Marvel RAID problem: Drive or controller

[jgourlay]

I have a RAID 1 with Marvel controller that keeps getting 'degraded'. I can 'rebuild' it with the same driveds and it will work for a while, then fails again. No data loss. How do I figure out whether (and which) of the two drives are bad or determine if it's the controller going bad?

 

[SkyNetRising]

Boot into RAID bios and check drive identification data. It should show, which drive it considers failed.

 

[jgourlay]

Right, I've done that a couple of times. It shows the array as "degraded", but doesn't show which drive is the culprit. Unless there is a setting/menu option somewhere deeper allowing diagnosis?

It will 'rebuild' successfully, boot up into windows. Then 5 to 20 minutes of use and the machine reboots and the array is 'degraded' again.

 

[SkyNetRising]

Can you make a photo or RAID bios screen?

 

[jgourlay]

Here are all the choices I think I have.

 

[SkyNetRising]

Since your drives are identical, you can distinguish them by serial only.
K1GVE0PB is the old drive.
K1GUHBGB is the one, that is being rebuilt.

Serial is printed on the drive BTW.

 

[jgourlay]

Thanks for that! Now it looks like both are fubar'd: it won't boot at all.

 

[jgourlay]

New information. I bought two new drives, set them up as RAID 1, and restored the system image (macrium) to the array. It would not boot, I'm guessing, because it was not named the same as what the image expected. However, in the RAID bios, it showed the 'old' array.

So, I deleted the 'new' array from one disk and used it to rebuild the 'old' array. Here is where is gets stranger. It still won't rebuild but the error is different. I can't capture this because it happens too fast. But:
1. set it to rebuild.
2. goes fine from somewhere between 30 seconds and a minute then the 'target' drive just disappears. Blink twice and it reappears as a an 'available' drive.
3. Blink again, and it the screen is back in rebuild mode but now it's 'paused'. If I 'resume' the rebuild it will go back through the cycle again and again.

At this point, if I reboot, it will not boot to windows and has a variety of blue screens. BUT, if I physically TURN OFF the 'target' drive, it will now reboot to windows and operate just fine as a single drive 'degraded' RAID array. Thoughts?

 

[SkyNetRising]

I'd install OS on a different (non raid drive, preferably SSD) and use raid for secondary storage only.

 

[jgourlay]

Thanks, and that's a good change, but would not solve the current issue. What should I try for this?

 

[SkyNetRising]

Why wouldn't it solve it?

• You install OS on a separate drive;
  Create raid on secondary storage drives;
  Restore data from system image on raid drives;

Profit.

 

[jgourlay]

Yes, sorry, I said it wrong. I still haven't been able to figure out if this is a hard drive or a RAID card problem. I mean, this setup has worked for years and across three different machines. So I'm trying to figure out what's wrong.