SoftRAID drives missing disks/failed upon screen waking
When I woke my laptop monitor this morning, I saw that several (but not all) of my disks were missing in my two SoftRAID RAID5 volumes. Both volumes said "Missing Disks" and one of them also said "Failed," which I've never seen before (and is very disconcerting).
Restarting the computer caused a panic/restart. Upon second restart, everything came back online. One of the drives rebuilt for 1.5 minutes and now everything appears as normal.
OSX 12.3.1/ SoftRAID 6.2.1/ MacBook Pro M1 Max/ All volumes connected via one thunderbolt daisy chain; running to the laptop via a TB2->TB3 adapter.
I have "Prevent Mac from sleeping with monitor off" checked, and "Put hard drives to sleep" unchecked.
Is there anything I'm doing wrong here? I'm very concerned about damaging my volumes.
You have not done anything wrong based on your description.
Is there any chance the cables moved while the computer was sleeping?
If not, I have to guess this was a one time event, especially that not all the drives were "missing". That would be very unusual.
@softraid-support The laptop shouldn't have moved; I'm the only one who uses my office. Cables all appeared secure this morning.
One very strange wrinkle about the missing disks: just one was missing from each volume, but all my enclosures have even numbers of disks (either two or four). Does that point to anything or just further muddy the waters?
that is what makes this seem like a one off event. When the problem is caused by Apple Sleep, all the disks in the enclosure(s) will eject and you will get a notification on each disk.
Less than all the disks is highly unusual. I do not have an explanation. Perhaps a minor "brown out"? Its recommended ini any case to get a power conditioner on your drives and computer, it can help your electronics live a lot longer by supplying consistent power.
I have an update. The problem has happened twice since my last reply.
Just as before, only some of the disks were missing. These two times, even numbers of disks were missing so were likely all from one or two enclosures.
The most recent event popped an error saying that a disk had hung during a read and needed replacing; however, upon restarting the computer all the disks and volumes appear normal with no errors.
I noticed some behavior when I was in the office today before I woke my laptop's screen that may be pertinent. It sounded as if all the volumes were spun down, then every few minutes, they would spin up for about 30s-1 minute and spin down again. This isn't behavior I ever noticed with my desktop, even when it was sleeping. It seems to be new behavior specific to the laptop.
I normally leave everything up and running during the night so that Backblaze can back up all my new files (I'm a photographer and routinely upload about 100GB/day to these volumes). I'm thinking I need to unmount and power-off all the volumes to avoid damaging anything.
Any ideas? I'm happy to provide any information you need to get to the bottom of this.
the sleep'wake cycle is Monterey. We all have to wait for Apple to fix this.
One or two drives ejecting is the real problem. Can you give all your disks SoftRAID disk labels?
I want to track two things:
Are all the ejects on the same disks?
Are they on the same "side" of the enclosure (Use Blink Disk light to check location of the drives)
I.e, in a 4 bay thunderbay AB and CD are on separate PCI buses. Do you ever get ejects on AD for instance? Separate ends?
@softraid-support Understood on the sleep/wake cycle. What an annoyance.
I'll work on labeling all the disks in SoftRAID. I do have a question: when using Blink Disk Light, all of my OWC enclosures blink the "Activity" light (or single blue light, depending on the age of the model) rather than an individual disk light. Is there a way to discern which disk is causing the blink rather than just which enclosure?
I may have explained poorly. Rather than the blue Disk 1 or Disk 2 lights blinking, it's the amber Activity light. In a two-bay enclosure, that just tells me one of the two is active but not which one.
I have a Thunderbay, which does seem to work as expected with "Blink Disk Light."
But I also use four Mercury Elite Pro Duals (newer and older TB2 models); none of those are blinking individual disk lights--rather they're all blinking just an "Activity" light which doesn't indicate which disk is in use. I can figure it out by taking each one apart one at a time and removing a single disk, but that isn't ideal with eight different disks.
It has four lights on the front: Power, 1, 2, and Activity. Activity is the only one that lights up during Blink Disk Light.
I'm having more problems now, which may or may not be related. I got a "Many SoftRAID Disk Errors" message today which said one of my disks encountered 2 or more read/write errors in the past 24 hours. I'm trying to figure out which one. I can see it in SoftRAID and it says "no errors" and has passed the SMART status test. It is currently rebuilding.
One extra issue; as soon as that happened, I attempted to unmount my other drive through SoftRAID so I could power them off and troubleshoot. As soon as I tried, my computer froze and had a kernel panic. I'm attaching the log of that here.
Any ideas what's going on?
I just pulled one off the shelf and the 2 middle LED's flash with lights with Blink Disk light.
Attach a support file, let me see what is going on. the panic was a file system (probably a hang reading the volume) problem.