Print

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

[Oman]

I know the original Deskstar 7K1000.B thread is locked, but I wanted to make sure some additional information gets reported.  The soft restart does get the drives spun up and recognized without issue but in testing 2 of the Deskstar drives I have found that there is an additional problem as well.  After about 10 minutes of I/O to the drive the NAS will fail the Deskstar drive, and then will fail the other drive as well (even if it is a drive that is not a problem drive).

This is with the 1.06 firmware.  Both of the test drives seem to have the same issue not matter what bay and if there is / is not another drive in the other bay and no matter what the RAID settings used.

Jon

[D-Link Multimedia]

Anyone else seeing this issue? I have the official firmware patched running and it has been running for well over a week now without any form of drive failure.

Can you give details on the environment...is it random? Is it always @ 10 minutes? As many details as possible would be best so we can attempt to reproduce.

[eric111]

Hi all,

New user here ;-)

I have exactly the same drives, firmware is 1.06, used RAID1 first and just have changed to 2 drives.
I don't have the problem mentioned posted by Oman.

But i do have the problem of recognizing the drives only after a soft reboot,
so i really want to have the new firmware as fast as possible ;-)

Thanks,
Eric

[ttmcmurry]

Resurgence of the "Deathstar" ?

The IBM Deskstar 75GXP, and several other models made around the same time, became infamous for their reportedly high failure rates. This led to the drives being colloquially referred to as "Deathstars".

Source:  http://en.wikipedia.org/wiki/Deskstar_75GXP

[khoult]

I have these drives as well, and they do have the usual problem with needing a restart after powerup to be detected.

I don't normally have the problem with the drives going away that you mention. With the standard 1.06 firmware they seem solid.

 I've been able to cause them to go away by installing twonky. With twonky enabled, the drives will disappear about 5 minutes after the restart. This seems to be due to the root partition filling up problem, not the drives themselves through. I'm working through the config changes to try and fix the full-root.

 If I disable twonky, all is good again (even leaving funplug in there).

[Oman]

I have 1.06 installed.  I spent some time testing and I think the problem is with I/O to the 7K1000 drives.  It seems the drive goes into "fail" when a large amount of data is written over a period of time.  I happen to be replacing two Maxtor 750GB drives with these 7K1000 drives in bay 2 as a backup drive.  The way I use the backup drive is to write to it from a backup program on Windows that writes a single very large file for the backup (Acronis True Image).  The drive does not seem to fail unless I start the backup process.  It then fails within about 10 minutes.

If I write and move a few medium size files around everything is fine.  When I move a giant file to the drive it will fail (say about 250GB).  When it fails the drive light goes from the usual blue I/O blinking to a red constant blink and then the NAS sends a drive failure email.  Within a second or so the other drive (the Maxtor 750GB) drive also goes into the red blink failure mode.

If I swap the 7K1000 for a Maxtor 750 the NAS works perfectly (so there are two of them installed) the NAS works perfectly.

If I take the 7K1000 and connect it to my Ubuntu box I can read and write the 250GB file without problems.  I have two of the 7K1000 drives and they both seem to work fine with my Ubuntu box and they both fail the same way within the NAS.

In order to ever see the 7K1000 drives I have to soft restart the NAS.  One of the things that occurred to after I did the testing last night was that I *do* have drive spin-down enabled in the NAS.  This is because the backup drive is used for about 6 total hours per week.  I will tonight do some more testing including disabling the spin-down.

If you have anything else you want me to try I would be happy to.  As well if you want to send me firmware to test I can do an NDA or whatever you need.


Edit: I am not using funplug or any other mods.  I have FTP enabled with one user, iTunes sharing enabled (on the Maxtor drive), one user and one share on the 7K1000 drive.



Jon

Anyone else seeing this issue? I have the official firmware patched running and it has been running for well over a week now without any form of drive failure.

Can you give details on the environment...is it random? Is it always @ 10 minutes? As many details as possible would be best so we can attempt to reproduce.

[D-Link Multimedia]

What kind of file could you possibly have that is 250GB in size as a SINGLE file? Even image files of OS's are usualy broken down unless Acronis is doing it as single file ?

Or are you talking about just transfering 250GB of data in one session?

[Oman]

Single file backups from systems.  I have them as large as 400gigs.  So long as the file system handles the file size I find that testing a single large file is better than many smaller ones.  It eliminates some of the variables in the testing process.

I also have a few virtual machine files that I use for testing software within various OSs that are in the dozen or so gigabyte size.

I'm a software engineer and a photographer.  I shoot all my work using Nikon RAW format, which are quite large.  Take a collection of 30 thousand or so RAW files and that adds up.  I can take the whole set of image files as well as their associated management DB and meta information and back them up to one single Acronis image file that can then be restored to any other machine at any time, completely intact.

Jon

What kind of file could you possibly have that is 250GB in size as a SINGLE file? Even image files of OS's are usualy broken down unless Acronis is doing it as single file ?

Or are you talking about just transfering 250GB of data in one session?

[Oman]

I reset to factory defaults and tried the 7K1000.  Other than the power-up problem the drives seem to be working.  The only two settings that appear to be different are the power save and the iTunes server.  I'll try turning them on individually to see what happens.

I assume that if you have resolved the boot start up issue then spinning up after power save should be working as well.  It is critical that the power save function works properly because the backup drive is only active 2 times per week.  Not spinning means that the drive is in a mechanically safe state, and also that wear is reduced (drive MTBF is almost completely mechanical in nature).



Jon

What kind of file could you possibly have that is 250GB in size as a SINGLE file? Even image files of OS's are usualy broken down unless Acronis is doing it as single file ?

Or are you talking about just transfering 250GB of data in one session?

[Oman]

I would like to say... Thanks for the update!  So far it seems to be working perfectly with the 7K1000.B


Jon

[eric111]

Yep, the same counts for me! Thanks!

Cheers,
Eric