Win10 PC keeps losing track of WD Black 4TB HDDs; plus file corruption issues

The Colclough

Prominent
Feb 18, 2017
3
0
510
I had a PC custom-built to my requirements in summer 2015, with an Asus X99 mobo, Corsair RM850 PSU, i7-5820K 6-core CPU, 4x 8GB Ballistix Sport DDR4 RAM, a Samsung 250GB SSD boot drive, LG Blu-ray drive, Akasa reader for SD cards etc, second-hand Nvidia GPU (670 series, I think), and - the bit that's giving me grief now - 2x 4TB WD Blacks to store the working files for my video production projects.

I initially installed Win7 Pro but immediately used the free upgrade to Win10 Pro, as that was cheaper than buying Win10 outright; don't know whether that will be relevant but I'm trying to cover all my bases.

I had the Blacks set up in RAID 1 (using the Windows RAID system rather than the BIOS one, as it was easier to set up), because I wanted the extra security of automated backups for my project files after spending quite a bit of money on one of my videos; earlier today, while looking for threads here relevant to my problem, I discovered one that mentioned Blacks aren't designed for RAID, but I wasn't aware of that when the machine was built so I went and RAIDed them.

Anyway, it all worked fine (give or take the usual Windows quirks) until December 2016. And then one day I decided to record a walk-through video for one of my Portal 2 custom maps, using Shadowplay and Audacity, which went fine up to the point where I tried to alter the in-game music volume - and BSOD. The PC hung again during the first reboot or two, but then came back online and seemed alright for a few days. Probably worth pointing out that my Steam directory and Shadowplay recording location were both on the RAID.

I then started getting error messages saying my RAID wasn't working. I started finding corrupted files (including the Audacity project for the ill-fated walkthrough video), and sometimes Disk Management would report one of the disks as Foreign and/or Missing.

Just before New Year, I physically identified which disk was which, and then broke the mirror, re-mounted the failed disk, and rebuilt the mirror from the other one. That seemed to fix things for a few weeks.

However, around mid-late January, I started getting errors again; most commonly, various programs would lock up while trying to read from, or (more often) write to, the RAID, and the system spent hours and hours resynching the two disks even though they'd already been synched as little as a few hours earlier and then had virtually nothing written to them.

I decided to break the mirror again, set the two disks up as independent drives (S: and T: ), and manually backup data between them while keeping an eye out for further problems. In the last week or two in particular, I've had several crashes as one or other disk became unresponsive, always requiring a reboot to get it working again, and after each reboot I've found an increasing number of corrupted files left over from the interrupted write operations - JPEGs with the bottom halves missing, and that sort of thing. Both disks are definitely affected.

Everything else in the system apart from the HDDs is working perfectly - no trouble with the SSD at all, notably. The HDDs are less than 2 years old, and so are well within warranty (unless I've voided it by having them in RAID?), but then again Blacks are supposed to be able to take a hammering and I don't think I've put them under an unreasonable workload anyway.

I'm pretty sure there's a causal relationship between that bad crash in December and the HDD failures I've had since, but I don't know which way round - whether the disks stopped and that killed Portal 2 and Shadowplay, or whether Portal 2 threw an error that somehow affected the disks, although my (admittedly only semi-educated) hunch is more the former than the latter.

I have no prior experience with WD products (used Seagates in my previous PCs) or with RAIDs (not that the disks are in RAID any more anyway), so I really don't know what's going on here. I'm in the middle of checking SATA cable connections and all that, but I was hoping somebody more experienced might be able to shed some light on likely causes?

I've tried to cover all of the pertinent information, but if I've missed anything important then I'll try and fill it in.

Thanks in advance!
 

The Colclough

Prominent
Feb 18, 2017
3
0
510


Sorry, forgot to mention I've already run Data Lifeguard (about a month ago), and it said both disks were working fine. I might give it another try and see if it changes its mind...
 

The Colclough

Prominent
Feb 18, 2017
3
0
510
update: I've tried running Data Lifeguard again, and it gave both disks a clean bill of health again. Haven't had any crashes in the last week or two, but then I haven't been writing much to the drives. Will revisit this next time the crashes start.