Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[andy-d]

Hello -

I've been running a DNS-323 without the ghost of a problem for more years than I care to remember, until today. I'm used to my router occasionally messing IP addresses around, making it inaccessible, so I wasn't too worried at first when I couldn't get access to it this morning. Several hours later, having failed to successfully reconfigure the network, I found the log page in the DNS-323 admin area that told me both hard drives failed yesterday early evening, three seconds apart from each other. It was plugged into a multi-block with surge protection, but clearly not effective enough protection as the 'power surge' warning light on the block is flashing.

This has not made me happy.

I have found a number of threads on here that I now need to read very closely (and will do) but I'm not very experienced in this sort of thing so it's going to be a steep learning curve. Hence this post, with one simple question:

What are the stupid things I must at all costs avoid doing?

To answer some of the more obvious questions: the data on it is both precious and important; my back-up regime lapsed some time ago; and yes, I know how stupid the combination of those two statements makes me. When I first discovered the problem it would not shut down using the front button and I did not think to use the admin area, so I pulled the power cord (it was all I could think of). After that, I've been using the admin shut-down. I can't think of anything else immediately to mention.

But as I say - I'm not posting to ask for help with the recovery (at least, not until I've exhausted all the possibilities elsewhere on these forums). I'm posting to ask if there are any obvious mistakes in what comes next that bitter experience has taught you all I'm in danger of making through ignorance.

Thanks in advance,

Andy

(Title edited as apparently one of the words in it can't be used. Apologies.)

[FurryNutz]

Link>Welcome!

• What Hardware version is your DNS? Look at the sticker behind or under the device.
• Link>What Firmware version is currently loaded? Found on the DNSs web page under status.
• What region are you located?

What Mfr and model is the main host router?
Is the DNS getting a Dynamic IP address from the main host router? I recommend using a Static IP address ON the DNS set to 192.168.#.32 or .23.

Can you plug the DNS into a different wall out let and or power strip?
I use a battery back up system (UPS) rather than a power strip.

[andy-d]

Thank you for the quick reply. Answers as best as I can:

What Hardware version is your DNS? Look at the sticker behind or under the device.

B1

What Firmware version is currently loaded? Found on the DNSs web page under status.

It came with 1.04 and I don't think I ever upgraded it - until today. While troubleshooting I installed 1.10 (which is when I found the log about the failed hard drives).

What region are you located?

I'm in England. (And I know the board says US-only, but there's so much knowledge here...)

What Mfr and model is the main host router?

Linksys WAG160-N

Is the DNS getting a Dynamic IP address from the main host router? I recommend using a Static IP address ON the DNS set to 192.168.#.32 or .23.

It was getting a dynamic one but because I have such regular trouble with it - often it ends up with two IP addresses, and did several times today - pretty much the first thing I tried after re-starting it was to assign a static IP. I didn't know about using 23 or 32 though, I used the one the router had assigned it.

Can you plug the DNS into a different wall out let and or power strip?

Hadn't thought of that - will try it. To be honest, though, I'm not sure how that will help. The DNS can be found using the Fing app on an iPad and LanScan on my Mac, it can be pinged, and the admin set-up is accessible on two different PCs and the Mac. It occasionally shows as a media server on the Win10 machine (although it's never been used as such). The network knows it's there. But the hard drives... not so much. It gives explicit times for the failure of each of them in its log.

[FurryNutz]

I would re-try the static IP ON the DNS. just make sure your routers default IP address pool is set to .100 thru .200. Then the Static IP address you set ON the DNS will work just fine. You'll always know where the IP address is and how to access it. Won't change.

How long have you had the DNS and these drives in the DNS?
What is the mfr and model of the drives.
What configuration is the drives using on the DNS? Standard or RAID#?
Do you have the data on this DNS back up someplace else? >http://forums.dlink.com/index.php?topic=52503.0

[andy-d]

I would re-try the static IP ON the DNS. just make sure your routers default IP address pool is set to .100 thru .200. Then the Static IP address you set ON the DNS will work just fine. You'll always know where the IP address is and how to access it. Won't change.

How long have you had the DNS and these drives in the DNS?
What is the mfr and model of the drives.
What configuration is the drives using on the DNS? Standard or RAID#?
Do you have the data on this DNS back up someplace else? >http://forums.dlink.com/index.php?topic=52503.0

I'll give the static IP a try in the morning (it's a bit too late in the evening now, almost 11pm). I still don't really understand why it might provide a solution when the DNS admin reports two drive failures, but I'm very aware that if I knew what I was talking about I wouldn't need to be here.

