NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Factory reset ReadyNas RN 204 without loosing data on Disk 2
Hi all,
I recently experienced a failed drive (disk 1) in my ReadyNas 204 (OS 6.10), which is fine as I have the data backed up. However, I have also been experiencing the "volume root is 100 full" message, so am considering doing a full factory reset, to restore to factory defaults. As I have a second drive in the unit with my media library. Can I remove Drive 2, install a new drive in Bay 1, do the factory reset, and then re-insert Drive2 in Bay 2, without it loosing its data? Or is there an alternative way of achieving this. Many thanks in advance
14 Replies
Replies have been turned off for this discussion
sixx wrote:
Can I remove Drive 2, install a new drive in Bay 1, do the factory reset, and then re-insert Drive2 in Bay 2, without it losing its data? Or is there an alternative way of achieving this.
Normally not. But in your specific case, there might be a way if the media volume is still mounted.
If it is mounted, you can try exporting the volume and remove the disk. Then power down the NAS, insert the blank disk, and power up. That will do a factory install (no need for the reset). Set up the NAS and make sure you change from XRAID to FlexRAID Then power down again and insert drive 2. Power up, and the NAS should import drive 2.
While I think this will work, it's not a procedure I've tried. You should definitely back up the data on Drive 2 first. Note the export might fail if the root is completely full.
sixx wrote:
However, I have also been experiencing the "volume root is 100 full" message
If you have ssh enabled, you might be able to clean this yourself. Paid support (via my.netgear.com) can also take care of it - so another option is to see if JohnCM_S and Marc_V can help you get per-incident support for this.
Can I remove Drive 2, install a new drive in Bay 1, do the factory reset, and then re-insert Drive2 in Bay 2, without it loosing its data? Or is there an alternative way of achieving this. Many thanks in advance
BEFORE you do ANYTHING, poweroff ReadyNas, get a PC with an SysrtemRescue CD (Partitionmagic DVD(USB stick), do cloning the OS disk and the disk #2 to empty same sized disks (eg. Disk #2 has 2TB, so buy a 2TB disk) clone it to that (either by a USB 3 cloning disk station or by ddrescue in LInux) and do the same to the OS disk (to another empty disk"
Never ever change e.g. Disk #2 by a new one without backup OS disk and Disk#2. ESPECIALLLY YOU HAVE ENCRYPTED DISKS OR RAIDS....
8ohmh wrote:
Never ever change e.g. Disk #2 by a new one without backup OS disk and Disk#2.
sixx did say he has a full backup the files on the failed Disk #1.
Cloning disk #2 is an option I guess, but it won't solve the full root partition. If the volume is still mounted, backing up the data files on it is IMO a better option - since it would allow just doing a factory default to get a clean OS, followed by restoring the media from the backup.
Generally I don't recommend cloning unless it is part of a data recovery effort (e.g, eliminating risk of more damage to a failing disk). One reason is that any sectors that can't be copied can't be repaired by RAID - RAID recovery depends on knowing which sector is errored, and cloning loses that information.
8ohmh wrote:
ESPECIALLY YOU HAVE ENCRYPTED DISKS OR RAIDS....
You raise a good point here - I was assuming no volume encryption. I don't encrypt my own volumes, and I really don't know if you can export/import an encrypted ReadyNAS volume.
The solution StephenB gave assumes the two drives are separate volumes. It sounds like that's the case, but please confirm so the advice given is correct for your situation.
Did you start experiencing the error messages before or after Drive 1 failed? While the process of creating a new primary volume and importing the old media volume will temporarily solve the full OS partition issue, you also need to figure out why it happened to insure it doesn't again.
There is one potential reason for cloning drive 2 before you do the export and that's that an exported volume can't be put into a NAS by itself as a primary volume, where it can before it's exported. At least, it can't be done with any tools or process made public by Netgear. Since the import is automatic, I've been unable to figure out what is done in the export process that might be able to be undone.
One thing @StephenB didn't expressly warn about is that the volume name and no share names can be the same on the new volume as are on the volume to be imported. If you're just essentially re-creating the failed Volume, that would likely have been the case anyway, but I though I should mention it.
I have, by the way, done an export and import. I originally did several as tests with my "sandbox" NAS, but I ultimately did it for real. One of my tests was doing exactly what you are doing, except the primary volume didn't really die. I ran into no problems, so I don't feel that a clone of the volume to be exported is all that important so long as you have the backup and are willing to go through the restoration process if it comes down to that. But if you are on the verge of a full OS partition, there may be a risk I didn't run into.
Hi all
Many thanks for your replys, much appreciated
The Ready NAS is set up as JBOD, with 2 separate Drives. Disk 1 6TB, Disk 2 10 TB..(Appologies, I should have mentioned this in my first post)
Disk 1 failed, however I do have all that data backed up. I also have a back up of Disk 2 Data.
The "volume root is 100 full" message, started appearing prior to the failure of Disk 1
My Linux / ReadyNAS OS knowledge is limited, but can follow SSH / OS Command instructions.
