Boot problem after Kernel upgrade...

[Mike Cisar]

I'm just wondering if anyone else has experienced a similar problem and/or
if anyone might be able to think of a less "manual" workaround than the one
I present toward the bottom of this email.

I did the upgrade to the kernel-2.4.22-1.2129 (and associated packages) this
morning and encountered a bit of a problem when rebooting the system after
the install.  When the system gets to the point of trying to fsck the root
partition it prints the error  "/dev/md1 is mounted. e2fsck cannot continue,
aborting" and then drops to the system maintenance prompt.  Booting back to
the older kernel works fine.  

I have had this problem in the past with a few Redhat 9 systems in the past
as well (and in retrospect probably Redhat 8 as well, but I didn't know it
then :-) and have made the following observations...
- doesn't on a new installation
- doesn't on a system upgrade (ie. RH8 to RH9, RH9 to Fedora1, etc)
- does happen the first time you do a kernel upgrade

The systems on which I have had this problem have had the following in
common...
- software raid with more than one partition
- disk quota enabled
- / partition is entered in fstab with the options
exec,dev,suid,rw,usrquota 

I have found that the workaround to this problem is to change the
"exec,dev,suid,rw,usrquota" entry in fstab to "defaults", then run mkinitrd
to generate a new initrd and finally set the fstab back to what it was
before.  After this process everything works just fine (until you do the
next kernel upgrade).

Obviously it's pretty easy to forget to regenerate an initrd using this
process after a kernel upgrade and come back hour s later to find your
computer sitting at a "enter root password or Ctrl-D" prompt.  Is there any
other workaround/solution that will let the kernel install process generate
a working initrd instead of having to do it by hand afterwards?

Thanks,
>>>>> Mike <<<<<