[sf-lug] Hayes Valley Mystery

Jason Turner jturner at nonzerosums.org
Fri Apr 18 10:27:03 PDT 2008

Asheesh Laroia wrote:
> On Wed, 16 Apr 2008, Tom Haddon wrote:
>> So the questions here are:
>> • How did the system boot from a device that it failed to mount (we know
>> it was booting from /dev/sda1 because the changes we'd made
>> to /boot/grub/menu.lst when we manually mounted /dev/sda1 before
>> rebooting were applied)?
> Sounds like it worked because GRUB's ext2/ext3 driver could read it; 
> Linux never had to, for it to boot.

Good gawd... how many times am I going to re-learn this fact?  I share 
my misunderstanding for the one or two people on the list that might 
share it.

GRUB/bootloaders in general(?) don't use /etc/fstab to locate files.  
The info they need is "hardwired" into their DNA -- the particulars of 
which and the degree to which that info is hardwired depend on the 
bootloader in question.  In light of that fact, Asheesh's answer looks 
perfectly logical.  Thanks, Asheesh.  Gimme another year or so though, 
and I may forget again...


More information about the sf-lug mailing list