156870 - Recursive frameset crashes in various different ways

Status: RESOLVED WORKSFORME

(Camino Graveyard :: General, defect)

Description

[Stephane Moureau]

That intentional javascript crash page (frameset loop) reacts very differently in Chimera 
(build 20020710 16:11).

Sometimes, it crash after it has draw 4 cells, see crash log.

Sometimes, its contents is drawn entierly (~20x20 cells) immediately after you've seen 
the 4 cells.
You get a grid, but the 'cells' are bigger than those obtained in M$IE 5.21.

IE draws them one by one, you see all the process until it crash when the status bar show 
the link ...?11 at that stage the cell are very very small, chimera has cells 2 or 3 times 
bigger than IE (? 8 -> 11).
Chimera becomes very slow, probably huge ram comsuption and seems to stay stable 
and is still working.

Sometimes, it stop saying that it can't reach the link ...?8 and an alert sheet don't stop 
appearing, you can't quit, stop, you must force quit.
The console in that case says:
2002-07-11 12:55:33.625 Navigator[957] *** WARNING: Method 
runModalForWindow:relativeToWindow: in class NSApplication is obsolete and will be 
removed in release GM ***

What's strange is that it was accessible few seconds before.

In Chimera during that slow drawing, you see in the Status bar that say ~'Read user file 
in the chimera profile'.

Comment 1

[Stephane Moureau]

Attachment: js2-crash log

js2-crash log

Comment 2

[Greg K.]

*** Bug 156869 has been marked as a duplicate of this bug. ***

Comment 3

[Greg K.]

Stephane, what Chimera build are you using? Always be sure to include your build
information in bug reports.

Comment 4

[Greg K.]

Also, Stephane, did you test this in Mozilla? What were the results?

Comment 5

[Stephane Moureau]

What's going on...
the build is on the first line and the attach is there too ?

Comment 6

[Greg K.]

Whoops, sorry Stephane. I don't know how I missed that.

Comment 7

[Greg K.]

using Chimera/20020708, the tescase runs for a time then an error sheet bearing
the text, "The file /crash/js2.html?8 cannot be found. Please check the location
and try again" is repeatedly shown, eventually also crashing at js_AllocGCThing.

Comment 8

[Greg K.]

Attachment: Crash report generated by Chimera/20020708 showing crash at js_AllocGCThing

Comment 9

[Greg K.]

...but, WorksForMe using FizzillaCFM/2002071208.

Comment 10

[Chris Casciano]

WFM Camino 2003061704... my guess is this got picked up in the move to the truck
if it was ok in mozilla proper in the last set of comments.

anyone argue with a resolution of WFM here?

Comment 11

[Chris Casciano]

well ok then... since noone objected -> WFM