Restores gone awry
-
We have 4 firewalls that have been taken out of production. The intent is to restore a basic config on them and then upgrade to the latest version and hold on to them for spares. Apparently that worked fine on the first 2 units. The second 2 units had the config backups restored and now the units won't boot. At all. I'm not the one doing this so I'm not sure what exact steps were taken. I'm told all they did was go into backup/restore and restore the config. These boxes have worked perfectly for years with no glitches but the hardware needed to be upgraded. Upon booting we now get:
Press F10 key now for boot menu Booting from Hard Disk... \ Can't find /boot/zfsloader / Can't find /boot/loader \ Can't find /boot/kernel/kernel FreeBSD/x86 boot Default: zroot/ROOT/default:/boot/kernel/kernel boot:I don't know how a restore of the config.xml could cause something like this, especially on 2 of them, so I'm open to suggestions. At this point we will just wipe and restore but I need to make sure I understand what happened and what can be done in case we see this in the future. Thanks!
-
@stewart Restoring just uploads the config file and reboots. Booting isn't part of the config.
Bad drive? How old are the PCs?
-
@steveits These are APU2 units. Probably 3 years old. I know a restore just puts the xml back into place but this happened on 2 units back to back. I find it hard to believe both drives crapped out together, you know? Something must have happened but I've no clue. Now I'm trying to figure out what was affected so I can determine what was done. So far I've created a bootable USB and can get into the recovery on it.
In df I can see listings for /zroot/var, zroot/tmp, zroot, and zroot/ROOT/default. Each of them show Size=106G and Avail=106G. They are essentially all mounted under /tmp/mnt_recovery in different folders. If I navigate to /tmp/mnt_recovery and run the entire folder is only 1.8M in size. It seems like the whole drive was wiped and I'm trying to determine how this was done. All they are telling me is they went into the Backup&Restore, selected the xml, and clicked Restore. If they did something other than that and managed to wipe the drive I need to know how they did it so I can stop it from happening again. I'm lucky this is done on units being pulled and not on production units.