D-Link Forums
The Graveyard - Products No Longer Supported => D-Link Storage => DNS-323 => Topic started by: halvar on November 10, 2013, 12:05:01 AM
-
Hello, I have a DNS-323 with 2 hdd Hitachi 2TB set as a RAID 1 for backup purposes. In the status screen the messages are bad:
Total Drive(s): 0
Slot Vendor Model Serial Number Size
Right - - - -
Left - - - -
In the logs the messages are:
HDD1 Hard Drive Has Failed
HDD2 Hard Drive Has Failed
I have attached the disks on my PC and the data is ok, only that in one disk the data is not update (6 days ago) In the other disk the data is up to date.
I have attached other several hdd in the DNS-323 (Seagate Barracuda 500GB and Seagate Barracuda 1000GB) and they are not recognised, and the logs says the same message HDD1 Hard Drive Has Failed
So I donīt understand if the DNS-323 has a problem, I've restore it to factory default settings and the problem is the same.
Thanks
-
Think about this for a minute...
So I donīt understand if the DNS-323 has a problem, I've restore it to factory default settings and the problem is the same.
1) The DNS-323 is reporting the original disks as failed, they test good in a PC
2) You've connected several - presumably KNOWN GOOD - disks to the DNS-323 and it does not recognize them
Isn't it obvious that there is a problem, and it's not the disks? It can only be the DNS-323 or, perhaps the power supply.
-
Thanks fordem, I canīt think about power supply due the dns-323 is power on and the web server running.
What power supply do you reffers to?
-
Ah, and the leds of the disks are blue. In the manual solid amber may indicate failure ???
-
The power adapter for the DNS-323 supplies 5v & 12v power - as far as I know the 12v power is used only by the disks, so theoretically - if it was not available, it would/should be possible for the DNS-323 to power up but not recognize any disks at all.
What I find puzzling is that you say the HDD lights are blue - if the unit is seeing the disks as failed, those lights should either be amber or off, and if it has been reset to default and is not recognizing any disks I would expect the lights to be off.
You can try a firmware update - even if it's to the same version - if that doesn't fix it, there's not much else you can do - replacement is the next step - my DNS-323 failed earlier this year (stopped booting, power light just stayed flashing) and I replaced it with a ReadyNAS.
-
Thanks, I have tried with 1.10 firmware and with Alt-F-0.1RC3 firmware with the same bad results.
If I deattach one hdd the DNS-323 says that has been removed.
It is a good device, but I think I have to change it :(
-
From the symptoms you describe it could still be a power supply problem with the 12v line not supplying enough current to run two disks correctly - if a disk does not get enough current at the correct voltage its firmware reports it as a disk failure.
The only reliable way to check the voltages from the power block is to do it under load, i.e. when connected to a load, your DNS-323.
I am assuming you have not tried with only one drive in place. Doing this will reduce the load on the power supply and may, if that is the problem, allow the unit to see the disk. It will also test the indicator circuitry - should be only one disk LED on - and so indicate if there is a problem with the units logic circuits.
-
Thanks, but I have tried with only one disk and the DNS-323 logs says that HDD1 has a failure.
The disk led is blue :-\
-
For background. . . has this DNS-323 been in service for a long time and is only now exhibiting issues or is this a new DNS-323 that's experiencing issues out of the box?
-
Good question, the dns-323 device has been working for 3 years without any problem.
-
halvar, how confident are you about opening the box to the extent of exposing the main board? I ask because our experience points to a power supply failure of some sort.
If you open the box it is possible to to get a volt meter reading at the power input socket with the unit on and no drives installed - should be 5v +- 5% and 12v +- 5%, anything much beyond that range and you need to replace the power block.
Then install one drive and take another reading - the voltages should remain almost the same, if either of them drops by a significant amount you have found the main problem.
The other thing to look for when you have the unit apart is capacitors with the tops domed up - they need replacing because they are faulty.
If you don't feel confident enough to do the test yourself anywhere that repairs electronic equipment should be able to do it for you, in fact they mat have the equipment necessary to test the output of the power block without taking the unit apart - we have a universal unit that tests up to three voltages simultaneously under varying loads.
-
Hello halvar and others
I have exactly the same problem but i restarted two times the device by disconnecting the power plug twice (not a good idea i've read !)
1. Did you find a solution for this problem ?
2. Do you confirm there is no danger to try to access the drive with an external docking station and plug it in a Windows 10 PC and then use Ext2IFS to backup the data ?
Thank you for your help
-
Patate, have you followed the instructions I gave in the post above yours concerning the power output of the power brick?
It is usually the loss, or reduction, of the 12volt supply line that causes the problem you are seeing. You could also check the voltage of the button cell that is used to retain memory setting when the unit is powered off - if it is much below 3volts it should be replaced.
It is not safe to connect any Linux formatted hard disk to a win 10 computer if you value your data. The safe way is to get a Live Linux DVD and boot the computer with that then attach the USB mounted disk and see if you can read it (works with RAID 1 disks, RAID 0 and other formatted disk arrays need special software to setup and read the software RAID of these units)
-
Thank you for your quick reply ivan.
My disks are in RAID 1 (hence the bad idea to switch off by removing the cable and then switch it on because I -think- auto rebuild is enabled).
When I turned it back on, the web server was accessible for some time but it is not anymore. It was no longer possible to turn off the NAS with the button so I turned off the power (without turning it back on).
Unfortunately, I do not have the skills to verify that (volts).
I have bought an USB SATA docking station and downloaded linuxmint-19.2-cinnamon-64bit.iso but how can I verify the iso file on Windows ? (for integrity and authenticity) [edit: I am currently following instructions on https://forums.linuxmint.com/viewtopic.php?f=42&t=291093 but it is not straightforward…]
Thank you again
-
The fact you have a RAID 1 array means that recovery is much easier since both disks should be identical.
Your problem sounds very much like a bad power brick but put that aside for the moment.
First thing you need to do is burn the ISO to a DVD, assuming your computer has a DVD drive - if no DVD drive then setup a bootable USB stick.
Now extract one disk from the NAS and mount it in the docking station.
Once that is done boot from the DVD or USB (may be necessary to tell the computer which device to boot from). when Linux Mint starts to boot it will ask for keyboard information and then ask if you want to try or install (select TRY unless you want to remove win 10).
With the Mint desktop up and running (should be something like the old win 7) plug in the docking station and an Icon should appear on the desktop. Double clicking in that icon will open the file manager showing the root level of the disk, assuming the disk isn't damaged (should show a single folder Volume1).
Opening the Volume1 folder should show all your data and allow you to copy it to another disk (drag and drop).
What you do from there is up to you (I would use a USB disk and transfer all the data to it for a backup)
With your data safe you have a choice, either find or borrow a new power brick to check if that is the problem. If the NAS boots up as normal - problem solved. If it doesn't then you will have to consider replacing it.
-
I've tried with the left drive: it is recognized as a 526 MB volume with 499MB free (it's a 1TB drive !). There is only a backup folder which is empty. I am feeling bad. Same with the right drive. Linux says it is ext3/ext4 (would not that be rather ext2 ?)
Here is the mail I received on 11/09. And I received it again after restarting the DNS-323 (by removing the power).
Left Hard Drive Has Failed
Right Hard Drive Has Failed
Sincerely,
Your DNS-323
It think I have firmware 1.05 or 1.06 (I have the first version of the DNS-323, from 2008 I think).
What can I do now ? Is there a chance to recover my data ?
Can I restart the DNS-323 without any disk to try to look after some params (firmware version
) ?
Thank you
edit: Is it possible that I got hacked? The day hard disks failed I received an email to pay otherwise my data would be destroyed but I erased the email. I do not believe it too much.
-
That sounds bad and yes it is possible you have been hacked especially if your NAS is/was open to the internet.
My 4 DNS-323 NAS are about the same vintage as yours but I run them behind a solid firewall with access to or from the internet (I also run them from a large UPS for power security).
There was a thread in one of the DNS NAS groups about the problems with hacking http://forums.dlink.com/index.php/?topic74592.0 (http://forums.dlink.com/index.php/?topic74592.0) I don't know if that will be helpful or not.
Sorry I can't be of much more help.
-
Thank you for your help ivan.
I think the "Backup" folder on the disk is the first of the volume...so maybe the raid1 is implicated.
I will wear the HD tomorrow in a data recovery center.
-
>>>it is recognized as a 526 MB volume with 499MB free (it's a 1TB drive !).
>>>There is only a "backup" folder which is empty.
It is possible that this folder was the first one on my volume.
Any idea about what happened ?
Is this the fault of RAID 1
(+-900$ for data recovery !)
Thank you
-
The only time I have seen a problem with a RAID 1 array in the last 20 odd years was when a disk physically died and someone pulled the good disk rather than the dead one (we were able to recover all their data from the 'good' disk and managed to get them to set up a valid backup system).
Unfortunately all small NAS boxes use what is known as a software raid, where the operating system does the necessary processing and can be a problem if the parity information is damaged in some way. Once the data structures are damaged or lost, data recovery becomes a long slow process and full recovery is most probably not possible.
In my old company (I am now retired) we had signs saying 'any data that you do not have at least two other copies of is not backed up', my department added to that with 'and a third tested copy held off site'.
I am very sorry that you appear to have lost your data, I think we have all lost data at some point, but you must consider the value of the data against the cost of possible partial recovery. That is something I can't advise you on.
-
I think I should have cut the power directly after receiving the mail "failed" for both drives.
But as the lights were blue and not amber, I turned off the power and turned it back on. Then the self-rebuild may have messed up.
I do not think I would use RAID1 again.
I have another NAS (Synology) and it uses SHR.
-
I finally was able to recover my data using Linux Live and mdadm.
There was only one bad sector on the right disk (left disk was fine).
I've tried to format the two disk and to do several hard reset but the NAS still don't "see" the disks (total drives: 0 and "HDD1 Hard Drive Has Failed" & "HDD2 Hard Drive Has Failed" in the logs).
There are 3 possibilities according to my researches:
- 3V button cell should be replaced
- the power supply unit should be replaced
- the NAS is dead