Does firefox upgrades need some pre/post install love to close and restart firefox during upgrades

Adam Williamson awilliam at redhat.com
Tue Nov 17 06:54:19 UTC 2009


On Tue, 2009-11-17 at 08:45 +0200, shmuel siegel wrote:
> Adam Williamson wrote:
> > On Tue, 2009-11-17 at 07:17 +0200, shmuel siegel wrote:
> >
> >   
> >> What are you trying to say. A set of users are saying that current 
> >> procedures cause data loss for them 
> >>     
> >
> > Um? I haven't seen anyone saying that.
> >
> >   
> Qoute from Adam Prybl earlier in the thread
> 
> "I also have troubles with this. The worst thing is that firefox often 
> after such update dies, without saving the tabs etc. as it does when it 
> is killed. And because it dies after update anyway, I would just suggest 
> to put a "killall firefox" in PREIN script. Because this will trigger 
> the firefox saving mechanism and would not cause a data loss."
> 
> And this from Gregory Maxwell
> 
> "After this burned me the first time in fedora 10 (?) I made my script
> that fires off my yum update killall firefox exactly as you suggest.
> 
> I know that killall firefox in PREIN sounds outrageous but it would be
> *strictly superior* to what fedora is allowing to happen today."
> 
> And this from Dariusz J. Garbowski
> 
> "Particularly a good idea for times when the user is working with a 
> session on his bank account / fulfilling payments / on his retirement 
> fund account, etc."
> 
> If these are inaccurate, then I take back what I said.

Well, that's not what I was thinking of as data loss, but now I get what
you mean. However, the alternative could potentially cause just the same
'data loss'. I just am not convinced the right way to address this is to
go around forcibly killing Firefox sessions, when part of the problem is
that - apparently - the session saving code doesn't always work
perfectly...it seems a bit premature to conclude that it's always going
to work fine.

-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net




More information about the test mailing list