Jump to content
RESET Forums (homeservershow.com)
BrainDrain

Any insights into weird disk problems with Microserver Gen 8?

Recommended Posts

BrainDrain

In brief, I've had the Gen 8 Microserver since March 2016. With cashback, it only cost me about £120 if I recall correctly — so I'm loathe to upgrade to Gen 10 when it is pushing £500...

 

I'm running Windows Server Essentials 2016. I use ACHI, not RAID — to get redundancy (mainly for ripped movies and client PC backups) I use StableBit DrivePool to mirror two large hard disks. Anything vital or irreplaceable I back up to other drives and to the cloud. I also use StableBit Scanner so that I get notifications when drives show signs of trouble.

 

But, I've been having trouble with my Gen 8 lately. It has been a saga, but the main points are:

 

  1. I had two Seagate IronWolf 4TB drives for nearly 3 years. Then a couple of months ago, Scanner started to send me notifications that indicated that both disks were rapidly accruing unreadable sectors. They did have over 700,000 load cycles each for some unknown reason. They were just inside warranty, and Seagate replaced them with no quibbles. I wanted a bit more room anyway, so rather than swap the replacements in, I bought two new 6TB drives instead.
  2. I've since had about SIX(!) replacement drives from Seagate (after some initial hiccups with their process they've been great). I've repeatedly been through something like this:
    1. I swap in the latest new 6TB drive replacement, and use DrivePool to mirror the data to it from the remaining 'original' 6TB drive (that has the data from the failing 4TB drives).
    2. I have lots of trouble with drives going 'missing' and then 'found' again — sometimes I'd get dozens of notifications about this from DrivePool in a matter of hours. I've tried different internal SATA bays, and an external USB-SATA adapter, but this doesn't seem to make much difference.
    3. When the mirroring does finally complete, and I let Scanner do its thing, it reports unreadable sectors on the new replacement disks too. 😞
    4. Then when I run the Long Generic test with Seagate's own Seatools software, the replacement disk would fail just like the others before it. (I tested one of the latest disk before the mirroring and it passed, only to fail after the unreadable sectors show up after mirroring). By the way, Seagate did mention that Seatools "isn't intended for this environment" (running on the server itself, or via USB adapter), which I don't really understand — are we all supposed to have some separate test rig just to test drives with Seatools?
  3. Note that those six replacement drives include one that was definitely from a different batch, and one that was an entirely different model (the Pro series). Same problems.
  4. Clearly, at this point, it is no longer reasonable to think that the drives are the root cause of the fault.
  5. StableBit have suggested that an unstable controller or dodgy cables/connections might cause this sort of fault — but I can't really test this because the Microserver isn't like a desktop PC where you can swap cables around and such, and it all seems pretty solid (however, see below).
  6. After the latest mirroring and 'unreadable sectors' notifications, I did take the cover off of the server and wiggle the SATA data and power cables. Weirdly, whether it is coincidence or not, I haven't gotten any more 'missing disk' or other notifications since then. I tried stress testing for a few hours with 'burst' tests on the disks, and didn't get any notifications.
  7. HP have been useless. They took a quick look at some log I sent, said they can't see anything wrong from the limited data those provide. And then they effectively washed their hands with it with some lame excuses — they can't/won't help because I'm using drives that are 'not compatible' (ie, not on their select list of pre-approved drives for their arse-covering). Bullshit — it is surely expected that people will use newer bigger drives at some point in the life of the PC, and it's all standards-based (SATA, etcetera).

 

This misses some details, but is long enough already. Can you offer any insights or advice based on your own experiences with anything similar?

 

Share this post


Link to post
Share on other sites
BrainDrain

In case anyone else comes looking with similar troubles: All these weird issues cleared up after I wiggled all the SATA and power cables. 🙄 I tried that before, but must've been more thorough in my desperation the second time.

 

I can't say for sure, but I suspect that the dodgy connection might've been where the motherboard joins the 1-to-many SATA cable. The disks are fine.

 

It's been stressful and odd, but I'm relieved to have a working system again.

 

As mentioned, Seagate have been brilliant. StableBit (DrivePool and Scanner software) were also great. HP were useless - I mean, I know it is out of warranty, but some troubleshooting information or something would've been nice rather than a cursory look at some logs and then "We can't help because you aren't using compatible [ie, pre-approved] disks".

  • Like 1
  • Thanks 1

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...