Mystery Slow Down - usually on waking from sleep

This time it actually led to a crash report. Again, happened when waking the computer from sleep.

SoftRAID shows the following error message and forced restarting SoftRAID:

SoftRAID Error

An internal part of the SoftRAID Monitor has stopped functioning properly. Please quit SoftRAID and relaunch it.

@eMouse

Instead, open Activity Monitor. Sort by CPU. What is taking up the most CPU or threads? is any process showing as hung?

This is in the panic log:

userspace watchdog timeout: no successful checkins from WindowServer in 128 seconds

service returned not alive with context : is_alive_func returned unhealthy : WindowServer initialization not complete (post IOKitWaitQuiet)

This panic means something is timed out, could be a disk, or at minimum a "write command" to a disk.

@softraid-support I've had activity monitor open, but it never indicates a definite issue with CPU usage. It will retroactively indicate system usage spikes when it does update, but with no particularly predictable process at the top of the % CPU list.

I found a discussion on the MacRumors forum that the SoftRAID SMART status was not necessarily accurate, and the user recommended using DriveDX. I installed that and it indicates that one of the drives in the Thunderbay 4 Mini array is having issues, while SoftRAID isn't indicating any issues. The exact problem is "UDMA CRC Error Count" with a raw value of 11 (all other disks in the array and connected to the computer show a value of 0). The description of the error indicates that it's probably not an error with the drive itself (which is why SoftRAID might not be detecting it), but with the drive's connection to the computer. Given that out of 4 drives in the enclosure, only this one is indicating such an error, that appears to point to an issue with the enclosure itself?

I guess I'll take it up with hardware tech support.

And here's another report from SoftRAID since hung and forced a self-reboot again. No OS error report popped up.

@eMouse

With regards to SMART, there have been two major studies with disks and the only two factors in both were related to reallocated sectors.There are lots of SMART parameters that may or may not point to failing drives, but have not been shown to do so in studies. this includes things like power on cycles, spin up speeds, temperature, etc. We are not trying to point to every possible potential issue with disks, we want to track ones specifically correlated to failures. DriveDx is fun to use, its geeky, etc. I do not know if they are tracking cummulative statistics or not. if they are, then they have data which could be very valuable to try to identify disks on a failure path. I do like that DriveDx has gotten SMART over USB working, while we have not, so that is a huge plus for that product.

Its possible that those errors are from cable, yes. But the stall is interesting, it generally means retries if it was disks, but then unreliable sector counts would go up. So what is causing this, I do not have any solid ideas.

In order to get some work done, I unmounted the Thunderbay, but left it connected, and haven't had any further issues. SoftRAID is still installed, though, presumably, it's not being engaged by anything. I've reconnected the enclosure to see if it stalls out again, and if that increments the UDMA CRC Error Count. In the mean time, I have reached out to hardware support to see what they can suggest to confirm if it's a hardware issue and how to handle it.

To give one more update, in the process of trying to identify if it was a problem with the physical enclosure, I reseated the drive that indicated some sort of potential error. Since then I haven't had any further issues with slow downs or crashes. It's looking like that was probably the problem.