Kevin McKinley wrote: > Andy wrote: > > advises one to do the following: > > "Be sure you say "no" when the script asks if you want > > to install a boot block using the existing /etc/lilo.conf." > > The short answer is "because when I said 'Yes' it didn't work". :) I always answer yes to that question and it always works for me. If you said no there you would need to run lilo manually after the installation. > A longer answer is that writing a new boot block with the existing lilo.conf > would simply re-write the current boot sector, which would boot the old > kernel. Are we talking about the same question? At the point that the question is asked if lilo should be run the new kernel is already installed, the symlinks pointing to it, and running lilo will set up the boot to go to the new kernel. I guess it all depends on what is in your existing lilo.conf file. But if you are just doing the normal default Debian installed lilo.conf file it should be fine. I have noticed a bug in the kernel-image*.postinst script which does not always get the /vmlinuz et al symlinks correct, leaving them pointing to the old kernel. I tried debugging that problem but have not closed to root cause yet so I have not reported it yet. I think perhaps it failed to update your symlinks when you had the failure. In that case you do need to correct the symlinks and run lilo again. I only see this when I am going forward and backward while hacking a new kernel. It normally works when just moving forward. > What you want to do is install a new boot block which refers to the new > kernel you just compiled. Which should be the default. Bob P.S. Let me second the review that your 'newbiedoc' is excellent.
Attachment:
pgprB2OMNGQLf.pgp
Description: PGP signature