Help needed: Not able to recover RAID1 on TS-253Pro after Power Outage

[Blackbird72]

Hello,

my NAS TS-253 Pro worked fine the last 7 years. And also Hot-plug worked when a disk failed 1.5 years ago without any issue.
Configuration: HD1: 3TB, HD2: 6TB Seagate (only 3TB used in RAID1)

What happend:
Last week we had an announced power outage, but the NAS was still shutting down, when the power was shut off.
After that the system had the expected error messages: that the system was not shut down properly and the file system is not clean.
BUT: Disk 2 was recognized as having 16TB (instead of 6TB).

I removed disk 2, put in an external USB and found it (in windows) having 6 TB, but unkown partitions, nothing to edit or delete (even with comand line tools). Also no chance to correct the disc connected to the NAS via USB...
With SeaTools I was able to access the drive (when connected directly to the controller, without USB) by a normal PC. All checks were positive (SMART, Random Read/Write etc.). I did a full erase and everything was fine.
Reinstalled to the NAS the disk is identified as 6 TB again but RAID rebuilt fails: I receive the error message:
Fehler 2022-08-10 20:32:18 admin 127.0.0.1 Storage & Snapshots RAID Group [Storage & Snapshots] Failed to assign RAID spare disk "Host: Disk 2".
(I tried this multiple times now)

All checks (Smart, IronWolf Smart, BadBlocks, test etc. are finished without errors). State of the disc is "Good" (Green).
HD2 is shown as 6TB "free space".

What irritating me:
RAID group 1: Seems to consist of HD1 and HD1 (again?) - the HD is mentioned twice.

As I had issues to delete the HD2 (6TB) I tried - before I opend the PC and used the SeaTools to delete -
to swap the drives (shut off system put HD1 3TB in slot 2, booting was fine, but rebuilt with hot plugged HD2-6TB in slot 1 started but did fail...). Maybe this was the time, when the HD1 was entered the second time to the RAID configuration - but I am not sure.

Is it normal, that HD1 is mentioned twice, if the second HD is not part of the RAID?
Is there a possibility two remove the "second time" HD1?
Might this solve my issue of rebuiling RAID 1?

How do I access the raid.conf (text configuration)?
I am not a Linux pro, but experienced enough two work with console, etc. if I know how to access it (Currently the damned graphical user interface does not show me a terminal or console) Is there a key-shortcut or do I need to install something?

Thanks for your help!!!

PS: My backup is current - I want to avoid to delete everything and to re-configure all the settings like admissions for multiple users, multiple shares,...
PPS: I use the German user interface - maybe some of my translations are not standard

[dolbyman]

1. Get a UPS asap.. NAS do not like powercuts .. as you just have noticed .. a very valuable investment
2. See this reply ... your NAS is a CAT2 and internal drives are not readable at all without expensive tools or special knowlage
3. A RAID "group" of two disks could be RAID1 (recoverable) or RAID0 (busted on single failure) .. so more info is needed
4. The 'Fehler' in your Error message sounds like it's set to German .. so this would be best posted here http://forum.qnapclub.de
5. Always attach Screenshots if you can (but if you decide to stay in an English forum .. make sure the UI is English!)
6. SSH is accessed via .. SSH, Use a separate client or built into whatever OS you are using (Windows/Mac/Linux) , worst case you have to traverse a SSH console that is default activated for a couple of years (you can disable that as well)

[JCNAP]

I'm not writing for writings sake, but I can't agree more with number 1.

My NAS was catastrophic after every power outage and I'm expecting more outages considering the outlook we have for 2023... especially if you live in the UK.
I'm Not an expert.

But my QNAP failed three times while having a new UPS because 'the NAS didn't pull enough Wattage'... so the UPS shut down... It didn't make sense to me but that was CyperPower's explanation - the company I bought the UPS from. it happened three times.

CyberPower is the brand. And it was a compatible model. But did not have Sinewave.

Pick an UPS that has SineWave.


CyberPower advised I upgrade to the CP900 with sinewave. I've been running smooth ever since.


