3:30 p.m.
$ rpm -q firefox xulrunner
firefox-6.0-1.fc16.x86_64
xulrunner-6.0-1.fc16.x86_64
crashes for me (and opens its own crash handler dialog) when trying
to browse pages at http://bikemap.net
This is _without_ Adobe's Flash plugin in case that matters.
Can anyone confirm?
If so, the current test update is bad.
--
Fedora release 16 (Verne) - Linux 3.1.0-0.rc3.git0.0.fc16.x86_64
loadavg: 0.09 0.12 0.12
3:46 p.m.
On 08/28/2011 11:30 PM, Michael Schwendt wrote:
$ rpm -q firefox xulrunner
firefox-6.0-1.fc16.x86_64
xulrunner-6.0-1.fc16.x86_64
crashes for me (and opens its own crash handler dialog) when trying
to browse pages at http://bikemap.net
This is _without_ Adobe's Flash plugin in case that matters.
Can anyone confirm?
If so, the current test update is bad.
Just a wild guess -- what version of glibc do you have? For a short
while updates-testing had glibc-2.14.90-6 that made lots of apps crash.
yum downgrade 'glibc*' should fix it if that's the case.
https://admin.fedoraproject.org/updates/glibc-2.14.90-6
4:02 p.m.
On Sun, 28 Aug 2011 23:46:30 +0300, KL (Kalev) wrote:
On 08/28/2011 11:30 PM, Michael Schwendt wrote:
> $ rpm -q firefox xulrunner
> firefox-6.0-1.fc16.x86_64
> xulrunner-6.0-1.fc16.x86_64
>
> crashes for me (and opens its own crash handler dialog) when trying
> to browse pages at http://bikemap.net
>
> This is _without_ Adobe's Flash plugin in case that matters.
>
> Can anyone confirm?
> If so, the current test update is bad.
>
Just a wild guess -- what version of glibc do you have? For a short
while updates-testing had glibc-2.14.90-6 that made lots of apps crash.
yum downgrade 'glibc*' should fix it if that's the case.
https://admin.fedoraproject.org/updates/glibc-2.14.90-6
$ rpm -q glibc --dbpath /root/mnt/1/var/lib/rpm
glibc-2.14.90-4.x86_64
Terminal output: __libc_res_nquery: Assertion `hp != hp2' failed
4:15 p.m.
On Sun, 28 Aug 2011 23:02:16 +0200, MS (Michael) wrote:
On Sun, 28 Aug 2011 23:46:30 +0300, KL (Kalev) wrote:
> On 08/28/2011 11:30 PM, Michael Schwendt wrote:
> > $ rpm -q firefox xulrunner
> > firefox-6.0-1.fc16.x86_64
> > xulrunner-6.0-1.fc16.x86_64
> >
> > crashes for me (and opens its own crash handler dialog) when trying
> > to browse pages at http://bikemap.net
> >
> > This is _without_ Adobe's Flash plugin in case that matters.
> >
> > Can anyone confirm?
> > If so, the current test update is bad.
> >
>
> Just a wild guess -- what version of glibc do you have? For a short
> while updates-testing had glibc-2.14.90-6 that made lots of apps crash.
> yum downgrade 'glibc*' should fix it if that's the case.
>
> https://admin.fedoraproject.org/updates/glibc-2.14.90-6
$ rpm -q glibc --dbpath /root/mnt/1/var/lib/rpm
glibc-2.14.90-4.x86_64
Terminal output: __libc_res_nquery: Assertion `hp != hp2' failed
https://bugzilla.redhat.com/730856
https://fedoraproject.org/wiki/Common_F16_bugs#glibc-dns-crash
3:49 a.m.
Kalev Lember <kalevlember <at> gmail.com> writes:
...
Just a wild guess -- what version of glibc do you have? For a short
while updates-testing had glibc-2.14.90-6 that made lots of apps crash.
yum downgrade 'glibc*' should fix it if that's the case.
https://admin.fedoraproject.org/updates/glibc-2.14.90-6
I downgraded to:
glibc.i686 2.14.90-4 @fedora
and just got a crash again (dump submitted via Firefox built-in reporter).
JB
4:23 a.m.
On Tue, 30 Aug 2011 08:49:37 +0000 (UTC), J (JB) wrote:
Kalev Lember <kalevlember <at> gmail.com> writes:
> ...
> Just a wild guess -- what version of glibc do you have? For a short
> while updates-testing had glibc-2.14.90-6 that made lots of apps crash.
> yum downgrade 'glibc*' should fix it if that's the case.
>
> https://admin.fedoraproject.org/updates/glibc-2.14.90-6
I downgraded to:
glibc.i686 2.14.90-4 @fedora
and just got a crash again (dump submitted via Firefox built-in reporter).
Yeah, the -4 release is faulty: https://bugzilla.redhat.com/730856
Comment 3 in that ticket mentions a scratch-build that includes the
upstream patch, but the build is called 2.14.90-6 which is the same
version-release as the broken test update mentioned further above.
Rule of thumb: If scratch-builds contain something not added in pkg git,
modify %{release} appropriately and add a string like .test1 or .1.test
9:43 a.m.
