My NV+ has been serving my apartment like a dream for over 3 years. I use it for media, computer backups (via Time Machine), and work storage. 4x2TB drives. Firmware is a little older to accommodate Time Machine for Snow Leopard. I'd grab you the exact version, but my NAS is currently unresponsive :S
3 days ago, Time Machine on my main machine notified me that it needs to setup a new image (or something to that effect). This has happened in the past - the Time Machine process usually does it's thing, creates a new image and copies over an entirely new backup (~150GB), and we're good to go. Unfortunately, this time around, it doesn't get too far in the copying over process (less than 3GB). The backup process stopped and I'm left with an unresponsive NAS (couldn't power down with the power button, couldn't access the control panel).
So I pulled the power cable (I realize now so many of my problems may stem from this - but it had been unresponsive for a long time).
The NAS fires up again, does a quota check, passes.
400GB of space has been freed up.... gulp. (went from 1.62TB/5.5TB free to 2.02TB free) It would appear that an NFS volume has been corrupted.
Through the control panel, I check the logs. Nothing out of the ordinary, asides from improper shut downs and scans starting.
I restart the device with a file system check. The file system check reaches 68% (which happens to be the approximate used capacity of the NAS) and doesn't make it any further than that. The scan began around 9:00AM, was at 68% by 10:30AM, and was still stuck there in the early afternoon (2:30PM), so I feel confident I'd given it enough time if it was a time intensive process.
Wind up pulling the cable again.
After the next startup/quota check, I find that the volumes are re-syncing themselves through the control panel. This process take the better part of a day, but it went smoothly/successfully.
I fire up RAIDar to see if it can tell me anything else. Everything appears to be best case: drives are full health, FS check is 100% - nothing out of place.
Wondering if the issues have been resolved via volume re-sync (aside from the corrupted volume), I try my Time Machine backup again. Same issue. Backup ends early on - NAS becomes unresponsive.
I've only spent an hour or two googling around looking for answers on this, but I'm hoping the wise wise forum members might be able to point me in the right direction on this.
My immediate interests are: - Guidance on proper diagnosis (could this be a bad sector?) - Possible solutions on why the Time Machine writes are failing (I've seen forcing a rebuild offered as an idea on here) - I realize it's a super long shot - but I'll lob out the confirmation question of "Is there such a thing as (inexpensively) recovering a corrupted NFS volume"?
What firmware? Have you removed and tested all the disks using vendor tools? You'll need to power down before removing and labelling ten and reinsert all with the Nas off. Do you have ssh enabled?
Yeah, I had tried upgrading to 4.1.9, but it broke my Time Machine backups, so I rolled back and have stuck with it. I don't recall having done any factory resets since initial setup.
I will go pick up a hard drive dock this afternoon and start testing drives. I was of the impression the NAS would have reliable disk error checking built in. I'm guessing I was mistaken? Or is this a fringe case?
If the smart data is perfect then the drives are probably ok, it's always worth checking in situations like this where writes and filesystem checks are failing.
The reason I asked about the last factory default is explained here:
So, my quota checks on startup are starting to choke (The NAS appears to be inaccessible). Once at 80.8% and currently at 87.8%.
Given that there's a decent number of files I'm hoping to copy - am I playing with fire with the 'pull the plug to restart the cycle'? Is there any point to waiting out a stuck quota check?
I checked the HCL, the Caviar Green is in fact on it.
Extended tests are complete. Disk 2 had a small number of bad sectors - so I repaired those. I tried firing things up after that - still a no go. My first attempt at scanning disk 3 resulted in the most serious warnings I got. It initially failed quick and extended tests, citing too many bad sectors - I was initially excited: yay, let's just replace the drive. Wanting to rule out a bad seating in the hard drive dock I was testing it in, I tried it in the other bay. My second attempt at scanning disk 3 resulted in passing both tests. Disk four passed everything uneventfully.
Firing up the NAS after all that, I was fortunate enough to access to my files again. I'm furiously backing up GB's and GB's of data now (my own doing, I realize - I deserve this discomfort).
I realize the sage advice would be to do a factory reset, start fresh, and hopefully restore some faith in this device. Given that it might be a few days/weeks until I round up 3.5TB worth of storage (and the time) to do the right thing - how freaked out should I be about this problem returning?
I know the core message will be "you fool, your NAS is not a backup", but if you could shed light between: - Hard drives do this from time to time, repairing the odd bad sector is a part of disk maintenance but it's not indicative of imminent failure and repaired disks can continue in a RAID5 environment where single disk failure has been planned for. vs - When hard drives start to show bad sectors, it's a black hole of unpredictability. Expect problems to return in short order - time to dispose of the disk.
Yes it is on it, but it has been out of favour with users for some time.
I'd suggest that running disks with bad sectors is not a good idea, especially if they are on more than one disk. The tolerance of most users here is 50. I don't think anyone is going to advise "repairing" sectors.
While I have the ear of a super knowledgeable NAS team, has there been any progress towards higher capacity drives for the NV+?
If I'm starting from scratch again, my interest would be to move up to 3-4TB per disk, but I remember 2TB being the cap for this device when I started fresh last time. Not sure if firmware updates can improve such things, just checking.
