Install the app
How to install the app on iOS

Follow along with the video below to see how to install our site as a web app on your home screen.

Note: This feature may not be available in some browsers.

DSM 7.2 Volume1 set as Read Only and then automatically repaired

As an Amazon Associate, we may earn commissions from qualifying purchases. Learn more...

Namaste

Subscriber
38
8
Hi everyone!

Something strange happened a few hours ago on my D920+. It started beeping, which is definitely not a good sign. I logged-in to check what was happening and found the four following notifications, published in that order and only a few seconds or minutes apart (translated from french, official wording may vary ;) ):
1) Container manager: Container PaperlessNGX-DB stopped unexpectedly. [Yes I run a Paperless stack with a DB]
2) Storage Group1 crashed and has been removed from the data cleaning task scheduling.
3) Volume1 has been set to Read Only ( :eek:)
4) Severity error message has been created.

My immediate action was to turn the device OFF, as I really had to go and couldn't investigate at that moment. Before turning it ON again I checked my Pushover app and Active Insight: I couldn't find anything, as if nothing happened.

After turning the NAS ON again, I could find these four notification, plus an additional one that said that "Volume1 was set to read only mode, but got automatically repaired and is now healthy." [I run a RAID5 array with an additional Hot Spare disk, I don't know if that helped to repair or not...]. That's all. Nothing else. The journals don't give any detailed information, all lights are green and everything seems normal in the storage manager, from what I can say at least. Each disk runs IronWolf health and complete S.M.A.R.T tests on a monthly basis: everything OK.

The only relevant action I could identify so far was to stop the Paperless stack. It is not critical to me. But damn I don't feel secure knowing something serious happened but having absolutely no clue, nor indication from the device itself about WHAT HAPPENED?

Has anyone an idea about what I could and should do to further investigate to ensure my system is healthy and will stay so?

The only thing I have to confess is that I put an additional "unofficial" 8Gb RAM memory into the device, but that was months ago and didn't cause any problem so far. Could that be the issue? What do you think about the notification sequence: could the paperless stack be guilty for such a problem?

Thanks a lot for your help and BR!
 
Could that be the issue?
Could be but then again it might not be. Running a memory test via Synology Assistant desktop app might give you some insight into that.

What do you think about the notification sequence: could the paperless stack be guilty for such a problem?
The sequence is valid. The fact that the app reported an issue 1st, then storage/volume makes sense. Also, if that is the only container your are running it makes sense that there was no other container calling out that it has an issue.

I run a RAID5 array with an additional Hot Spare disk, I don't know if that helped to repair or not
If one of your 4 drives is configured as hot spare, then the volume might have been degraded and the system kicked the faulty drive out and placed a hot spare in its place. Still, this will take time to rebuild, so how long did it take for the system to get back to "normal" status? Check what drives are in use and confirm that the hot spare is in use. Also, if this is true, then one of your initial 3 drives of that RAID5 should be flagged as problematic in some shape or form.

Each disk runs IronWolf health and complete S.M.A.R.T tests on a monthly basis: everything OK.
This doesn't mean that the drive can't fail immediately. Its unlikely but not impossible.
 
Hi, thank you for your answers!
Open Storage Manager. What does it show?
Well, yesterday after 1st reboot the system managed to repair the volume automatically and everything was back to normal. The Storage Manager did not show any warning or issue. All green and "healthy".

But this morning, after a new boot (my NAS goes OFF during nights and wakes up every morning) the warning appeared again and is now remaining. It has not been able to repair automatically a second time. Now Storage manager says this: (translation below the print screen)
Capture d’écran du 2025-01-14 17-40-36.webp

(Translation of the "Suggestion" section : "Volume 1 has been set as read only, probably because of unexpected file system errors. We suggest you save your data immediately [but all packages have been deactivated, including HB. But I have fresh backups and access to File Manager...] and move them to another one."
But I don't have another volume available and it doesn't let me create a new one.

Regarding disks health:
Capture d’écran du 2025-01-14 17-41-14.webp


If one of your 4 drives is configured as hot spare, then the volume might have been degraded and the system kicked the faulty drive out and placed a hot spare in its place. Still, this will take time to rebuild, so how long did it take for the system to get back to "normal" status? Check what drives are in use and confirm that the hot spare is in use. Also, if this is true, then one of your initial 3 drives of that RAID5 should be flagged as problematic in some shape or form.
That's what I would have expected too, but the above print screen shows that all disks are reported as healthy. Note: disk 1 has about 40 i/o errors, but that's an old story and that value has been stable for months now.

Also, if that is the only container your are running it makes sense that there was no other container calling out that it has an issue.
I run several. Is there a way to identify if this container has been the cause of the problem, or it's first victim?

Running a memory test via Synology Assistant desktop app might give you some insight into that.
I'll try tomorrow, thank's.

Now the main questions on the table are:
  1. Do I have any other option than deleting the volume, reinstalling a new one and hope my backups are healthy?
  2. Would you suggest to replace disk 1, even if it is reported as healthy?
  3. I would really like to identify the root cause of the issue in order to avoid this mess happening again in a few days, but the information I got so far are not sufficient to this regard. What would you suggest?
Thank's again to all!
BR
 
Last edited:
1. Delete and recreate volume. Restore from your tested backup.
2. No.
3. You may never know. It could be software or hardware related. Perhaps somewhere in Synology's logs (via SSH access) the cause may be noted. It is very unlikely a reputable docker image/container could cause this.
 
Volume 1 has been set as read only, probably because of unexpected file system errors.
This is the most probable cause of the issue, because it is clear that the rebuild has not happened in such a short time.

Considering that you are turning on/off your NAS it is clear that something is happening during boot causing your NAS to lock up the volume.

