I have a single hard drive in my Debian Etch server. On install, I selected to use LVM with a separate home directory. I tried to add a second internal drive to the server today, and on reboot, it wouldn't boot up. I get an error message saying mail is the hostname of this server. It also says that a uuid device can't be found. How do I get this machine to boot properly? If that isn't possible, how do I get a backup of the /var/www directory before I do a fresh install of Etch? I don't have a backup of the web directory, yet...that's what the second drive was for!
After booting to a recovery CD, I started ssh and tried to backup my websites. Nothing was there. /var/www did not exist! So, I did a fresh install of Etch...but, how to avoid this from happening again when I try to install the second hard drive?
I hadn't even gotten that far. All I did was plug in the drive and turn the machine back on. That's when it wouldn't boot. I'll check the how-to section more throughly before I try again to plug in the second drive.
No I didn't. I actually hadn't found that how-to. Now that I've re-installed (after the problem of no /dev/mapper/mail-root and no websites)...I did the Etch install with LVM configured with separate home directory. Now, when I did the re-install, I had the second drive plugged in. Now, after the install, the drive doesn't show up on the system Disk 1 (boot disk) is 120 GB Disk 2 (second drive...not showing) is 80 GB Any ideas how I should proceed next? Falko, thanks for everything!
If it doesn't show up, you should control alll plugs again. If that doesn't work, it might be a driver problem. What kind of HDD is this?
It's a Western Digital drive. If I pull the plugs and plug them back in, what are the odds of the original problem coming back? I already would have done that, but was trying to not run the threat of having the same problem.
I tried to re-seet the plugs, and found the problem. The IDE cable connection came apart in my hands! That might explain why it didn't work. I got a new cable and just went through the back-up how-to, and everything went GREAT. Thanks, Falko. Any chance of a how-to to set this up as a cron with a rotation of the backups to keep 1 or 2 of the backups around? Thanks again!
You mean something like this? http://www.howtoforge.com/linux_lvm_snapshots You could use a find script like this to rotate the backups: Code: #!/bin/sh for file in "$( /usr/bin/find /home/somedir -type f -mtime +2 )" do /bin/rm -f $file done This will delete all file older than two days, for example.