I have an NV+ with 4.1.8. It's been running for several years. Recently it refused to boot with the message "found bad disk" on the LCD. Based on info from searching around in the forum, I pulled disk 1 and the system would boot. Disk 1 is bad and won't read on another system. I purchased a new disk and plugged it in. when I boot with the new disk it doesn't seem to try and recover the data -- I get "no volumes".
Is the new disk on the HCL? http://www.readynas.com/?cat=37
Have you tried to remove disk 1 (the new disk) and boot the NAS?
The new drive is ST2000DL003. It shows on the HCL list. I plan to replace the other disks as well if this works, and double my capacity in the process...
If I remove the new drive the unit boots. I don't see the old shares at all at this point. If I put the old one back in boot fails.
The other odd thing thing is that looking back at the log it was drive 3 not drive 1 that reported SMART errors.
mdgm wrote:
Was it purchased on/after Aug 21, 2007? I would suggest you contact NetGear tech support
that seems like a bother... I was hoping that it would be straightforward -- replace failed drive and then replace the others...
I'm going to continue to document my travails in hopes this helps someone...
So, after a total of 3 hours with tech support, plus a useless 24 hour wait dictated by the first tech support rep, the result is that they gave me an RMA for the chassis because he somehow determined that slot 1 was bad and gave me an RMA, which I then happily paid $50 to get overnight.
Well, I have since determined conclusively that there is no problem with slot 1 in the unit by installing a new drive in slot 1 of the "defective" chassis and I am able to format it and access it just fine. So the problem was the drive.
I have determined this conclusively also -- that drive 1 of my 4 which were in the unit was bad -- by installing all 4 drives in another NV+, and having it fail to boot even after a "firmware reset". Finally, I placed a new (unformatted) drive in place of the defective hard drive 1, and performed a "firmware reset" using the back panel switch.
In this state, the unit boots, and I see my old shares listed in frontview, but the shares are not accessible and raidar shows a yellow "warning or dead" indicator for the unit.
Unfortunately, in frontview, the button for "resync" is grayed out, so what is the next step? Anyone? Call tech support again? (last time it was 45 minutes on hold followed by a very nice and helpful fellow, who unfortunately reached an incorrect conclusion...)
claykin wrote:
menlomike386 wrote:
[BUMP] Can anyone point me to steps to troubleshooting a defective hard drive with "found bad disk error"? TIA
When you plug the "dead disk" into a PC can you see it? Even though you can't read it?
Windows asks if I should format it. When I say "no" the disk doesn't show in disk management -- in fact disk management doesn't finish loading, so I'm guessing it's totally dead. I believe that when I plug the other drives into a windows PC I get a linux partition, a swap partition and an unknown large partition.
Well when you got that "no volumes" error, I immediately concluded that if possible cloning the dead disk e.g. using dd_rescue on Linux, if possible was probably what should be tried. If you can't even do that, unless there's some way to fix the problem without repairing that drive it looks like any data recovery attempt would be quite expensive as you'd need a 3rd party data recovery service to attempt to repair the drive. Not sure if what you've subsequently tried would affect chances of data recovery.
Just my 2 cents, but if drive 1 fails and you can boot and access everything with 2 thru 4 installed, then you simply hot plug the new drive 1 in and it is added and resync begins. Not sure I would boot with a new drive installed, and by "firmware reset" I really hope you meant OS Re-install and not the Factory Default function using the rear reset switch.
Just my 2 cents, but if drive 1 fails and you can boot and access everything with 2 thru 4 installed, then you simply hot plug the new drive 1 in and it is added and resync begins. Not sure I would boot with a new drive installed, and by "firmware reset" I really hope you meant OS Re-install and not the Factory Default function using the rear reset switch.
[list=a:2zfeztz0]
the unit won't boot at all with disk 1 empty
when I replace disk 1 with a good drive I get the unit up, but it does not resync and does not show my shares
yes, I meant OS Re-Install. Have not done a factory reset
Summary:
Disk 1 out, others in: "Checking root FS" on the LCD and "file system check" in RAIDar with 3 green indicators for disks 2 - 4. This quickly changes to just "booting" and only the first indicator as blue with the others not there (not even gray) and then both the LCD and RAIDar change to just say "booting" forever.
4 drives in OR disk 4 out, 1,2,3 in: "Checking root FS" followed by LCD: "found bad disk" (singular) and RAIDar: "Bad disks detected" with the first indicator blue and all others gray.
disk 2 out, 1,3,4 in or disk 3 out, 1,2,4 in: "Checking root FS" Followed by "found bad disk 1" on the LCD and "Bad disks detected" in RAIDar with disk 1 showing yellow. The disk 1 access indicator is flashing and I hear what sounds like a seek-reset... seek-reset cycle ongoing.
