5884 - NECKO: crash going to netcenter's small business page

Status: VERIFIED FIXED

(Core :: Networking, defect, P3)

Description

[Akkana Peck]

Went to home.netscape.com, clicked on the "Does size really matter?" link
halfway down the right column which in 4.51 would have taken me to
http://home.netscape.com/netcenter/smallbusiness/index.html, but instead I
crashed.  cd is 0 in net_ProcessHTTP.  Happens every time, but only when I click
through home.netscape.com; doesn't happen if I just enter the URL manually
(maybe netcenter is feeding me a different url because of the user agent?  I
can't seem to get the apprunner browser to tell me what it thinks the link
should be).

#0  0x40198e46 in net_ProcessHTTP (ce=0x86bc658)
    at /builds/mon/mozilla/network/protocol/http/mkhttp.c:3621
#1  0x40261f17 in NET_ProcessNet (ready_fd=0x84c9be8, fd_type=2)
    at /builds/mon/mozilla/network/main/mkgeturl.c:3355
#2  0x40269ea5 in NET_PollSockets ()
    at /builds/mon/mozilla/network/main/mkselect.c:320
#3  0x4028bb02 in nsNetlibService::NetPollSocketsCallback (aTimer=0x825b168,
    aClosure=0x8075cc8)
    at /builds/mon/mozilla/network/module/nsNetService.cpp:1263
#4  0x40177de9 in TimerImpl::FireTimeout (this=0x825b168)
    at /builds/mon/mozilla/base/src/gtk/nsTimer.cpp:73

Comment 1

[Gagan]

Deferring till Necko lands...

Comment 2

[Akkana Peck]

Maybe related to 5196?  Jfrancis says this, which sounds suspiciously similar to
the stack trace of this bug:
I recently tracked down a bug (now handed off to gagan) where
net_ProcessHTTP() managed to get called recursively.  The inner call
detected an error and released the connection data.  The outer call was
quite offended by this chicanery and died.  When you get the crash, you
only see the outer call on the stack so it's not so obvious what's going
on.

Comment 3

[Gagan]

Per DP's suggestion marking these till M8. Though Necko lands with M7, we will
be able to verify it for M8.

Comment 4

[James Everingham]

I'm moving this to target M9, Necko will be enabled somewhere during late M8 or
early M9.  We will need to get on this and it cannot be postponed past the M9
milestone.

Comment 5

[leger]

Changing all Networking Library/Browser bugs to Networking-Core component for
Browser.

Occasionally, Bugzilla will burp and cause Verified bugs to reopen when I do
this in a bulk change.  If this happens, I will fix. ;-)

Comment 6

[Gagan]

Should be fixed with Necko. Pl. verify.

Comment 7

[Gagan]

Should be fixed with Necko. Pl. verify.

Comment 8

[Paul MacQuiddy]

well, this link is gone so I can't test it. Redirects are working now, though.

Comment 9

[leger]

Bulk move of all Networking-Core (to be deleted component) bugs to new
Networking component.