I understand that installing is your focus now. But do you have a concept now?
See my post above.
On a dual HDD NAS can I use the 2nd HDD as the backup destination of the 1st HDD on Odroid HC4?
-
- gelöst
- OMV 5.x
- viewmo
-
-
it won't matter either way.
I took your advice and I've decided to run with Debian Odroid netboot. OMV is more deeply intergrated into the operating system than I thought (system updates are handled by OMV now, I think.)
Those tweaks to fstab (noatime, commenting out swap etc.) from 'Getting_Started-OMV5.pdf' should go along way to prolonging the microSD card.
Now I'm trying to figure my way round OMV. Very nice web interface.
The Neboot Installer for ODROID-HC4 is the version I've customized
Thanks for the great work on the Debian netboot installer. I didn't realize it was your work. And thanks for the videos too.
and http://ppa.linuxfactory.or.kr which is mine.
Hope you don't mind me asking, but I presume you work for Hardkernel? Or are the N2/C4/HC4 Debian netboots personal projects of yours?
But do you have a concept now?
See my post above.
Once I went for the HC4 I had made my mind up to have both backup and original drives on the single HC4 device. I'll just have to take my chance and hope nothing catastrophic happens that would take out both drives.
It's not the perfect backup solution but I guess "perfect is the enemy of the good" for me on this. Hopefully it will be good enough.
Thanks for your initial post. I did read it. It gives a good run through of a backup setup.
-
Now I'm trying to figure my way round OMV. Very nice web interface.
You're up! Great. We (on the forum) appreciate it when someone adds to the user knowledge base. You've done that with the Odriod HC4. (If you will, please change the title of this thread to include Odriod HC4. Thanks)
Don't forget to clone your SD-card at significant points, for a guaranteed fallback position, as you configure your server. As mentioned before, an SD-card cloning process is in the User Guide, under OS backup.
And there's a drive-to-drive backup "walk through" based on Rsync, under Full Disk Mirroring / Backup with Rsync. It will create a functional mirror of the source drive. It's easy to set up and you won't have to leave the GUI to do it. If you have a data drive catastrophe, at some point, the backup is also dirt simple to restore. -
You're up!
I guess I am...
We (on the forum) appreciate it when someone adds the user knowledge base.
No problem gald to help, even if its only a very small way. Thank you guys for all your help.
an SD-card cloning process is in the User Guide, under OS backup.
I haven't read that part yet but I have come across a great tip for cloning microSD cards: fill the empty space zeros using /dev/zero.
Then when you create the backup image you can compress it with xz and the backup will go from 16GB to (in my specific case) to 500MB !!!
Same image. Debain and OMV and all the stuff is still there. Just the empty space has been squashed to almost nothing.
And Gnome Disks doesn't need the image to be decompressed. It can work with .img.xz files directly when re writing to a card. I was well pleased with that tip when I discovered it.
I haven't even set up the drives yet. I'm planning on using LUKS on them . I have to learn about the LUKS plug in.
I changed thread title. Hope that's what you meant. Let me know and I'll edit it if required.
-
viewmo
Hat den Titel des Themas von „On a dual HDD NAS can I use the 2nd HDD as the backup destination of the 1st HDD?“ zu „On a dual HDD NAS can I use the 2nd HDD as the backup destination of the 1st HDD on Odroid HC4?“ geändert. -
Once I went for the HC4 I had made my mind up to have both backup and original drives on the single HC4 device. I'll just have to take my chance and hope nothing catastrophic happens that would take out both drives.
It's not the perfect backup solution but I guess "perfect is the enemy of the good" for me on this. Hopefully it will be good enough.
Thanks for your initial post. I did read it. It gives a good run through of a backup setup.
Well, the only thing you are protecting against with your current setup is:
a) One single HDD failure
b) An accidental delete that you notice very quickly
Scenarios that you do not protect against:
1) power surge (e.g. lightning strike)
2) accidental delete that you notice a year later
3) Virus/Malware
4) many others
What you could improve:
For 2) do versioned backups (borg-backup, rsnapshot)
For 3) ensure that only one drive is accessible from other machines. Use different credentials for your linux machine in combination with (2)
You could also later add an external drive and/or an offsite backup.
But if your data is valuable, I would not say that what you do today can be deemed "good" (in the sense of the enemy of the better). It is close to no protection, in my view.
Regards,
Hendrik
-
I agree with henfri that you should really consider versioned backups. Also an external HDD which you just update occasionally (and maybe store offsite) would preserve most of your data.
To protect against malware also btrfs with read-only snapshots could be used. But btrfs is a whole chapter itself and should only be used when you are ready to experiment and get familiar with...
-
I changed thread title. Hope that's what you meant.
Yep. Good. Now it will be easier to search.
