0

I have a broken NAS. The system is not broken but the hard drive is. The NAS system is ASUSTOR AS5108T with 8 bay of HDDs. The NAS originally purchased with 8 units 3TB WD GREEN (I know it is wrong to having WD Greens for frequently-accessed NAS but thats what shipped until I know about this case). The HDDs also setup with RAID 5.

  1. When the NAS is coming, our IT guy immediately setting up all computers to backup to the NAS. Until this point I did not know much about the NAS let alone the WD Greens. I am just a normal employee following orders from the IT guy.

  2. Once all computers were connected to the NAS, IIRC the IT guy told all employees to do backup. Backup size is various, but lets say all employees has ~50GB of data, x ~20 employees. Until this point I did not know much about the NAS let alone the WD Greens. But, ALL computers are doing backup altogether. (imagine all those WD greens enforced to receive data from 20 computers!)

  3. And I exactly remember when all computers are still doing backup, the NAS was down for a while. I don't know what the IT doing but the NAS is eventually back on.

  4. ...weeks later, the NAS is down. From what I've known, the NAS is down (not dead). I have external HDD provided by the IT before the NAS was purchased so I regularly doing backup there.

  5. Just by this week, I am asking about the NAS to the new IT guy. The old IT guy was moved to another location. Starting this point, I know the details about the NAS as I described in the first paragraph of this post.

At the moment my new IT guy did not have guts to turning the NAS on because he said the last time NAS turned on, it cannot rebuild the RAID for around 15%. And oh now the NAS is now equipped with 7 WD Green and 1 WD Blue (the Blue was intended to make the rebuild success because the original WD green placed there is dead). Both me and the IT guy afraid there will be a data loss when turning the NAS on again (Me and others still have some data there which needs to be accessed).

I suspect the rebuild failed because the WD Green's power saving feature which prevents the RAID system to completely rebuild.

I am aware that the power saving feature on the WD green can be turned off (https://wdullaer.com/blog/2015/04/05/hack-your-wd-greens/) so I am thinking theoritically this can help the rebuild. My plan:

  1. Unplug all the HDD from the NAS and plugging it into my Linux system without mounting it
  2. Disabling the power management on the HDD via idle3-tools for each HDD.
  3. Plug all HDDs back to the NAS and do the rebuild.

I believe this action is not modifying filesystem and RAID inside. Will this work?

I need suggestion to recover data from the NAS from you guys. Once important data obtained I will switch to WD Red.

Let me know if you guys need more details.

Thanks!

8
  • Okay, why downvote? Jun 9, 2017 at 4:43
  • 1
    Your plan probably won't hurt, but we'd need to know more about how it failed to really say.
    – Bill Weiss
    Jun 9, 2017 at 5:53
  • The best way to ensure that there will be no further data loss is to take a copy of each drive with dd to an external location, and work with the image files. Jun 9, 2017 at 7:09
  • Thank you - I like to bookmark cases like yours for future users asking about using RAID 5 - it's been essentially dead/useless for nearly a decade now yet people still don't do their homework, still use it and pay the price. Take some relief in knowing that your pain may act as a cautionary-tale for others.
    – Chopper3
    Jun 9, 2017 at 8:24
  • @BillWeiss My IT did not allow me to turn on the NAS again but what is he saying the rebuild just failed and the NAS doing nothing anymore. Jun 12, 2017 at 6:03

2 Answers 2

2

It is not possible for the rebuild to fail due to a disk entering power saving mode. After all, a rebuild is a very stressful operation, so the disks literally have no time entering standby.

If the rebuild really fails, it must be due to some other problems, for example: - an URE problem on one of the original disks (WD Greens) - a NAS hardware failure (ie: bad RAM) - a broken replacing disk (the WD Blue)

I strongly suggest you contacting a local hardware expert / sysadmin to assist you with the case.

1
  • Thank you. The IT saying that he would choose the company to learn the $3000 lesson for data recovery. Jun 12, 2017 at 6:04
0

If rebuild fails at 15% it means one of the other drives has at least an unreadable sector.

A way to check this is to remove them all, connect them one by one via USB adapter without altering anything on them, have them separately scanned for bad sectors, isolate or repair the bad sector and then put them back together.

You should also recommend 2 things for your IT people: never use WD Greens or Seagates as heavy-duty NAS and don't use RAID 5 as it takes quite a lot to rebuild. Use RAID 10 or other variations instead.

1
  • Thanks for the tips -- unfortunately the IT did not allow me to unplug all the HDDs. I've just describing why in the new comments above. Jun 12, 2017 at 6:05

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .