JMFord wrote:
Along with that the the event log shows {Delayed Write Failed} messages.
Well done for spotting this. Unfortunately that is a very general Windows error message, basically saying that a write which had been cached on the host (often a filesystem metadata update) could not be
successfully written to the storage at a later time - but at the level of that message, it doesn't know (or care) exactly why the write failed. The consequence will be exactly as you describe - parts of the filesystem will become inaccessible to Windows, as these filesystem updates fail. Are there any other odd event log messages shortly before those ones?
JMFord wrote:
The strange thing is that it I reboot the machine the files return in all their glory.
That is expected for intermittent errors, until the next filesystem metadata update failure occurs.
JMFord wrote:
From what I can gather I mostly likely have a firmware, controller, or power supply problem, but I can't be sure.
Actually the list of suspects is longer (e.g. could be internal drive problems) - unfortunately
any problem which causes an I/O failure on a metadata update, can cause the results which you describe. A drive firmware problem is highly unlikely, as this drive has been working OK for a while.
JMFord wrote:
Are there any further tests I can run to troubleshoot this further?
As the data on the drive isn't needed and we don't have to worry about the possibility of "diagnosing the drive to death", I suggest gathering the full SMART data from the drive, including raw attribute values (smartctl (specifically smartctl -x) from smartmontools is one suitable utility, but there are many others), so that this data can be reviewed. The advantage of running smartctl -x is that the drive's [standard] internal error log is also collected, and that can give better insight about a potential problem than just the SMART data, in some cases.
After collecting that SMART data (and ideally the internal error log) from the drive, you could download & run the full WD DLG (Data LifeGuard) test and see what result you get. However if the problem with the drive is specific to
writes, then it is possible that this test will pass (as always, a test which passes does not prove that everything is OK).
Hope that gives you some ideas.