So I suspect there's something wrong with disk 1 AND one of the others, but I'm not sure which of the others (probably disk 4?) So I guess my data is gone... 😞
Waiting on a call from NTGR to telnet into my unit...
I suspect you likely have one or more bad disks. You already determined that disk 1 is a problem. I can appreciate your desire for troubleshooting, but be careful testing the way you are. You've identified a disk problem, what you don't want is the NAS to kill your existing volume on one of these power up attempts.
Regarding your defective disk, I pointed you in the direction of Spinrite for more than just to test. Spinrite is known for its ability to bring back disks from the dead. Seriously. The disk must be electrically working and recognizable in BIOS, and if it is, Spinrite is worth a shot. If it doesn't work for you GRC will refund your $ if you request it.
Run Spinrite at level 2 on all 4 of your disks (I recommend you do one at a time). Make sure you mark where each disk went. They must be returned to the same slot in the NAS or your X-RAID is GONE!
With Spinrite, use a PC and preferably an Intel motherboard SATA controller. If you run into problems set your SATA controller to IDE mode for the purpose of running Spinrite on the 4 disks. Be sure to return it to AHCI mode after you've finished testing all 4 (if thats where it was set) or your host disk on that PC will fail to boot.
Keep in mind that if the disk(s) are recovered you should boot your NAS and immediately copy off your important data. Sometimes Spinrite can get you going just long enough to get your data...
I am speaking from personal experience. Spinrite works!
So, claykin, I decided to try your advice and since all my drives were seagates, I went with the seatools first.
Fortunately I have a windows PC which is somewhat disassembled since I am converting it for use by my kids. This made it fairly easy to hook up the drives and run seatools.
I tested all the drives and drives 1, 3 and 4 ALL failed the basic self test! So as the seatools documentation recommended, I used the DOS boot CD image they provide to boot and run the "long test" to "repair" the disks. I had to run it twice on disk 1, and after I was done, disk 4 had no partition table, so I tried a partition repair tool and that didn't work, so I copied the MBR from disk 2 to disk 4, but that didn't work either.
Anyway, once I plugged disks 1, 2 and 3 back into the ReadyNAS and did yet another "OS re-install", the unit booted right up!! As I write this I'm copying off several years of Quicken data backups, family photos and iTunes purchases!! Yay!!
:woot: :woot:
Given a total of some 200 bad sectors "repaired" by seatools, I'm guessing some of my files may be corrupted, but a few hundred sectors is a tiny fraction of 400+GB, so most are saved!
With X-RAID e.g. on an NV+ (v1), the parity disk (usually disk 4 in a 4-bay system) does not have a partition table. The lack of a partition table on this disk is not a problem unless you're trying to do data recovery attempting to mount the volume in a PC. However as you've noticed you only need 3 working disks to recover data.
X-RAID is found on the Sparc devices e.g. NV+ v1 and not to be confused with X-RAID2 (found on x86 and ARM e.g. NV+ v2) which is a different RAID format.
You can have problems with multiple disks at once. It's unfortunate, but it happens. Would be a good idea to start regularly backing up data.
Good to hear Seatools helped you boot the system. :thumbsup:
If you have trouble copying files off the NAS, and those files are important to you, I still highly recommend Spinrite.
Now, before you begin using your NAS again, what is your plan for disks? Since 3 of your 4 Seagate Barracuda Green disks are considered bad by Seatools, I'm thinking you need to investigate further before relying on these disks. Did your 4 disks have the latest firmware? See here. http://seagate.custkb.com/seagate/crm/s ... cId=218171
If the firmware looks OK, do you have any idea why 3 of the 4 disks could be bad? Was the NAS "manhandled" while powered on? Is there sufficient airflow (check for NAS air vents clogged with dust)?
I had a business colleague who had several disks fails in his Readynas over the course of a year. Upon further investigation we found that the nighttime office cleaners were picking it up by the handle and cleaning the desk just about every night. They were obviously "manhandling" it. Since we ceased that behavior not a single disk failed.
If your disks are still under warranty I suppose you could get refurbs from Seagate. I've had some real odd experiences with Seagate over the last couple of years with refurbs. I have had 4 occasions where ES and ES.2 disks needed to be exchanged. In all four cases the first refurb sent was unacceptable, meaning they were found to have serious SMART errors after one round of pre-testing. One of the refurbs sent even had a dent in the cover and was all scratched up like someone dragged it across the floor. All were labeled Seagate re-certified and shipped from Jabil in Texas. I'm suspecting their recertification criteria has changed over the years as I never used to have such issues with them.
May I recommend you consider Enterprise disks? The only trouble is that timing is bad. With the floods in Thailand disk prices are through the roof and likely won't come down until Summer. If I had to recommend a consumer 2TB disk it would be the HGST 7K3000.
