Classicamiga Forum Retro Edition
1 2 3
Thread: NAS problems
Demon Cleaner 19:46 9th June 2015
One disk on my Synology 2413+ started with bad sector counts some time ago, and since 3 days it gives me I/O errors.

I get an health report every first day of the month, and since the 1.08.2014, disk 1 has 1 bad sector count.

Originally Posted by :
Disk1:
Disk Reconnection Count 0
Bad Sector Count 1
Disk Re-identification Count 0
I thought that it might be not so bad, and it also stayed like this for a while. But in the health report of January this year, the bad sector count had increased to 50. One month later it was at 83, where it stalled until this day.

Unfortunately since last Saturday I also get I/O errors from that disk, and that once a day:
Originally Posted by :
Dear user,

The hard disk 1 on Synology2413 had an I/O error, but it is working properly now after several retries. It might have been caused by bad sectors. If this error occurs again, please back up your data and run the S.M.A.R.T. test on your hard drive to examine the hard drive status.

Sincerely,
Synology DiskStation
S.M.A.R.T. test is normal and doesn't show anything unusual, status is green.

As the bad sectors get locked, and at the moment the count is not increasing, should I be worried? Do you think I should change the disk anyway?

You know that I'm running the NAS as one single volume in Raid 6 mode. Could I also use f.ex. a WD Red disk as replacement, as I'm using 12 x 4TB Hitachi Deskstar™ 5K4000 (HDS5C4040ALE630), which are not available anymore, does that matter?

Here's an excerpt of my log file:


[Reply]
Harrison 01:31 10th June 2015
I would definitely swap the disk with a new one if possible as it is slowly signalling failure. I think as long as the replacement drive is the same size it should be fine and just rebuild the set and reintegrate the disk into the set. With RAID 6 if you take the disk out the NAS set should function OK until the new disk is ready.

How old is the drive? Most drives have 3-5 year warranties.
[Reply]
Kin Hell 09:01 10th June 2015
Golden rule for ANY hard drive showing bad tracks or sectors is quite simple. Bin it & replace it asap.

What actually causes the bad sectors are bits of the ceramic platter/s coming away from the rest of the surface. These bits are always floating around inside the hard drive & are just waiting to start even more carnage. They get dragged across the platters during normal access procedures on the heads causing the continuous damage.

It really does not matter what mode the drive is configured in. Throw it in the bin or return it under warranty if applicable.

Also, S.M.A.R.T. isn't always as clever as it claims to be. It is a burden on your systems performance too, as it's always looking for problems. I never run S.M.A.R.T. on my systems as you'll know when something is wrong when trying to access the drive either on read or writes. Modern NAS boxes will give you an audible Beep, tho' S.M.A.R.T. isn't detrimental to NAS boxes like Windows based PC's are.

Reads might read okay but then you might have an issue when accessing the file/s just read. (usually corruption)
Writes will give you an instant I/O error every time.

Also on a personal note, I would not be running Hitachi Death Star Hard Drives in a NAS box & more so in a computer in any instance. They were good in the early days of low storage sizes: 20GB or < & went tits up when Maxtor came out with lower than 1" profile 40GB 3.5" HD's.

One good thing here is you're finally getting rid of one Death Star hard drive. They could be beyond their service limit & hopefully not available anymore. Hitachi Hard Drives are the worst hard drives you could ever run in a computer. They probably only did this well cos of the 5400rpm. The 7200's were nothing short of Nasty & their fail rate when in any RAID mode is alarmingly high from my personal experiences & close colleagues.

For your NAS box, pull them all & replace with WD Red immediately. Pro Red's are not necessary.
[Reply]
Harrison 10:50 10th June 2015
Those drives were most likely actually manufactured my WD as they purchased the Hitachi 3.5" HDD business a long time ago, but have now since sold it on again to Toshiba, hence the reason the drives are no longer available.

Originally Posted by :
For your NAS box, pull them all & replace with WD Red immediately
That would be very expensive for DC to do considering his NAS has 12x 4TB disks in it.

