45227 - Back button does not work (in frames context) at this url

Status: VERIFIED WORKSFORME

(Core :: DOM: Navigation, defect, P4)

Description

[Sander van Rijnswou]

mozilla m16 solaris 2.7 build 2000062115
mozilla m16 win 98 build 2000062108

First of all the back button

1.goto www.securityfocus.com
2. click on forums->mailing list->bugtraq

The new page will take over the browser, but both communicator and
explorer put this in a frame. Pressing the back button will leave you
on this page. If there was some history before securityfocus it will
eventually get there but the first of page of securityfocus will never
be reached.

Under solaris the system is now instable.

3.  click on addres bar and remove directories so that the addres now
reads www.securityfocus.com again (press enter)

this will always produce a crash. Instead of (3) you could also go to
www.slashdot.org for a crash. BUT if you go www.w3.org there will not
be a crash.

Last few lines in console

Document: Done (2.988 secs)
*** check number of frames in content area
Document http://www.securityfocus.com/forums/bugtraq/intro.html loaded successfully
JavaScript error:
http://www.securityfocus.com/forums/bugtraq/intro.html line 19: parent.ads has
no properties

Entry at index 0 is www.securityfocus.com
Entry at index 1 is http://www.securityfocus.com/
WEBSHELL+ = 8
WEBSHELL+ = 9
WEBSHELL+ = 10
WEBSHELL+ = 11
Segmentation Fault - core dumped

Stacktrace (gdb, bt)

#0  0xfda553bc in nsDSURIContentListener::DoContent ()
from /home/sander/mozilla/package/libdocshell.so
#1  0xfdbd5804 in nsDocumentOpenInfo::DispatchContent ()
from /home/sander/mozilla/package/components/liburiloader.so
#2  0xfdbd5218 in nsDocumentOpenInfo::OnStartRequest ()
from /home/sander/mozilla/package/components/liburiloader.so
#3  0xfdb85b0c in nsHTTPFinalListener::OnStartRequest ()
from /home/sander/mozilla/package/components/libnecko.so
#4  0xfdb83b0c in nsHTTPCacheListener::OnStartRequest ()
from /home/sander/mozilla/package/components/libnecko.so
#5  0xfdb6c54c in nsDiskCacheRecordChannel::OnStartRequest ()
from /home/sander/mozilla/package/components/libnecko.so
#6  0xfdb3e69c in nsOnStartRequestEvent::HandleEvent ()
from /home/sander/mozilla/package/components/libnecko.so
#7  0xfdb3e0f0 in nsStreamListenerEvent::HandlePLEvent ()   from
/home/sander/mozilla/package/components/libnecko.so
#8  0xff2fb5b0 in PL_HandleEvent () from /home/sander/mozilla/package/libxpcom.so
#9  0xff2fb4fc in PL_ProcessPendingEvents ()
from /home/sander/mozilla/package/libxpcom.so
#10 0xff2fc2b4 in nsEventQueueImpl::ProcessPendingEvents ()
from /home/sander/mozilla/package/libxpcom.so
#11 0xfe0fed24 in nsAppShell::SetDispatchListener ()
from /home/sander/mozilla/package/components/libwidget_gtk.so
#12 0xfe0fea34 in keysym2ucs ()
from /home/sander/mozilla/package/components/libwidget_gtk.so
#13 0xfde83c38 in g_io_unix_dispatch (source_data=0x265c40, current_time=
0xffbef668,
user_data=0x11b520) at giounix.c:135
#14 0xfde858f0 in g_main_dispatch (dispatch_time=0xffbef668) at gmain.c:656
#15 0xfde861f0 in g_main_iterate (block=1236, dispatch=1) at gmain.c:877
#16 0xfde86400 in g_main_run (loop=0x255a28) at gmain.c:935
#17 0xfdfbcef8 in gtk_main () at gtkmain.c:476
#18 0xfe0ff2ec in nsAppShell::Run ()
from /home/sander/mozilla/package/components/libwidget_gtk.so
#19 0xfecd8bd8 in nsAppShellService::Run ()
from /home/sander/mozilla/package/components/libnsappshell.so
#20 0x153e0 in CheckForNewChrome ()
#21 0x158e4 in main ()

Comment 1

[leger]

Adding crash keyword

Comment 2

[Andreas Franke (gone)]

Sander, are you still seeing this with recent builds?
I tried both PC/Linux 2000071208 and Sun-Sparc/Solaris 2000070921, but I could
not reproduce any of the mentioned problems.

Comment 3

[Andreas Franke (gone)]

Well, I have to admit that I the back button sometimes doesn't do anything at
all, but I think this is a known problem. When I tried to query bugzilla for the
bug number, mozilla crashed with a "Bus Error"... :(

Comment 4

[Andreas Franke (gone)]

I could not find a bug about the "back button does not do anything" problem.
Confirming this bug for this reason: After step 2. the back button doesn't work.

Bug 40116 "Crash in nsDSURIContentListener::DoContent loading this URL" seems to
be about a similar crash as the one reported here, but that bug is also marked
verified fixed.

Comment 5

[Andreas Franke (gone)]

Sending to history, removing crash from summary/keywords for now, cc self.
Sander, please put that back (or, better, file a new bug for that) if you are
still seeing the crash.

Comment 6

[Andreas Franke (gone)]

Doh. Forgot to reassign to new owner.

Comment 7

[Rich Burridge]

I've changing the Platform/OS to PC/Windows98 which is a tier-1 platform
(Solaris isn't).

This bug will therefore get higher visability, and a greater chance of being
fixed.

Comment 8

[Claudius Gayle]

rich this is probably a dupe of another SH bug that radha has - there are several floating around about
Session History and frames.

Comment 9

[Blake Ross]

Marking dup of bug 32444, the general frame/SH problem.  Please be sure to 
retest this once that's fixed.

*** This bug has been marked as a duplicate of 32444 ***

Comment 10

[R.K.Aa.]

32444 is marked fixed, and with linux SEA 2000102021 i still can't use the
back-button at securityfocus.com. I can barely use the history dropdown list on
back-button either. Some will work - others not - and all have the same name.
There is no crash. There is no back-feature either. back+frames in mozilla is a
pure mess.

Comment 11

[Azrael AOD]

If this is not fixed and bug 32444  is fixed, then this cannot be a dupe, I 
have asked there for someone to take a look.
Could those seeing this bug perhaps take a moment to check out bug 32444 and 
see if they are affected by that one too. help decide if these bugs are the 
same and if they are indeed fixed or not.

Comment 12

[Claudius Gayle]

I'm reopening this bug to separate it from the bug it was duped of. That was general and this is more
specific, let's try to keep it that way :-)

Comment 13

[Viswanath Ramachandran]

nav triage: p4. 

Comment 14

[Asa Dotzler [:asa]]

*** Bug 63482 has been marked as a duplicate of this bug. ***

Comment 15

[Paul Chen]

nav triage team:

Marking nsbeta1+

Comment 16

[Radha on family leave (not reading bugmail)]

This is working now. I presume the last checkin I made just before RTM was
pushed out the door fixed this.

Comment 17

[sairuh (rarely reading bugmail)]

mass-verifying WorksForMe bugs which haven't changed since 2001.12.31.

set your search string in mail to "EmperorLondoMollari" to filter out these
messages.