-
Well, the only thing you are protecting against...
a) Is what I set out to do. If I could afford it I'd go with a better backup solution, but a) is pretty much what my budget and bandwith covers for now.
Does 'rsnapshot' not help against b)? I was planning on using it.
Power surge: What is that? I was going to ask earlier. I see it mentioned in US based discussions quite a bit. I've never heard it mentioned, nor any one who suffered loss from it over here.By lightning strike do mean the consequential power failure is the issue or a dramatic surge in electrical energy through the house wiring?
Would the fuse system for the building not protect any electrical energy surge? If not, surely a house fire would be inevitable?
Electrical storms in Ireland aren't nearly as common nor severe as in the States and elsewhere. I think the chance of it happening are as likely as being hit by... well, lightning.
3) Malware for Linux is much less common. The HC4 isn't connected to the internet. I'm pretty cautious about what software I run. Hopefully the chances of that are low too.
I hear what your saying: there are better backup solutions. And for anyone else planning their systems and reading this your advice is great.
But I have my constraints. I weighed up my choices and for now at least I'll be running the HC4 dual HDD backup. And I'm happy with it.You could also later add an external drive and/or an offsite backup.
Down the road I might just do that.
Hell, I felt pretty pleased with myself that I had even considered and planned for backups right for the start.
I guess I'm playing with the big kids here. Standards are higher.
To protect against malware also btrfs...
I really like the idea of btrfs (from what little I know about it).
Is it ready for prime time yet? I thought it still had issues. You really don't want issues with your filesystem. I don't think I'll be experimenting with it. Progress seems so slow on it. I thought Facebook were developing it. With all their resources you'd think it'd be done by now. -
viewmo
Hat das Label OMV 5.x hinzugefügt. -
Power surge: What is that? I was going to ask earlier. I see it mentioned in US based discussions quite a bit. I've never heard it mentioned, nor any one who suffered loss from it over here.
The vast majority of issues with consumer electronics are power related problems caused by the AC power line. While you may not notice it or see it, there are high voltage high frequency spikes coming in on your power line. It's not enough to trip a breaker or blow a fuse but more than enough to damage electronics.
Today's electronics (mostly CMOS) is very sensitive to high voltage spikes. If one of those spikes gets past the protection provided by your power supply, the resultant damage can create serious frustration. You see, CMOS can be damaged, but it doesn't always fail completely. A spike can cause an arc over, blowing a CMOS gate. Thereafter the circuit may still work and test out fine, but problems may occur intermittently. (That's the worst of all possibilities, but it happens all the time. I'd much rather rather have a complete failure.)
Anyone can become a victim of power line issues. It doesn't matter where you are or how reliable your local power is. If a truck hits a utility pole, lighting strikes a power line even several miles away, a sub-station switches lines, a transformer blows, etc.; these events can cause high voltage very short duration transients on the line.
The best protection is an UPS. The second choice would be a good surge suppressing power strip.
Myself, I have whole house surge suppression, mounted in my power panel, AND I use an UPS.
-
Is it ready for prime time yet?
Well that depends on perspective I would say. RAID 5/6 is still officially marked as unstable. On the other hand some major Linux distributions (openSUSE and Fedora) use it as their primary filesystem. So there seems to be confidence that it is stable enough. I myself use it on all my systems and did not have a problem yet (for what that counts...).
-
Is it ready for prime time yet? I thought it still had issues. You really don't want issues with your filesystem. I don't think I'll be experimenting with it. Progress seems so slow on it. I thought Facebook were developing it. With all their resources you'd think it'd be done by now.
What I've noticed, on this forum, is that when a user has a BTRFS problem and posts it, it's not uncommon to see it go unanswered.
I've had my own "not so good" experiences with BTRFS, that were related to sudden power losses. The FS went "read only" more than once. (That was bad news for me. The application was portable where power losses were to be expected.) That was about 2 to 3 years ago. Whatever the issue was, it seemed to be fixed in later kernel releases. It's working fine now.
In any case, when I tried to use BTRFS utilities to correct the problem, in one instance I bricked the filesystem. While I might have done the wrong thing in the wrong way, the lesson learned was FS utilities were not yet mature and little is known about repairing BTRFS, if needed. FS utilities seemed to be something of an after thought, to aid developers.
So, as it would be in any case, real backup (as in a second full and independent copy of your data store) is recommend. -
Hope you don't mind me asking, but I presume you work for Hardkernel? Or are the N2/C4/HC4 Debian netboots personal projects of yours?
Sorry for late response. I work for Hardkernel and the Netboot Installer is started as my personal project. It's unofficial support by Hardkernel but the features in the Netboot Installer and packages on the OS setup with Netboot Installer are pretty much for ODROID only.
-
So now the guinea pig, sorry tester is gonna be hit by electricity spikes, lightening, his house could burn down, prior to that he could get a virus or be zapped by malware, hopefully he won't go out and get hit by bus.
