662544 - IPv6 Site unreachable only with firefox

Status: RESOLVED DUPLICATE of bug 621558

(Core :: Networking, defect)

Description

[Oufcom]

User-Agent:       Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; fr; rv:1.9.0.19) Gecko/2011032020 Camino/2.0.7 (MultiLang) (like Firefox/3.0.19)
Build Identifier: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; fr; rv:1.9.0.19) Gecko/2011032020 Camino/2.0.7 (MultiLang) (like Firefox/3.0.19)

First, sorry if i did a mistake somewhere, i'm new here, here's the "bug"

http://bat.oufcommunication.fr is unreachable with any recent gecko/firefox type browser with ipv6

don't work with on Mac OS X with IPv6 :
Camino 2.0.7 (Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; fr; rv:1.9.0.19) Gecko/2011032020 Camino/2.0.7 (MultiLang) (like Firefox/3.0.19))
Firefox Aurora 6.0a2
Firefox 4.0.1 (current official)

work with on Mac OS X with IPv6 (or IPv6 disabled in option) :
Shiira 1.2.2 (defaut configuration) 
Firefox Aurora 6.0a2 (network.dns.disableIPv6; with true option)
Firefox 4.0.1 (network.dns.disableIPv6; with true option)
Chrome 11
Opera 9.24
Internet Explorer mac 5.2

It's possible there's something wrong with the configuration of the site (host on Apache/1.3.37 + bind dns) (seems happens with Apache/2.0.64 too), but it's strange that the bug only happens with gecko based browser,

by the way i'm available to do all test needed (even test on Apache 2)

Reproducible: Always

Steps to Reproduce:
1. go on http://bat.oufcommunication.fr
2. 
3.

Actual Results:  
unreachable, no error showned

Expected Results:  
page showed

you can test with network.dns.disableIPv6 with true option, it should work well

ipv6 connexion used : free (free.fr)

if you need anymore information, please contact me

Comment 1

[Oufcom]

Main about:buildconfig (firefox 4)

about:buildconfig
Source

Built from http://hg.mozilla.org/releases/mozilla-2.0/rev/fca718600ca0
Build platform
target
i386-apple-darwin10.2.0
Build tools
Compiler 	Version 	Compiler flags
gcc-4.2 -arch i386 	gcc version 4.2.1 (Apple Inc. build 5646) (dot 1) 	-Wall -W -Wno-unused -Wpointer-arith -Wcast-align -W -gdwarf-2 -isysroot /Developer/SDKs/MacOSX10.5.sdk -fno-strict-aliasing -fpascal-strings -fno-common -pthread -DNO_X11 -DNDEBUG -DTRIMMED -gdwarf-2 -O3 -fomit-frame-pointer
g++-4.2 -arch i386 	gcc version 4.2.1 (Apple Inc. build 5646) (dot 1) 	-fno-rtti -fno-exceptions -Wall -Wpointer-arith -Woverloaded-virtual -Wsynth -Wno-ctor-dtor-privacy -Wno-non-virtual-dtor -Wcast-align -Wno-invalid-offsetof -Wno-variadic-macros -Werror=return-type -gdwarf-2 -isysroot /Developer/SDKs/MacOSX10.5.sdk -fno-strict-aliasing -fpascal-strings -fno-common -fshort-wchar -pthread -DNO_X11 -DNDEBUG -DTRIMMED -gdwarf-2 -O3 -fomit-frame-pointer
Configure arguments

--target=i386-apple-darwin10.2.0 --with-macos-sdk=/Developer/SDKs/MacOSX10.5.sdk --enable-application=browser --enable-update-channel=release --enable-update-packaging --enable-tests --enable-official-branding --enable-debug-symbols=-gdwarf-2
Source

