Re: Testing Fedora - small (?) suggestion.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Sun, 2006-11-12 at 11:53 -0500, Jesse Keating wrote:
> On Sunday 12 November 2006 11:28, Gilboa Davara wrote:
> > By -design-, Anacnoda bombs out if anything is broken.
> 
> Actually I strongly question this.  Anaconda seems perfectly happy to install 
> packages with broken deps.  its when things are missing for %post installs 
> that things go haywire.
> 
> > If you maintain that there's no way to create/detect/exclude a stable
> > tree image, Anaconda should be able to detect and remove missing deps
> > -and- do graceful recovery (as opposed to graceful exit) when something
> > is broken but was undetected during the initial scan.
> > This should ensure that unless something is broken within Anaconda
> > (which is interesting and worthwhile by itself), most install attempts
> > will end with a bootable image - no matter which package configuration
> > was selected. Missing stuff (or broken stuff) can always be installed
> > using "yum --exclude XXX update"
> 
> But in my new world of rawhide, where anaconda knows nothing about packages 
> until you start the install and it pulls the latest repodata from whatever 
> repo this becomes a bit different.  Anaconda doesn't know anything until you 
> run it, not when the boot images were created.

Can it be fixed by pointing Anaconda to a second, automatically
screened, repodata?

- Gilboa

-- 
fedora-devel-list mailing list
fedora-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/fedora-devel-list

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [Fedora Announce]     [Fedora Kernel]     [Fedora Testing]     [Fedora Formulas]     [Fedora PHP Devel]     [Kernel Development]     [Fedora Legacy]     [Fedora Maintainers]     [Fedora Desktop]     [PAM]     [Red Hat Development]     [Gimp]     [Yosemite News]
  Powered by Linux