Apart from deleting the volume and starting fresh + backup restore, replacing the drive might not have an effect as it is not clear that the drive is an actual issue here.

While the drive is on the official supported 3rd party list for the 920 model, the added RAM might be a problem when contacting Synology support. Still, you could try and open a case with them, to have a look as this might be a minor issue that would not required such a radical step as a rebuild.

If you do not get support, you could, for testing purposes try and remove the added ram and then shut down and boot up. See if there will be any change. Still, removing the volume lock atm will probably be only possible via support, or as stated, by destroying and starting fresh.

I run several. Is there a way to identify if this container has been the cause of the problem, or it's first victim?
I would say the container going down was the result of that "system error", but without checking the logs it is just a guess.
 
Hi - exactly the same thing happened to me last week. Also a DS920+ with ironwolf drives. All the data seems to be there and intact, but the critical error and read only.

Did you manage to find any way of a fix without reinstalling the volume?
 
Hi - exactly the same thing happened to me last week. Also a DS920+ with ironwolf drives. All the data seems to be there and intact, but the critical error and read only.

Did you manage to find any way of a fix without reinstalling the volume?
Wow, what a coincidence (?)!
Nope...I deleted the volume and restored from an external USB drive. I didn't want to loose too much time, as several persons in my family are relying on some services (mainly Drive, Photos, contacts and calendar).

The recovery process went well and the main functionalities are back, but important ones are still unavailable (resuming backup tasks and Docker containers).
 
@Rusty @Telos : thank's a lot for your advices!

As mentioned above I didn't (couldn't) loose too much time and decided to restore quiet quickly. I didn't manage to run a memory test as the device could not be found by Syno Assistant. Didn't understand why, but anyway!

I restored from an external USB drive. Configuration, packages and shared folders, but still not a bare metal restore. It went quiet well for all main services: Drive, Photos, Calendar, etc. but I have issues with other services and no idea how to get them back properly.
  • Hyper Backup: package restored and the tasks appear in the left column. Nevertheless:
    • USB backup : system says it doesn't have the appropriate rights to use that folder. The same happens if I try exploring the disk with File Station, despite my account having the appropriate permissions.
    • Remote off-site backup: my NAS is now unable to connect to the second one. I received an alert from the second one, saying someone made too many failed attempts to connect. It actually was my 920+...
  • ABB: I had to reinstall the package (I think HB doesn't allow to backup this one) and all tasks vanished. Is there a way to recover the tasks, connections, etc. and resume from where it stopped?
  • Docker: also have issues with this one, but I could at least recover the compose files I was using, and will try to recover by myself before asking for help.
And last but not least, so I can be better prepared for next time :) Would another kind of backup (e.g. bare metal HB or snapshot/replication) have helped avoid these issues?

Have a nice week-end all and thank's again!
 
USB backup : system says it doesn't have the appropriate rights to use that folder. The same happens if I try exploring the disk with File Station, despite my account having the appropriate permissions.
After the new install of DSM, have you created a new account (even if its called the same) or have you during the install choose to restore from backup?

Or was the "restore" done after you have configured a new DSM instance and then installed HB in order to restore the backup?

Remote off-site backup: my NAS is now unable to connect to the second one. I received an alert from the second one, saying someone made too many failed attempts to connect. It actually was my 920+...
This might be due to a "fresh new account" that is not a part of the old backup task

ABB: I had to reinstall the package (I think HB doesn't allow to backup this one) and all tasks vanished. Is there a way to recover the tasks, connections, etc. and resume from where it stopped?
Backup tasks will be restored if you have backed up the folders that contain the data and task configurations. Use ABB to reconnect to the existing task/folder and the data will be accessible again as it will be the task itself

Docker: also have issues with this one, but I could at least recover the compose files I was using, and will try to recover by myself before asking for help.
What exact problem?
 
After the new install of DSM, have you created a new account (even if its called the same) or have you during the install choose to restore from backup?

Or was the "restore" done after you have configured a new DSM instance and then installed HB in order to restore the backup?
I did not reinstall DSM. At least not that I know... Only deleted the volume, created a new one and restored from USB disk with HB.
This might be due to a "fresh new account" that is not a part of the old backup task
I did not create a new account as DSM was not reinstalled. But yes, configuration (including users) has been restored. I'll investigate. That's actually not a huge issue, as I could just delete and recreate this user on both sides. It is just a bit strange, unconvenient and, most importantly, I fear that I will face the same issue resuming the backup tasks on the off-site NAS than on the USB disk. Future will tell, I'll let you know.
Backup tasks will be restored if you have backed up the folders that contain the data and task configurations. Use ABB to reconnect to the existing task/folder and the data will be accessible again as it will be the task itself
I'll try.

What exact problem?
None anymore :) Just as with ABB, I guess I am sometimes "over thinking" things. I thought I would not be able to "recover" all my compose files, data etc. on a fresh Portainer install, but it actually went straight forward. And you know what? There is actually a very positive aspect to this (small) disaster : it helped me clean the mess I created in Portainer through years "automatically"! I know only have the few, well running containers I really use, and all the things I tested months or years ago. Yeah! :D

Thank's a lot Rusty for your help not only here, but also on Blackvoid. All your writings definitely helped me "survive" this event with no really big issue, thank's to a pretty good backup strategy I learned mainly from you (HB, SR tutorials, etc.).
 

Create an account or login to comment

You must be a member in order to leave a comment

Create account

Create an account on our community. It's easy!

Log in

Already have an account? Log in here.

Popular tags from this forum

Welcome to SynoForum.com!

SynoForum.com is an unofficial Synology forum for NAS owners and enthusiasts.

Registration is free, easy and fast!

Trending content in this forum

Back
Top