WD Reds are definitely the way to go though and I would recommend replacing them with them as they fail/start showing errors.
[Reply]
Demon Cleaner 21:52 10th June 2015
Have to check tonight when I bought the NAS and the drives, maybe I will have a look in the forum, must be a thread about it. I guess they're 2-3 years old (according to the forum, I ordered it the 10th of September 2013, so still warranty).

I never had any problems with my Hitachi HDDs in my Qnap, which I sold to my friend here at work, and I just asked him, and they're still running fine.

Problem is that I heard that the new disk needs to be at least as big as the old one, it must have exact the same amount of sectors. The Hitachi is 3726.02GB but I guess I have to find a datasheet to check about the sectors (EDIT: The Hitachi datasheet doesn't show it).

WD Red has 7,814,037,168 sectors.

- - - Updated - - -

Other question.

As there's warranty on the disk, could I take it out, send it in, wait for replacement, and if I get it back, build it in again? Would the NAS during this time then run in Raid 5 mode?
[Reply]
Harrison 01:33 11th June 2015
With Raid 6 you can suffer a maximum of 2 concurrant drive failures and still maintain the integrity of the array. Therefore taking a single disk out of the array would keep it working. As for it being reduced to Raid 5, in as much as only 1 more drive can fail and the raid array stays intact is correct, so in that regard it would act just like Raid 5.

One thing you could have done with a 12 bay NAS is set 1 or 2 drives aside as hot swappable spares, so they could come into action as soon as a drive fails.
[Reply]
Demon Cleaner 12:23 11th June 2015
I considered doing the hot swap when I set up the NAS, but in the end I decided to go with a normal Raid 6.
[Reply]
Harrison 12:46 11th June 2015
I suppose the downside of hot swap drives is you lose the 8TB of space the 2 drives provide. It depends how much redundancy you want I suppose.

Storage is never fall proof or 100% safe is it. Backups are still the safest way to protect data, but even they can fail. When will someone invent a 100% safe backup media?
[Reply]
Kin Hell 15:06 11th June 2015
@ DC

Fully appreciate where you are with the Sector thing & again I'll add....that's very poor of the Hitachi Hard drives.

I have a 4 bay Qnap which use to have 4 x 3TB Seagate Barracuda's in there. I run two of them as JBOD's & the other two in RAID 1.
I acquired the device like this from a close friend who also had one for himself. About 12 weeks ago, one of my Barracuda's failed & thankfully it was one of the RAID 1 drives.
Went & bought a 3TB WD Red, popped it in, updated the drive management, the NAS box did it's thing & re-built the array.

If Hitachi drives are not sticking to Industry standards, it would appear you are in a never ending cycle of using their drives & what are you going to do when they are no longer available?

Might be an Idea to jump ship now so to speak. Lets say you got a WD Red, could you transfer data from one of your Stand-alone Hitachi's to the WD Red & then throw the donor Hitachi back into your RAID config?

Not sure your RAID config will allow this, but just a thought fella.

PS. My close friend's identical NAS box killed all 4 of his Barracuda's in 5 days after mine died. He was not a happy chappy to say the least.

/Kin hopes this story doesn't soon destroy his remaining 3 Barracudas......
[Reply]
Demon Cleaner 19:05 11th June 2015
Originally Posted by Kin Hell:
Went & bought a 3TB WD Red, popped it in, updated the drive management, the NAS box did it's thing & re-built the array.

If Hitachi drives are not sticking to Industry standards, it would appear you are in a never ending cycle of using their drives & what are you going to do when they are no longer available?
I didn't mean this only with the Hitachi disks, but I thought this was a COMMON problem? I mean a common NAS problem, not related to the disks.

One guy from the Synology forum told me just to order one WD Red, pop it in, and if it doesn't work, just send it back. I didn't want to do that, as I think that's crazy, not gonna order to try out something, I wanna know for sure in advance.
[Reply]
Tags:Array
1 2 3
Up