I have an old Sparc based ReadyNAS NV+. There was a power cut this week and the NAS went down without being shutdown cleanly. It now will not reboot. It hangs, I left it for 24+ hours on "Checking FS". It is a totally stock device and has been running flawlessly for years. I am not sure when I last updated the firmware via raidiator. I did a firmware reinstall by holding in the reset button for 5 sec. That seemed to work fine, but the device still does not boot. I opened it up and cleaned out nearly 10 years of dust and reseated the RAM. Still no luck.
Everything is nominally backed up, but I would prefer to not have to do a factory reset. Is there something else I can try?
If I want to get the data off what do I need to do to stick the disks in my linux box (I have enough spare SATA connectors) to recover the data? It seems like from what I read it uses a software raid.
I finally got the drives installed into my linux box. My boot device is /dev/sdb and the four drives from the NAS are /dev/sda, /dev/sdc, /dev/sdd, and /dev/sde. I see four partitions sdx1, sdx2, sdx3, and sdx5 on all the devices but sde which shows nothing. Running pvdisplay finds volumes on sda5, sdc5, and sdd5, and obviously nothing on sde. Running lvdisplay shows me the logical volume C. So it looks like I lost a disk, but the RAID may make the logical volume still readable.
I then checked the drive health by running smartctl -a
sda:
SMART overall-health self-assessment test results: failed Drive failure expected in less than 24 hours. SAVE ALL DATA.
Apparently there have been nearly 33000 errors on the drive.
sdc: passed with no errors
sdd: passed with no errors
sde: passed with no errors
So it seems like sda is failing/has failed.
I haven't tried to mount the devices yet. Should I stick sda, sdc, and sdd into the NAS and see if it can boot with the 2 good disks and the one hopefully failing disk? Do I try and mount the LV directly from my Linux box? Do I just give up now? Is it worth trying to reconstruct the partitions table on sde with the hope that maybe only the partition table was lost?
Good to know about the parity drive. I have never really looked at how RAID works before. As for email alerts, my inbox is pretty much the equivlent of /dev/null. I will try putting the three good drives back into the NAS and rebooting.
So I put the 3 drives in that passed the smartctl tests. The ReadyNas booted up quickly without problems and reported that 0 TB free of 0 TB total. I then put the 3 disks back into the Linux box and booted up. I got errors about missing devices from pvdisplay and lvdisplay. I then shutdown, put in the 4th disk and booted the linux ox up again. I was then able to mount the volume with fuseext2 -o ro -o sync_read. I am now copying the data from the volume to my new NAS. I am copying the data in small chunks so that if/when the disk crashes I will know what else I need to restore from backup.
Once I get the data off, if the disk hasn't crashed, Is it possible to wipe the OS partitions to see if that fixes anything?
I suspect that if you wipe the OS partition you will either get a corrupt root or a factory reset. You could try an OS reinstall. Note you can do this from ssh (the post on that is here: viewtopic.php?f=18&t=62028#p357734). Not sure if that will work from ssh if you do the touch on the linux box, but it seems to me it should (assuming the clocks are close).
The other option is to put in the replacement disk and do a factory reset. You'll need to rebuild the NAS and back-copy the data.
i am facing exact same situation as beeper. My NV+ just hanged out of the blue when I was doing some file transfer. Upon forced power cycle, the unit will just hang at checking FS - 39% and after which will freeze again.
I am trying to recover some photos of my family and is looking to mounting the 4 disc X-RAID array on linux. Since the 4 discs are labeled in sequence in the NV+, must I follow the same sequence to mount them onto my pc?
