firefox-3.0_2,1 no reactions

[Guido Falsi]

On Fri, Jul 11, 2008 at 07:14:26AM +0200, Helko Glathe wrote:
> Am Freitag 11 Juli 2008 03:05:18 schrieb Guido Falsi:
> > Doug Barton wrote:
> > > Guido Falsi wrote:
> > >> I just disabled the attack and forgery site checks
> > >
> > > Exactly which prefs are you referring to there?
> >
> > in the security pane, the second and third option, they read:
> >
> > Tell me if the site I'm visiting is a suspected attack site
> >
> > and
> >
> > Tell me if the site I'm visiting is a suspected forgery
> 
> Disabling this options doesn't fix the problem.

>From private emails with Josh Tolbert there are some problems with
NFS locking too. Most probably firefox is heavily using it.

At home I have an nfs server and 2 clients and all using statd and
lockd(and since the inclusion of the in kernel locking code I have
to restart statd removing it's status file in /var/db when I reboot
a client on both the server and the client, but I don't have the
knowledge to debug this. I think a recwent commit to -current has
fixed this though. I'm waiting for it to be MFCed).

NFS locking works usually, but when I don't remove the statd.status
file after rebooting a machine the calls to lockd seem to block
indefinitely and the app just sits there, maybe this is what is
happening to you.

Disabling those functions and removing the sqlite files from the
.mozilla subdir was needed to me to solve the slowness problm,
because the app was rereading via network the whole files each time
a page loaded.

Please let me know if some specific information is needed. I hope
I vcan help shed some light.

My machines are all 7.0-STABLE cvsupped around a month ago.

-- 
Guido Falsi <mad at madpilot.net>