I hit a kernel panic after the upgrade.
The fix was easy enough for me to find and correct, out of space in the efi part due to windows dual boot stuff in there too. I removed the old kernel rebuilt and I was good.
I wo der how many people hit that and were stumped.
My 2 concerns to date are
1 just a bad file manager. Its dumbed down for people and I hate it.
2 lack of support for veracrypt volumes.
First, thanks. And no, I haven’t tried their support yet, because the situation hadn’t become urgent enough yet.
I can restore an image, but that doesn’t let me test, unless I restore to yet another external drive with enough space, then chroot to it, I guess. I didn’t have another 1 TB drive at hand to try.
I didn’t try to boot to the clone, because I read a warning about booting with two different connected drives having the same UUID. I wouldn’t use my daily drive laptop to try that out in case something serious went wrong. All I found online were vague warnings and I wasn’t prepared to ignore them. Also, I know software, and I could imagine some potentially serious errors due to optimistic assumptions in software about the uniqueness of drive UUIDs.
I can’t disable my laptop’s internal drive through BIOS and I’m not prepared to open the machine to physically disconnect the drive.
That’s where I am now. It looks like the best option to test would be to backup to an image on USB drive A, then restore to USB drive B, then boot to a live installation and chroot. I mean… OK, but that’s not what I hoped for from a so-called bootable backup.
Am I missing something obvious? Am I making a clearly invalid assumption?
The real problem here is that you are not trusting the backup tool you selected. You are right to be paranoid of course…I am. Testing the backups is something we all should do on a schedule.
You don’t want to restore the entire image for fear it won’t work, and currently have no other option to test it.
Leap of faith with your untested restore is not adviseable either.
Ugh. Either you need to find another machine with enough space to teat the restore 1st, disconnect the drive and mount,… Don’t know how you could proceed otherwise.
Thank you. Of the folks who have engaged me over the past few years trying to find an adequate backup system, you’re the first to tell me that I’m indeed not missing anything and that this cannot be made to work the way I’m expecting. I find that comforting.
Now I have a way to move forward: to back up a 1 TB drive, I can partition a 2 TB drive in half, back up to an image on one, restore that image to the other, chroot, and check that that worked.
Wasteful, but workable.
Thank you.
You’re welcome. Like many, I dumpster dive for a lot of my equip. If you van getcyohr hands on almost any other machine with large enough drive, that would be good enough to test your clonezilla restore.
On my daily driver, I use nextcloud for document/photo sync so that I immediately have 2 copies. Then a weekly backup to external storage 1 time per week of my home folder.
Remember to test the backups every so often. Make sure you keep a copy of data offsite too(fire,theft,etc)
And its morbid, but I left instructions for my kids on how to gain access to backups in case I’m gone so they can keep pics and bids from my stash.
Indeed, I have a lot of redundancy for critical documents, including a restic backup to Blackblaze and many many git repositories. What I don’t have is the “Get back to work quickly when the internal drive dies” that SuperDuper gave me on my Mac. That’s the missing piece I’d like.
And, as you noticed, since I was struggling with testing the Clonezilla backup, I couldn’t possibly trust it. Now, at least, I have a plan.
And yes, I have onsite backups, offsite backups, and a succession plan.
Oh yes, I tried testing my Clonezilla restore on an older laptop, but after logging in, the keyboard and trackpad stopped working. I wondered if that was a hardware incompatibility problem, since the keyboard and trackpad worked fine at the greeter screen. 🤷