This is a new piece of hardware, the guy's on a budget, does it work, is is stable, is it good alternative SBC, it certainly looks it.
-
So now the guinea pig, sorry tester is gonna be hit by electricity spikes, lightening, his house could burn down, prior to that he could get a virus or be zapped by malware, hopefully he won't go out and get hit by bus.
Or, maybe COVID mutates and attacks Armbian! (gasp!....)
-
You see, CMOS can be damaged, but it doesn't always fail completely.
So the result a user might see might see would be unusual hardware behavior but without ever knowing what the cause was? Damn.
It doesn't matter where you are or how reliable your local power is.
I honestly thought it was dues to crappy US electrical utility companies with crappy standards (rolling black outs in California, Enron etc.) I need to research power surge protection some more.
I have whole house surge suppression, mounted in my power panel
I was hoping this was standard practice. I guess not.
some major Linux distributions (openSUSE and Fedora) use it as their primary filesystem.
Wow, didn't know that. That's quite encouraging, progress is being made. I still think I'll be leaving it to mature for a while longer.
I myself use it on all my systems and did not have a problem yet (for what that counts...).
Always good to hear real world experiences. Thanks.
Sorry for late response.
No need to apologize, you're under no obligation. I'm sure you're plenty busy. Thanks for your Debian netboot project. It got me out of a jam.
This might be a bit cheeky (feel free to ignore) but I don't suppose you publish the source code for your project?
... hit by electricity spikes, lightening, his house could burn down, ...get a virus or be zapped by malware ...get hit by bus.
Or, maybe COVID mutates and attacks Armbian! (gasp!....)
Only need a plague of locusts and I'll have my own personal apocalypse.
-
I need to research power surge protection some more.
Actually crashtest point is relevant, all my PC's, TV's, Home Theatre are on anti surge plugs, or plugged into anti surge adaptors, but as to using a full blown anti surge on your fuse box most sparks would say it's not necessary this side of the pond.
-
So the result a user might see might see would be unusual hardware behavior but without ever knowing what the cause was? Damn.
I agree! Damn!... I've seen this a few times in my lifetime, with items I've owned. There's no repairing most of them. They have to be thrown out.
_______________________________________________
I got a couple primers on this issue way back in the day. After CMOS became mainstream (replacing all TTL) and was being shipped in regular plastic envelopes and stuck into styrofoam (not static protected), whole batches of components were mysteriously failing.
CMOS works by using voltage to generate electrostatic fields, to turn off a stage or to forward bias it. Since there's no current flow to speak of, just the generation of a field using voltage, CMOS drastically reduces power consumption when compared to TTL. That's the up side. Now for the down side.
To find out what was happening with the failed components, NASA did a "micro grinding" operation on a few CMOS gates. They put them under an electron microscope and found that the gate was blown. That means the gate's insulator had been arc'ed through and, in the process, very tiny particles of the metal oxide used in the gate where randomly distributed within the insulator hole.
(See the poor artist rendition below.)
With tiny metal particles sprayed through the insulator, the gate is now far more susceptible to arc over under normal operating voltages and conditions, but it may still work. Some fail completely (shorted out). Some work as if nothing had happened. Others may work fine for awhile and fail. Still others may work but act up in an intermittent and bizarre manner. The last is the worst of all possibilities and, all to often, very common.
Lately, protections from power line spikes have been designed into power supplies but that protect is limited. Since you'll be using a power brick (they're not known for being excellent power supplies) a good surge suppression power strip is a real good idea.
My preference for panel mounted surge protection is due to the presence of electronic controls and processor brain boxes in nearly everything. Even refrigerators and washing machines have electronics these days. If protection is in your panel, everything is protected, TV's, stereo, etc.
Now, I feel that I should caution you - surge suppression won't save you from a maniacal bus driver,, or a plague of locusts.
__________________________________________________I honestly thought it was dues to crappy US electrical utility companies with crappy standards (rolling black outs in California, Enron etc.) I need to research power surge protection some more.
In the US, crap utility companies can be a consideration but that greatly depends on the location. California? They have a lot more going on than what's in the news. Their rolling outages is just a "symptom" of their real (childish) problems.
-
I honestly thought it was dues to crappy US electrical utility companies with crappy standards (rolling black outs in California, Enron etc.) I need to research power surge protection some more.
no, you need a Backup.
Surge protection can help, but there is no guarantee.
-
What a coincidence:
-
Well lock down is over and I just went out and bought three surge protected extension leads.
Thanks for the heads up on power surges.
Is it normal that OMV removes the ordinary user's SSH login abilities? When I set up Debian I was able to SSH in as an ordinary user. Now I can only ssh in as root?
Jetzt mitmachen!
Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!