The first thing I did was to set up a userid via the web interface. I then shut the MB down, removed the drives and inserted them into my existing Fedora 11 system.
My system SATA ports are hot plugable. Is there any problem inserting the drives into a running system, or should I shut down first.
- I fumbled around with mdadm(8). While I am familiar with LVM, I am not familiar with mdadm. In any case, I ended up doing a 'mdadm —scan —auto-detect —detail' and it recognized the 4 RAID1 partitions.
- I then mounted the first partition (eg. mount /dev/md124 /media/mybook).
- I then used vi to edit /media/mybook/etc/shadow to clear out the root password.
- I then added telnet to /media/mybook/etc/inetd.conf.
- I then unmounted the array (umount /media/mybook).
- What I didn't do was to stop mdadm (eg mdadm -S).
- I then placed the drives back into the MB in the wrong order (eg. B into the A slot and A into the B slot).
- At this point, when I powered on the WD it failed to boot, and I discovered by error.
- I then placed the drives in their correct slots, but was not able to boot.
At this point, what I want to find out is other than a hardware failure, what steps did I take (or fail to take) that could have bricked the unit so I can do it properly when I get the unit back. I'm planning to place the unit in a rack in a computer room to use as a backup server for Linux systems (strictly rsync), so I need to hack the root password. Another thing I probably should have done was to enable sshd through the web interface first, so all I really need to do is to zero out the root password.