I'm not sure exactly when we got the DNS but it's at least 10 years and perhaps 12. Same drives throughout, but I don't know what they are as I haven't opened it up yet. Again, tomorrow.

I've always believed it was RAID, one mirroring the other, but I can't find proof of that now as the DNS admin doesn't actually believe it has any drives fitted, and therefore won't provide any information on them.

With respect to back-ups, I used to do a nightly automatic back-up to a separate drive - but that lapsed. I don't know how long ago. Could be as long as a year ago. The drive still exists, but its PSU seems to have been fried in the same surge and we don't have another of the same specification. We do believe the drive itself is functional, though, and part of my shopping trip tomorrow will be to get a new PSU for it.

I realise the lack of a proper back-up system leaves me open to a lot of perfectly legitimate criticism. There's not a lot I can say at this point, except that nothing anyone can say to me could be harsher than what I've been saying to myself.

[FurryNutz]

The age could be one reason...

I'm not sure exactly when we got the DNS but it's at least 10 years and perhaps 12. Same drives throughout, but I don't know what they are as I haven't opened it up yet. Again, tomorrow.

[andy-d]

The age could be one reason...

Well, certainly, none of us are getting any younger

This morning I've powered it up again and reconnected it to the network. It's on a static IP and can be found by a LAN scanner on its usual address. The admin area works fine, but that's all. A second pair of 'HDD2 Hard Drive Has Failed - HDD1 Hard Drive Has Failed' log entries have appeared. It's reporting an internal temperature of 96F/36C, which is lower than yesterday, but there's no warmth coming out of the unit. The drives are not making any noise (usually they sound like log chippers, and always have). Under 'total drive(s)' it still reports zero, and 'physical disk info' has empty entries for both left and right.

Can hard drives be 'parked' by power surges? Could it possibly be that the unit simply needs to somehow rediscover the drives and all will be well, and I can go off and play with the unicorns that will undoubtedly exist in that impossibly magical world? Yeah, obviously not...

[andy-d]

What is the mfr and model of the drives.

Samsung HD502IJ (both of them).

[ivan]

A couple of thing to check.

1) check the power supply (brick) and make sure it is supplying both 12v @ 3A and 5v @ 3A (you might not be able to check the Amps without proper test equipment).

2) with the unit powered on check to see if the drives are spinning (you can do this by removing the front panel and placing one end of a small screwdriver on a disk and the handle in your ear, you should hear a noise if the disk is spinning).

Reasons for these checks:

1) since you are able to access the webUI it indicates the NAS electronics are mostly working.
2) the fact that the webUI indicated the disks have failed points to the 12v supply from the power supply has failed (the disks are the only things that use the 12v supply).
3) if you can't hear the disks spinning it also indicates the 12v supply has failed or the disks are really dead. 

You don't say how your disks are setup, RAID 0/1, JBOD or Single disks.  This information is needed to help point you in the right direction if recovery is necessary (replacing a duff power supply results in the NAS coming back on line without problems in about 60% of the cases).
 

[andy-d]

Ivan, thank you for that. The power supply failure sounds like good advice as something to pursue (I'm thinking of writing all this up afterwards as a guide for someone in my situation). I think however that, since my last post, I've got to a point where I can rule that out without specifically checking for it.

Following the advice elsewhere on this forum, I installed a third-party application that would allow my Win10 box to read Linux-format disks - not Ext2IFS, which crashed the PC on installation, but DiskInternals Linux Reader, which seems pretty good. And I went out and bought a docking station.

The result: one drive is readable in DiskInternals Linux Reader, one isn't.

The left of the two drives as you look at the front of the DNS is the readable one. I'm guessing that's HDD1, which is the one that failed second. DiskInternals Linux Reader gives an error message saying it's damaged but repairable.

The other shows up as a drive but can't be read. It also has some mottling to its casing over the disk itself - could that be heat damage? I have no idea, but the other one has only a small amount of it.

The drive that can be read seems completely intact - I can drill down to individual files and so far haven't found any part of the disk I can't access. So I'm very hopeful that I can copy its contents to another drive and then thank the stars for my lucky escape. I'm not taking anything for granted yet, but at least I know everything seems to be there.

Still some work to do, but it looks encouraging.

Oh, and the other back-up drive that I thought hadn't run for a year? March, as it turns out. Could have been worse.

Thank you both - I need to get on with fixing this now! Will report back on progress...

EDIT: I see from another thread here that HDD2 is the left drive. That kind of makes sense in this context, I guess. Anyway, after an 11 hour transfer it looks like I have managed to copy the whole contents of the drive (more than 350,000 files) onto back-up storage.