Mac classic platform/OS, which is no longer supported
by mozilla.org. Please re-target it to another platform/OS if this bug applies
there as well or resolve this bug.

I will resolve this bug as WONTFIX in four weeks if no action has been taken.
To filter this and similar messages out, please filter for "mac_cla_reorg".

Comment 11

[Bob Clary [:bc] (inactive)]

-> default assignee for old netscape assigned bugs.

Comment 12

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

This probably should have been killed with comment 10.  If this bug still exists in a recent app and you can provide relevant info, please feel free to re-open and move this to Core:Java Embedding Plugin.