.... sounds like I'm selling it. Not a salesman either.

[OneCD]

But my QNAP failed three times while having a new UPS because 'the NAS didn't pull enough Wattage'...

What size (capacity) UPS was this?

[dolbyman]

My 1288X is on this one (bought @ Costco)

https://www.apc.com/us/en/product/BN150 ... s-4-surge/

gives that large NAS 45 minutes bridge time .. no problem (had a power outage over the weekend) with the sine approximation

My 951X and 419p+ share this

https://www.apc.com/us/en/product/BE550 ... s-4-surge/

Bought it long time ago at staples on a sale .. changed the battery pack twice already .. still great (also just sine approximation)

So there is no need to spend tons of money.. pretty sure that there is plenty of comparable NAS models in Europe as well.

[JCNAP]

But my QNAP failed three times while having a new UPS because 'the NAS didn't pull enough Wattage'...

What size (capacity) UPS was this?

It was 700VA... I've upped it to 900VA.
I doubt most home NAS need even half of that.

[JCNAP]

My 951X and 419p+ share this

https://www.apc.com/us/en/product/BE550 ... s-4-surge/

I should have gone for something like this.

[OneCD]

It was 700VA... I've upped it to 900VA.

It’s odd that increasing the size fixed the issue. I guess the larger size model has better load detection.

I use a 1500VA APC, but with lots of gear connected to it. Only runs at about 20% capacity when everything is powered on at the same time.

[JCNAP]

It was 700VA... I've upped it to 900VA.

......I guess the larger size model has better load detection.

That must be it... nothing to do with the VA or the Sinewave.... contradicting myself from my previous post... but it makes sense.

[Blackbird72]

Thanks for the feedback and hints.

To 1. Even though power outages are seldom in this region (third time in ten years and two of them were announced previously) I will get a USP - have one sitting in the office of my wife wich is not used anymore - just have to check compatibility. But this does not solve my current problem.
To 2: I assume there is a missunderstanding - I did/do not want to access the drives from the windows PC when I took it out of the NAS - I just wanted to completely delete it - including removal of all partitions. This worked fine with the SeaTools.
To 3: The RAID was configured to mirroring. RAID1.
To 4&5: Correct - I am German and User interface was configured to German - changed now to English.
To 5: Some Screenshots:
- Health Status of disk 1
- Health Status of disk 2
- RAID Group showing as member of the RAID1, two times disk 1
- Error Log (with red numbers indicating the following steps) of
1. removing the disk 2 from the NAS
2. Deleting the disk 2 with SeaTools on an other PC
3. Re-Entering the disk 2 to the NAS
4. The NAS automatically adding the "free space" (but not adding it back again to the RAID1)
5. The NAS failing to add the disk 2 to RAID 1 after manual start of Rebuild RAID by admin.
To 6: I will read into SSH tomorrow - did not use it for more than a year.
I was somehow blocked in mind - trying to get access via the web-fronted to the internal configuration files of the NAS... -> So thanks for the hint.

1. Get a UPS asap.. NAS do not like powercuts .. as you just have noticed .. a very valuable investment
2. See this reply ... your NAS is a CAT2 and internal drives are not readable at all without expensive tools or special knowlage
3. A RAID "group" of two disks could be RAID1 (recoverable) or RAID0 (busted on single failure) .. so more info is needed
4. The 'Fehler' in your Error message sounds like it's set to German .. so this would be best posted here http://forum.qnapclub.de
5. Always attach Screenshots if you can (but if you decide to stay in an English forum .. make sure the UI is English!)
6. SSH is accessed via .. SSH, Use a separate client or built into whatever OS you are using (Windows/Mac/Linux) , worst case you have to traverse a SSH console that is default activated for a couple of years (you can disable that as well)

Can the "double disk membership" of disk 1 to RAID1 be the cause of my issue?

[dolbyman]

you can always use diskpark in windows (option clean) to clean a disk of all QNAP partitions

Just to reiterate, your goal is to start the NAS from scratch ? So take out both disks (2TB and 6TB) clean them and start the NAS from scratch (the OS is on the disks, so without disks, nothing will be left).

