17557 - [MLK] Build: 1999102808 cases page fault error

Status: RESOLVED WORKSFORME

(Core :: CSS Parsing and Computation, defect, P3)

Description

[solomonr]

When I got to the w3c page, as the css is redrawing the page I get the page
fault error.

Question, is Java enabled in this version, if so how do I get it to run.

Thanks.

Comment 1

[leger]

chrisd, can you reproduce this?

Comment 2

[Christine Hoffman]

Using build 1999110208 on Win 98, unable to reproduce bug. Marking bug
WORKSFORME. Also, I don't see javascript on this page but to enable Java or
Javascript, go to Edit/Preferences, click on 'Advanced' and then check the
appropriate Java enable boxes

Comment 3

[solomonr]

Yesterday, 11-3-1999, using build 1999110208, when I went to the
www.w3c.org/syle/css/Test series of pages I able to get the browser to run out
of memory. If you are to run a series of the CSS tests from the W3C eventually
the DOS window stop displaying the messages and you get a Window's message
saying that I should close other applications that you have open. I wasn't
running anything else when I tried the tests.

Comment 4

[leger]

Clearing WorksForMe resolution due to reopen of this bug.

Comment 5

[Pierre Saslawsky]

salomonr: could you provide more details about how to reproduce the bug? In
particular, how many pages have you visited before running out of memory?

Comment 6

[Pierre Saslawsky]

More info from the reporter:

======
I first started at www.w3c.org. I then selected the CSS1 test option. I
then selected the first test on that screen and continued to choice the
next option on the top of the screen.

I was also able to get it to crash via www.freerepublic.com web site.
This was by choosing the read article option and read new articles and
hitting the back button. This took about thirty tries.

Here is the Configuration from system info:

Microsoft Windows 98 4.10.1998
Clean install using MSDN CD /T:C:\WININST0.400 /SrcDir=D:\WIN98 /IZ /II
/IS /IQ /IT /II /NR /II /C  /U:xxxxxxxxxxxxxxxxx
IE 5 5.00.2314.1003
Uptime: 0:00:03:10
Normal mode

GenuineIntel Pentium(r) II Processor Intel MMX(TM) Technology
128MB RAM
88% system resources free
Windows-managed swap file on drive C (2208MB free)
Available space on drive C: 2208MB of 6130MB (FAT32)
======

Chris: I can't reproduce this bug. Within the same session, I browsed through
many articles on freerepublic.com and then I scanned the entire CSS1 test suite
back and forth without problem. Could you have a look? A solution would be to
make a list of all the URLs in the CSS1 test suite and ask Bindu Sharma to write
a script to scan continuously through them. She already has something very
similar to that.

For now, I'm marking this bug as dependent on bug 17390 "leaking documents and
webshells while running editor" (ignore the "while running editor" part) and
setting a milestone of M20.

Comment 7

[Pierre Saslawsky]

Chris, could you try to reproduce this crash? Salomon wrote me today:
---
With this week's build I was able to get the browser to blow up
consistantly. First I went to sportingnews.com and then to fastball.com.
On most occasions the fastball.com site never loaded. The active gif on
the lefthand side of the screen seems to display incorrectly, maybe this
is related. Also when a site, i.e. yahoo.com, comes up the screen is a
mass of colors, is this correct or is this a video card issue?
---

Comment 8

[Christine Hoffman]

Using 12/2 build on Win 98, cannot reproduce consistent crashes. Loaded
www.sportingnews.com, then loaded www.fastball.com. No problem loading
'fastball.com'. There is something wrong with the animated gif - will have to
look at that tomorrow. Went back to W3C.org to flip through the CSS1 test suite
and could not get application to crash. On the www.yahoo.com page, there is a
splash of lines crossing the window prior to the page loading. This is
happening on other pages as well but not as pronounced as on yahoo.com. Will
check to see if this has been reported. On the crash issue, however, unable to
reproduce.

Comment 9

[Pierre Saslawsky]

The problem with the animated gif has been reported as bug 21903.
For the rest, sorry Salomon, I'm closing the bug as Remind because we can't
reproduce the crashes here. As we release more betas with TalkBack enabled, we'll
be able to gather much more information about crashes.

Comment 10

[Christine Hoffman]

Verified remind

Comment 11

[Pierre Saslawsky]

Reopening my REMIND bugs

Comment 12

[Pierre Saslawsky]

...and closing this one as WorksForMe.

Comment 13

[Christine Hoffman]

Changing QA contact