In simple terms, I was enquiring about how feasible it would be to do the following:
1. Insert replacement disk into Bay 1, Remove Disk 2 from Bay 2 (Both JBOD)
2. Restore OS to Factory Reset, (to clear anything causing the 100% full message)
3. Transfer from Disk 1 back up, data back to new Disk 1
4. Insert Disk 2 back into Bay 2 - without loosing its data and avoiding the time and hassle of having to also transfer the files from the disk 2 backup
I hope the above makes sense,
Again many thanks for your suggestions so far.
sixx wrote:
In simple terms, I was enquiring about how feasible it would be to do the following:
1. Insert replacement disk into Bay 1, Remove Disk 2 from Bay 2 (Both JBOD)
2. Restore OS to Factory Reset, (to clear anything causing the 100% full message)
3. Transfer from Disk 1 back up, data back to new Disk 1
4. Insert Disk 2 back into Bay 2 - without loosing its data and avoiding the time and hassle of having to also transfer the files from the disk 2 backup
As I tried to suggest before, I think the following procedure should work:
- Export disk 2 from the volume settings wheel and then remove the disks.
- Insert replacement disk into bay 1, and do a factory install (if the disk it blank, there is no need to use the boot menu).
- Reconfigure the NAS, and restore disk 1 files from backup
- power down the NAS, and reinsert disk 2.
When the NAS is powered up, disk 2 should be imported. Note there is no import control in the web ui.
sixx wrote:
The "volume root is 100 full" message, started appearing prior to the failure of Disk 1
My Linux / ReadyNAS OS knowledge is limited, but can follow SSH / OS Command instructions.
The tricky part on cleaning this yourself is finding the files that are causing the problem. Since disk 2 is media, I am wondering if you are running plex. Plex transcoding files sometimes aren't deleted (at least that was an issue some time ago). If you let us know what apps you have installed, we might be able to offer concrete suggestions on what to look for.
Anyway, to look for yourself, you first enable ssh from system->settings->services. You log into the NAS as root, using the admin password. For instance from the Windows search bar (or terminal if you use macOS) you can enter
ssh root@nas-ip-address
The OS partition has a number of mount points, and they get in the way of investigating the problem. You overcome that by remounting the root onto /mnt and then examine /mnt.
mount --bind / /mnt
You can undo this later on by entering
cd /
umount /mntNote it is umount, not unmount. Also, this command will fail if you are in /mnt (which is why I put the cd command in front of it).
du -csh * is a good way to find folders that are too full. Basically you start with /mnt and then descend into the largest folder and repeat.
Here's what I see on my RN526 - comparing yours with this should help get you started:
root@NAS:/mnt# du -csh *
0 apps
6.4M bin
0 boot
0 data
24K dev
11M etc
30M frontview
0 ftp_ban.tbl
0 home
4.0K homes
33M lib
4.0K lib64
0 media
0 mnt
4.6M opt
0 proc
40K root
0 run
11M sbin
0 srv
0 sys
0 tmp
273M usr
675M var
1.1G total
root@NAS:/mnt#
StephenB wrote:As I tried to suggest before, I think the following procedure should work:
- Export disk 2 from the volume settings wheel and then remove the disks.
- Insert replacement disk into bay 1, and do a factory install (if the disk it blank, there is no need to use the boot menu).
- Reconfigure the NAS, and restore disk 1 files from backup
- power down the NAS, and reinsert disk 2.
When the NAS is powered up, disk 2 should be imported. Note there is no import control in the web ui.
I have performed export/import experiments, and it worked fine. The steps I was doing were to figure out a way for people to recover from an accidental export of the primary volume, so my test would be more accurately duplicated by adding a couple steps:
- Boot the NAS with just Drive 2.
- DESTROY the now non-existent volume from Drive 1. This will make Drive 2 the primary drive.
- Now proceed as above.
I've also done an export when the primary drive is fine, but yours isn't, and I don't know if that will make any difference.
Note that once you have exported the drive 2 volume, it cannot be made back into a primary volume -- it must be imported into a unit with an existing volume and that unit must already be in FlexRAID mode. That existing volume cannot have the same name as the one being imported, and no share names can be duplicated, either.
When you are done, you will have to go in and set permissions for all shares on the imported volume, including doing a permissions reset for the existing files.
The wrinkle in this is that the full OS partition may limit your ability to do the steps leading to the export. volume_util, accessible via SSH, seems to have the same capabilities, but I've not used it for this. If you can at least partly clean the OS partition first, it would probably help. Not knowing that you won't get stuck somewhere in between with a non-importable volume is one good reason to insure you have a backup of the files. As I said in a reply to 8ohmh , you could also make a clone of Drive B before doing anything since some of the steps are irreversible, but I doubt you'd have any better luck trying again on a clone, so I don't think it adds a lot. You just have to be prepared to start everything from fresh if it goes south, restoring the media files from backup.
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