If you want to keep your settings, you could remove the storage volume (or pool) and keep the first disks intact, then just clean the second disk again and create a new pool/volume on the two disks (apps will mostly need to be reinstalled as well)

[Blackbird72]

My target is TO AVOID starting from scratch
- Disk 1 (3TB) is working fine and I would like to keep it that way. I would like to keep the data, the file shares, the user configuration and so on...
But safety is gone - as disk 2 (6TB) is not connected to the RAID1 anymore.

I tried to use the hot swap functionallity to get disk 2 (after cleaning it) back into the RAID1 of volume 1. This does not work as I expect it.
It feels like an error 99: "change user" (or admin of the NAS in this case).

To reiterate it: I am trying to avoid a new set up at all costs because the effort would be high.

Question: Deleting the volume and setting up a new one also deletes all the file-shares created, correct?
PS: This is what will take days to re-share and then re-connect on PCs, on network printers scanning to file-shares, ...

PPS: My back up strategy is file based (windows xcopy of file shares to external-hd-drive which is disconnected after back up and stored in an other building - to be safe) - so all settings will be lost if I need the back up...
...but all the important files will be there...

[dolbyman]

Deleting the volume would remove all data volumes (not the shares though, so shares can be edited later to point to the correct spot on the new volume again, this way assigned user/group permissions can stay in place)

How many local users and file shares does this NAS have that it would take days to re-setup ? (Hope the NAS is not web exposed for anything)

[Blackbird72]

If you want to keep your settings, you could remove the storage volume (or pool) and keep the first disks intact, then just clean the second disk again and create a new pool/volume on the two disks (apps will mostly need to be reinstalled as well)

I am not sure if I get the point here correctly. To re-iterate:
- If I remove the volume the data will be gone (I would use the back up to fill it again).
- By "keep the first disk intact" you mean I do not remove it from the bay and do not delete all partitions. So the operating system and (most) settings are still available.
- Disk 2 needs to be removed from the bay again and "deleted", put back in the bay afterwards. I would create a new static volume with RAID1 (and hopefully be able to use both disks again)
Correct?

It is still strange to me that disk 1 is listed two times in the datavol1 management (see attachment). As "NAS Host: Disk 1" and a second time as "NAS Host: Disk 1 (Spare)". I expect to be able to rebuilt the RAID after removing this second "NAS Host: Disk 1 (Spare)" - but I do not know how to do it.
Is this entry normal, if the RAID is degraded?
If not, is there a method to get rid of this "double connection" to the same disk?

PS: I did not only try to Rebuilt the RAID1 manually. I also tried to "trick" the system by setting disk 2 as an "enclosure spare" - which also faild:
Error 2022-08-11 22:46:52 admin 127.0.0.1 Storage & Snapshots RAID Group [Storage & Snapshots] Failed to assign RAID spare disk "Host: Disk 2".
Can this be, because disk1 is also nominated as spare?

[dolbyman]

I am not sure if I get the point here correctly. To re-iterate:
- If I remove the volume the data will be gone (I would use the back up to fill it again).
- By "keep the first disk intact" you mean I do not remove it from the bay and do not delete all partitions. So the operating system and (most) settings are still available.
- Disk 2 needs to be removed from the bay again and "deleted", put back in the bay afterwards. I would create a new static volume with RAID1 (and hopefully be able to use both disks again)
Correct?

Correct

It is still strange to me that disk 1 is listed two times in the datavol1 management (see attachment). As "NAS Host: Disk 1" and a second time as "NAS Host: Disk 1 (Spare)". I expect to be able to rebuilt the RAID after removing this second "NAS Host: Disk 1 (Spare)" - but I do not know how to do it.
Is this entry normal, if the RAID is degraded?
If not, is there a method to get rid of this "double connection" to the same disk?

Something weird must have happened .. maybe some corrupted entry in the raid config

Whatever you tried could have caused this issue