FC4t2 no good without LILO

Guy Fraser guy at incentre.net
Thu Apr 14 20:52:35 UTC 2005


On Wed, 2005-13-04 at 13:21 -0400, Dan Williams wrote:
> On Wed, 13 Apr 2005, Guy Fraser wrote:
> > On Tue, 2005-12-04 at 19:46 -0400, David Zeuthen wrote:
> > > No, of course not. 
> > > 
> > > If something is wrong with app X the solution is to fix app X, not ship
> > > an alternative to app X just because some users thinks it work better
> > > for them. As a comparison, it's not like we ship the FreeBSD kernel just
> > > because some users think that it has features that the Linux 2.6 kernel
> > > doesn't.
> > Then get the problems fixed... DUH!!!
> 
> Well, it also requires interaction with _users_ (such as yourself) for engineers 
> to determine what the exact problems are and how to fix them.  Obviously Red Hat 
> doesn't have the budget nor the time to acquire every computer ever made to test 
> on, which is where users can help out quite a bit.  Users need to 
> _constructively_ respond to engineers with problem reports that include specific 
> information and steps to reproduce the problems.
> 
> Again, _constructive_ problem reports and interaction.  Bugs don't get fixed if 
> engineers can't figure out the problem or don't have the same hardware, or if 
> users can't constructively describe their problems.  If you help out by adding 
> your configuration and steps to reproduce to existing bugzillas, that can make 
> the difference between the engineer finding similarities that enable him/her to 
> fix the issue, and not being able to fix the issue because he/she doesn't have 
> enough information.
> 
> Dan
I tried for a couple weeks, but the "engineers" could give me 
nothing that would help, and at the time they had the snotty 
attitude.

Yesterday I was told that add-on PCI SATA {Non-RAID} controllers 
were not standard. We have been told many times on the list that 
on-board SATA RAID is no good and shouldn't be used. It has also 
been said that add-on PCI SATA RAID is also no good. Since I have 
never found a Motherboard that supports SATA on-board that does 
not have RAID, I have no idea what the "engineers" have been using 
to test SATA. It seems like just another runaround.

I really don't see why LILO can't be kept around until the 
problems with GRUB have been resolved. It is not in the 
best interest of anyone involved with the Redhat Fedora Core 
Community Project, too alienate users for no real reason.

I have heard a lot of FUD and hokum about why LILO can't be 
supported, but every reason appears to be based on inaccurate 
information.

You want a test bed :

Get a mother board with onboard PATA and SATA RAID then add 
a PCI ATA133 card and a PCI SATA card with or without RAID.
Run many simulations adding and removing non boot drives, 
then change the location of the boot drive and test again.
When you can boot off any drive that the BIOS will let you 
boot from, and be able to change non boot drives with out 
problems, then you can consider GRUB ready.





More information about the test mailing list