NV+ Booting Issues and Possible Data Loss (16344829)
Hey guys,
My NV+ has stopped booting successfully, it would always hang and just say "Booting...". I contacted support and they helped as much as they could, but since my unit is no longer under warranty, they couldn't really solve my problem.
What I have figured out is even though all 4 drives passed SeaTools tests, drive 3 somehow had a corrupted partition table. I was instructed to remove that drive and try to boot without it by support. I've done so, but it still just hangs at "Booting...". I am able to ssh into the box in the state that it is in and here is a dump of dmesg after it has stopped booting (hdg is the removed drive):
Add dead device not present and allow master sector. Link drive hdg to its hwgroup. hdg: max request size: 512KiB hdg: use capacity 2930277168 sectors (1500301 MB) Drive support hpa, still should not change max addr. :<6>hdg: 2930256688 sectors (1500291 MB), CHS=65535/255/63 ::<6> hdg: (Fix non bootable low sector,flag=0xffffffff) ide3 at 0x300-0x307,0x308 on irq 34 idedisk_deaddisk_init on hdg ide-disk: hdg: from special init, need to reset values. hdg: max request size: 512KiB hdg: use capacity 2930277168 sectors (1500301 MB) Drive support hpa, still should not change max addr. :<6>hdg: 2930256688 sectors (1500291 MB), CHS=65535/255/63 Change X_RAID running mode from 0 to 1 ::::Update backup SB. X_RAID: recovery thread got woken up ... No drive to use, stop recovery. :<6> hdc: hdc1 hdc2 hdc3 < hdc5 > :<6> hde: hde1 hde2 hde3 < hde5 > :<6> hdg: :<6> hdi: hdi1 hdi2 hdi3 < hdi5 > kjournald starting. Commit interval 5 seconds EXT3 FS on hdc1, internal journal EXT3-fs: mounted filesystem with ordered data mode. linked, 1000mbps mode :::::: CMD to offline/lost_intr chn(2): 80f07ae0 ::<4>EXT2-fs warning (device hdc1): ext2_fill_super: mounting ext3 filesystem as ext2 Adding 255968k swap on /dev/hdc2. Priority:0 extents:1 across:255968k Adding 255968k swap on /dev/hde2. Priority:0 extents:1 across:255968k Adding 255968k swap on /dev/hdi2. Priority:0 extents:1 across:255968k enable_irq(11) unbalanced from f804059c :<6>hdc: cache flushes supported :<6>hde: cache flushes supported
CMD to offline/lost_intr chn(2): 80b17ae0 :<6>hdi: cache flushes supported :<6> hdg: (Fix non bootable low sector,flag=0xfffffff5)
Does anyone know of anything else I could try? At this point, it seems my data may be unrecoverable.
Re: NV+ Booting Issues and Possible Data Loss (16344829)
That option may only be available to those with units under warranty, I don't know. As a fellow ReadyNAS user though I'd suggest inquiring as to whether there's any option for that. It'd definitely be better than trying to fix the problem yourself.
Another option would be to try a local data recovery company that understands how to recover data from Linux devices that are using RAID.
Re: NV+ Booting Issues and Possible Data Loss (16344829)
I am still waiting on a response from Netgear if an out of warranty service is even an option.
In the meantime, I acquired a brand new 2TB drive (my NAS contained 4x1.5TB), so I should be able to use the 2TB drive to replace the dead 1.5TB one under normal conditions. I installed the new drive in the slot and turned on the NAS, and of coarse it still hangs at Booting... I then sshed into the box and looked around and this is where things get weird. Both fdisk and hdparm report the new drive in slot 3 as the old one. fdisk shows it as a 1.5TB drive and hdparm even shows the serial number of the old drive. Does anyone have any idea why this may be happening? Is there any way to force the NAS to redetect the drives and (perhaps) attempt recovery?
The only thing that keeps me going on this are the following lines in dmesg:
X_RAID: recovery thread got woken up ... No drive to use, stop recovery.
This is pretty much the last thing that happens before it stops booting. My thought is if I can get the NAS to see the new drive, it may attempt a recovery using it.