On Tue, 30 Aug 2011 11:23:25 +0200
Michael Schwendt <mschwendt(a)gmail.com> wrote:
On Tue, 30 Aug 2011 08:49:37 +0000 (UTC), J (JB) wrote:
> Kalev Lember <kalevlember <at> gmail.com> writes:
>
> > ...
> > Just a wild guess -- what version of glibc do you have? For a
> > short while updates-testing had glibc-2.14.90-6 that made lots of
> > apps crash. yum downgrade 'glibc*' should fix it if that's the
> > case.
> >
> > https://admin.fedoraproject.org/updates/glibc-2.14.90-6
>
> I downgraded to:
> glibc.i686 2.14.90-4 @fedora
> and just got a crash again (dump submitted via Firefox built-in
> reporter).
Yeah, the -4 release is faulty: https://bugzilla.redhat.com/730856
Comment 3 in that ticket mentions a scratch-build that includes the
upstream patch, but the build is called 2.14.90-6 which is the same
version-release as the broken test update mentioned further above.
Rule of thumb: If scratch-builds contain something not added in pkg
git, modify %{release} appropriately and add a string like .test1
or .1.test
Sorry about that (I made the scratch build). I was sure hoping it would
have been fixed and pushed out in a real update long before now. ;(
kevin
12:48 p.m.
On Tue, 2011-08-30 at 08:43 -0600, Kevin Fenzi wrote:
On Tue, 30 Aug 2011 11:23:25 +0200
Michael Schwendt <mschwendt(a)gmail.com> wrote:
> On Tue, 30 Aug 2011 08:49:37 +0000 (UTC), J (JB) wrote:
>
> > Kalev Lember <kalevlember <at> gmail.com> writes:
> >
> > > ...
> > > Just a wild guess -- what version of glibc do you have? For a
> > > short while updates-testing had glibc-2.14.90-6 that made lots of
> > > apps crash. yum downgrade 'glibc*' should fix it if that's
the
> > > case.
> > >
> > > https://admin.fedoraproject.org/updates/glibc-2.14.90-6
> >
> > I downgraded to:
> > glibc.i686 2.14.90-4 @fedora
> > and just got a crash again (dump submitted via Firefox built-in
> > reporter).
>
> Yeah, the -4 release is faulty: https://bugzilla.redhat.com/730856
> Comment 3 in that ticket mentions a scratch-build that includes the
> upstream patch, but the build is called 2.14.90-6 which is the same
> version-release as the broken test update mentioned further above.
>
> Rule of thumb: If scratch-builds contain something not added in pkg
> git, modify %{release} appropriately and add a string like .test1
> or .1.test
Sorry about that (I made the scratch build). I was sure hoping it would
have been fixed and pushed out in a real update long before now. ;(
Your scratch build has also now expired. I just built myself a 4.1 with
the patch included...it's pretty easy to do.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | identi.ca: adamwfedora
http://www.happyassassin.net
2:31 p.m.
Date: Sun, 28 Aug 2011 22:30:14 +0200
From: mschwendt(a)gmail.com
To: test(a)lists.fedoraproject.org
Subject: Firefox crash
$ rpm -q firefox xulrunner
firefox-6.0-1.fc16.x86_64
xulrunner-6.0-1.fc16.x86_64
crashes for me (and opens its own crash handler dialog) when trying
to browse pages at http://bikemap.net
This is _without_ Adobe's Flash plugin in case that matters.
Can anyone confirm?
If so, the current test update is bad.
I'm running the same packages, save the 32-bit versions (and with Flash plugin).
Firefox
did crash on me this morning the first time I started it. Since then, it's been
rock-solid,
including navigating to that page.
~ imag1narynumber
--
Fedora release 16 (Verne) - Linux 3.1.0-0.rc3.git0.0.fc16.x86_64
loadavg: 0.09 0.12 0.12
--
test mailing list
test(a)lists.fedoraproject.org
To unsubscribe:
https://admin.fedoraproject.org/mailman/listinfo/test
10:31 p.m.
On Mon, 2011-08-29 at 15:31 -0400, imag1nary number wrote:
I'm running the same packages, save the 32-bit versions (and
with
Flash plugin). Firefox
did crash on me this morning the first time I started it. Since then,
it's been rock-solid,
including navigating to that page.
AIUI, it's very difficult to have a dependable reproducer for the bug,
because it happens when the server returns some fairly unusual responses
to a DNS query - SERVFAIL, NOTIMP or REFUSED. It's rare for a server to
be in a state where it will always give out such a response, but when it
happens, it's usually the case for a few minutes at a time. So some
query will seem to be a 'reliable reproducer' for half an hour, then
it'll suddenly quit being one.
Nevertheless, the bug does seem reasonably well understood and there's a
patch for it; just glibc...being glibc...seems to be the roadblock to it
getting in, up or downstream.
http://sourceware.org/bugzilla/show_bug.cgi?id=13013 has been sitting
there for over a month now.
I'm about < > this close to just pushing it into glibc myself with
provenpackager privileges...
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | identi.ca: adamwfedora
http://www.happyassassin.net
4655
days inactive
4657
days old
9 comments
6 participants
participants (6)
-
Adam Williamson -
imag1nary number -
JB -
Kalev Lember -
Kevin Fenzi -
Michael Schwendt