Firefox performance sucks!

[Rodd Clarkson]

On Sun, 2007-08-26 at 14:59 +1000, David Timms wrote:
> Rodd Clarkson wrote:
> > On Sat, 2007-08-25 at 20:51 -0600, Kevin DeKorte wrote:
> > 
> >>> So what else can I test?
> >> It sounds like you may have a DNS resolution problem. Might want to make
> >> sure the DNS server you are using is the best you can. I actually have a
> >> caching DNS server on my network, but even my ISPs is not that slow.
> > 
> > Actually, I've done a little more testing and I copied ~/.mozilla to
> > ~/.mozilla.org and all of a sudden firefox was running nicely again.
> > 
> > A little more looking around (copying files from the old .mozilla to the
> > new .mozilla I discover that something in my prefs.js file is causing
> > the problem.
> > 
> > I've been using (through various upgrades) this prefs.js we before
> > firefox became the default in Fedora/Redhat so it appears there's
> > something in there that it doesn't like.
> > 
> > I'm now trying to figure out what it is.
> Do you have ipv6 enabled ?  ifconfig...
> 
> Do you have cache settings {moz} set really large ?
> 
> If the problem is at all network based, then running wireshark, capture, 
> and then  trying to load the page gives the answer to:
> - how long till the dns is requested /responds ?
> - is it waiting for ipv6 results ?
> - which sites is it actually getting connected to:
> - is that site fast.

I'm positive the problems are due to one of the settings in prefs.js
now.

I realized I was having the same issues with seamonkey, so I moved the
prefs.js file to prefs.js.old and it recreates the file on start up and
then it loads pages at a much more acceptable speed.

I've done the same with my firefox prefs.js (using all my other original
settings) and it's vastly improved (and running like I would expect it
too).

I'm not sure what to look for in the prefs.js file, but if someone else
is intersted in grabbing my original (broken?) prefs.js file and trying
it on their install (to see if it causes the same grief) then that might
be interesting to do.

Should I file a bug against this?


R.
-- 
"It's a fine line between denial and faith.
 It's much better on my side"