tags 444767 unreproducible
close 444767
thanks

* Matthias Bläsing (matthias.blaes...@rwth-aachen.de) wrote:
> Hey,
> 
> Am Sonntag, den 26.10.2008, 17:57 -0400 schrieb Eric Dorland:
> > * Matthias Bläsing (matthias.blaes...@rwth-aachen.de) wrote:
> > > Package: iceweasel
> > > Version: 2.0.0.7-2
> > > Severity: normal
> > > 
> > > --- Please enter the report below this line. ---
> > > 
> > > Hey,
> > > 
> > > egroupware showed very long login times for our setup, where it was
> > > configured to run over ssl (tests without SSL didn't show this error). I
> > > thought I would hunt a bug in egroupware, but then I tested with
> > > links-ssl, konquerer, IE7, ... and got "normal times" (~ 6s). With
> > > firefox/iceweasel/epiphany(the gecko version) I get time ranging from
> > > 12s to 24s (most request take approx. 24s).
> > > 
> > > So at some point I installed LiveHeaders and saw that the get requests
> > > after a 302 Redirect took approx 15s to complete. So I had a look at the
> > > redirect function of egroupware. This is not optimal, but at least for
> > > the apache case we have it does this:
> > > 
> > > Create a Location Header with the new url and an empty body only
> > > containing two \n. Well these broke firefox/iceweasel. I commented them
> > > out and it now the times are down to ~6s also on iceweasel.
> > > 
> > > According to the RFC (2616) a message-body should be send along with the
> > > header. So Iceweasel should not stall because of this.
> > 
> > I have hard time believing this is the problem. Is it still the same
> > with Iceweasel 3.0?
> 
> actually I could not believe it either, but the measurements at that
> time were unmistakable. I can't reproduce it anymore, as the installed
> version was fixed (the old package was lost in the time since then) and
> the egroupware from lenny works flawlessly with Iceweasel 3.0.
> 
> Lack of time does not allow me to build an environment to experiment
> with an old firefox and an old egroupsware setup or a test script.
> 
> So feel free to close this bug as unreproducible.

Thanks Matthias, closing.

-- 
Eric Dorland <e...@kuroneko.ca>
ICQ: #61138586, Jabber: ho...@jabber.com

Attachment: signature.asc
Description: Digital signature

Reply via email to