Built from http://hg.mozilla.org/releases/mozilla-2.0/rev/fca718600ca0
Build platform
target
x86_64-apple-darwin10.2.0
Build tools
Compiler 	Version 	Compiler flags
gcc-4.2 -arch x86_64 	gcc version 4.2.1 (Apple Inc. build 5646) (dot 1) 	-Wall -W -Wno-unused -Wpointer-arith -Wcast-align -W -gdwarf-2 -isysroot /Developer/SDKs/MacOSX10.6.sdk -fno-strict-aliasing -fpascal-strings -fno-common -pthread -DNO_X11 -DNDEBUG -DTRIMMED -gdwarf-2 -O3 -fomit-frame-pointer
g++-4.2 -arch x86_64 	gcc version 4.2.1 (Apple Inc. build 5646) (dot 1) 	-fno-rtti -fno-exceptions -Wall -Wpointer-arith -Woverloaded-virtual -Wsynth -Wno-ctor-dtor-privacy -Wno-non-virtual-dtor -Wcast-align -Wno-invalid-offsetof -Wno-variadic-macros -Werror=return-type -gdwarf-2 -isysroot /Developer/SDKs/MacOSX10.6.sdk -fno-strict-aliasing -fpascal-strings -fno-common -fshort-wchar -pthread -DNO_X11 -DNDEBUG -DTRIMMED -gdwarf-2 -O3 -fomit-frame-pointer
Configure arguments

--target=x86_64-apple-darwin10.2.0 --with-macos-sdk=/Developer/SDKs/MacOSX10.6.sdk --enable-application=browser --enable-update-channel=release --enable-update-packaging --enable-tests --enable-official-branding --enable-debug-symbols=-gdwarf-2

Comment 2

[Justin Dolske [:Dolske]]

Does it still fail if you try in safe mode or with a new profile? What exactly is the failure ("unreachable, no error showned")? Do you get an error page, or just the browser just sit and never load the page?

Comment 3

[Matthias Versen [:Matti]]

What happens if you enter the IPv6 in a working browser :
[2001:41d0:1:a3dd:91:121:108:42] ?
You should get a security warning because the site redirects to https and the IP is not in the certificate.

I don't have IPv6 and used http://web-sniffer.net for testing.
I get a "no route to host error" for http://[2001:41d0:1:a3dd:91:121:108:42] !

http://[2a02:2e0:3fe:100::8] works however = http://web-sniffer.net seems to have a working IPv6 setup.

I'm pretty sure that your old IE doesn't support IPv6 and Opera and Chrome could just prefer the IPv4.

Comment 4

[Matthias Versen [:Matti]]

I found another online tool: http://4or6.com/traceroute 
I tested with bat.oufcommunication.fr and the output for the last hop is:
11  rbx-1-6k.fr.eu (2001:41d0::531)  3157.559 ms !H * *

!H is Host unreachable

Comment 5

[Oufcom]

@Justin Dolske

safe-mode do the same.

"Do you get an error page, or just the browser just sit and never load the page?"
browser try to load, sit and never load anything, don't show any error or nothing



@Matthias Versen

don't misunderstand me, i don't say there's no problem to reach the site in v6, in fact, there is one (and probably more that one since 2001:41d0:1:a3dd:91:121:108:42 does not ping6 too), in any case, main server (apache 1) doesn't not support ipv6, and that's why browser should switch into ipv4 (seems that's the other browser -excluding shiira and internet explorer who does not support v6- do, i guess).

. http://[2001:41d0:1:a3dd:91:121:108:42]:90 (apache2) should work, and that's not the case (that show there's a problem in server configuration too)
i can work to solve network configuration problem, but do you prefer that i wait during the problem is reproducable ?

example of one site (on same server, but not on same network interface) who works correctly with firefox and others browser
http://[2001:41d0:1:a3dd:91:121:108:45]:90
= http://www.culturepaysmontlucon.fr:90 (on apache 2/v4 or v6) and 
= http://www.culturepaysmontlucon.fr (on apache 1/v4)

i'm still not sure it can be considered like a "bug", it can just be firefox being more strict than the other browser ... (it's strange there's no error shown in any way)

i let you choose if you consider it like a bug or not, and if i can or not work for now to make 2001:41d0:1:a3dd:91:121:108:42 reachable

by the way,
thanks for so quick answer,
regards

Comment 6

[Oufcom]

sorry for asking again,
do i have to let the server configuration stay like it is now (if someone want to have a case reproducable) or there's no need of it ?

Thanks

Comment 7

[Matthias Versen [:Matti]]

I think it's not necessary to keep your configuration.
The case is clear: 
You have a domain with an IPv4 and an IPv6 DNS entry but IÜv6 isn't working.
That is basically not a browser issue but we might implement a failover and this could be a dupe of an already existing bug.

Comment 8

[Matthias Versen [:Matti]]

should be fixed with bug 621558