dolbyman wrote:linux md is well documented .. and if you use static volumes instead of pools you also take lvm out of the equation
Interesting... a few hours ago you said you didn't understand why suggesting basic RAID-1 was a "helpful tip." Seems you have answered your own question. Avoid the LVM/storage pools, you can often get your data back relatively easily with mdadm. And with RAID-1, you can do it with just 1 surviving disk, instead of trying to hook up 3 or more with RAID-5.
where did I mention raid1 here ?
linux md is used for all raid types
no idea why we are turning in circles here...raid is not a backup...raid recovery should never be part of your recovery strategy ...keep a backup of your data in a separate location (bankvault ..friends house)
everything else is splitting hairs for argument sake
fredrogers wrote:who is trying to be responsible with his/her data by purchasing a NAS, without realizing that the default or recommended configuration actually puts his/her data at significant risk of loss
That is just plain bull **.
How is it bull? The original poster would be able to get his data back if he hadn't used the recommended storage pool feature which added LVM. Had he stayed away from LVM, he could almost surely get his data back with mdadm. I'm not the only one on this thread who has noted that.
MrVideo wrote:
fredrogers wrote:With that said, using your scenario, let's say a 6-bay unit stops working. Drives are fine, the NAS isn't. What *exactly* is your recovery scenario? Have you tested it? Have you actually recovered a failed RAID-5 or 6 drive array before? How long did it take? How much did it cost? Do you have all the equipment to hook up all the drives, in your home right now? Does your backup contain files you saved an hour ago? Yesterday? Two weeks ago? If your backup fails during recovery (not impossible, when restoring 10-20TB+ of data), what's the backup to the backup plan?
Wow, you sound like a defeatist.
And you sound like someone who doesn't want to admit his recovery plan isn't actually very well thought-out.
dolbyman wrote:linux md is well documented .. and if you use static volumes instead of pools you also take lvm out of the equation
Interesting... a few hours ago you said you didn't understand why suggesting basic RAID-1 was a "helpful tip." Seems you have answered your own question. Avoid the LVM/storage pools, you can often get your data back relatively easily with mdadm. And with RAID-1, you can do it with just 1 surviving disk, instead of trying to hook up 3 or more with RAID-5.
where did I mention raid1 here ?
linux md is used for all raid types
no idea why we are turning in circles here...raid is not a backup...raid recovery should never be part of your recovery strategy ...keep a backup of your data in a separate location (bankvault ..friends house)
everything else is splitting hairs for argument sake
If your backup is in a bank vault or friend's house, it is unlikely to contain your latest data. It is probably missing the last few days or weeks of data. Do you visit your bank every single day? Of course not. Just one reason why "backups" are not an ideal solution when a NAS fails.
Yes, let's get back to insulting him (like storageman) and reminding him how stupid he was for expecting his 2-month old file storage device to, you know, safely store his files!
dolbyman wrote:no idea why we are turning in circles here...raid is not a backup...raid recovery should never be part of your recovery strategy ...keep a backup of your data in a separate location (bankvault ..friends house)
I totally agree. fredroger's idea to only use RAID1, without LVM (storage pool) is just plain bull.
I *NEVER* said don't keep backups. Of course you should keep backups. Anyone who is reading this: make backups!
But the original poster doesn't have them. So the question is what to do now? All you did was tell the guy he was out of luck, and further, you said "A bad workman blames his tools, you should have kept backups." Yea, that is kicking a guy when he is down. How about just keeping your mouth shut if that is all you have to contribute.
Using RAID-1 and avoiding storage pools does not prevent anyone from making backups. But as I have stated many times here, and which you haven't disputed, is that a backup is likely to not be 100% perfect -- probably because it doesn't hold the very latest data, or it fails itself during recovery. With RAID-1 and no storage pool, YOU CAN STILL RECOVER. Even you yourself stated that.
I'm still trying to recover my data with a recovery software in Windows (EaseUS Data Recovery), it's very slow and do not recognize my directory structure, but at least I have a copy of my data.
As soon as it will finish, I will try your solutions. Thank you.
In Linux we got the following error (for both HDDs):
WARNING: Unrecognized segment type tier-thin-pool
LV tp1, segment 1 invalid: does not support flag ERROR_WHEN_FULL. for tier-thin-pool segment.
Internal error: LV segments corrupted in tp1.
I asked QNAP if I can send them the HDDs together with the NAS so they can fix also this issue, they told me (after 2-3 days) they don't care about the HDDs and they will accept just the NAS, even if the HDDs are corrupted cause of their NAS... Such a delusion with QNAP...
That 231P2 must still be under warranty.
Don't mess with those disks, wait for Qnap/reseller to send you back replacement/repaired unit, then try the drives in that.
Might be nothing wrong with the disks still , as long as all the tools you've tried haven't messed with the partitions.
I don't think you ever said at the start of this saga, how do you know the 231P2 is faulty? Did you test turning it on without drives? Does it actually boot?
One evening I found the NAS turned off by itself and after that it does not turn on anymore, obviously I've already tried without HDDs. The only signal of life is the LAN port's LED which turns on when connect to an Ethernet cable.
QNAP service told me in this case it's faulty and I have to send it back for repairment.
Yes anyway I guess I have to wait and test again after repairment, hoping HDDs will work again, with all my data.
