All posts are moderated, sorry. It prevents spam posters, which there are a lot, even on a specialized forum like this.
I asked about encryption, as there is a known kernel bug with encryption on SoftRAID volumes,we are trying to get fixed on Apple's end.
I do not know what is causing your panics, and it is frustrating, as we cannot investigate without a functional debugger on the M1. (there is no way to trace through the events to see what happened)
1.
Kext user-space log filter changed from 0xff2 to 0xfff.
Kext kernel-space log filter changed from 0xff2 to 0xfff.
Kext library architecture set to arm64e.
Executing: /usr/bin/kmutil load -p /Library/Extensions/SoftRAID.kext
Error Domain=KMErrorDomain Code=71 "Incompatible architecture: Binary is for x86_64, but needed arch arm64e
Incompatible architecture: Binary is for x86_64, but needed arch arm64e
Incompatible architecture: Binary is for x86_64, but needed arch arm64e" UserInfo={NSLocalizedDescription=Incompatible architecture: Binary is for x86_64, but needed arch arm64e
Incompatible architecture: Binary is for x86_64, but needed arch arm64e
Incompatible architecture: Binary is for x86_64, but needed arch arm64e}
2. com.apple.quarantine
After initially migrating my original account over from my Intel mini, I realized this might be part of the problem. I deleted the entire drive via Recovery, then restored via Apple Configurator. That went well. I created an account, logged in, rebooted to recovery to set the kernel permissions, then rebooted and verified full disk access, approved the developer, and installed SoftRAID driver. Didn't improve anything.
What's interesting to me is that I see the exact 541 kext extensions in the System/Library/Extensions of my other fresh install M1 mini, and also the same as my work Intel MBA. The arcMRS and PromiseSTEX extensions are the only two with no/no for notarized/loaded, which led me to believe these drivers were causing the problem like you mentioned in this thread earlier (or another, can't recall as I've looked at all the M1 threads here in the past 2 days). However, I see no way to delete system loaded extensions on an M1 Mac via terminal.
Trying again with HFS+
It should take ~6+ hours to copy ~900GB
I'll see what happens
This should be near the top of Apple's priority list. We do not have any idea when. They do not give estimates.
I have performed intensive data tests on a Thunderbay RAID 5 volume on the M1 and did not see any issues. that does not mean there are none, but we use a similar test to Apple's internal tests. And we run it for long periods of time.
If you boot into internet recovery mode, you can do it in terminal with rm -r then refresh the cache's. Let me investigate if there is something else you can do to make this work.
I guess you did it with HFS+
I'm trying it now. Can it be some HW issues with Thunderbay enclosure, or you would be able to see it with report file?
We definitely tested with APFS also. It is unlikely to be the enclosure, it would have different symptoms. Its possible a disk could be hanging, but SoftRAID should report an error if so.
It is very hard to know. Hope your file copy succeeds.
I guess another test is the same copy using "finder"
I tried running the commands from recovery mode. The two kext files remain. The other two kext commands are not recognized in recovery terminal, but I ran them upon reboot. Then installed the SoftRAID driver again.
The SoftRAID driver then appears in the system list as Notarized and Loaded, but as soon as I plug in either one of my RAID10 SoftRAID devices my system panics and restarts.
I'm at a loss. I don't understand how it's still loading an x86 version of the driver each time. It doesn't make any sense that there'd even be a way for the x86 driver to be installed on the M1. But this is clearly where it's breaking down IMO. Whether that's due to the other two non-loaded system extensions or the SoftRaid driver installation process, I have no idea but this is a fresh install with all system apps and only SoftRAID installed and I've hit the same result over 20 times now.
This is Thunderbolt, and 11.1, correct? Directly connected? No other devices?
If you connect only one disk (no volumes mount) do you get the panic?
I don't think this message means anything, as I think it is complaining of the "fat" nature of the driver. (I am checking, will post when I get an answer, which may not be today)
What if you disable Automatic rebuilds in RAID preferences?
Well this is good news! Disabling Automatic Rebuilds in RAID preferences has my Thunderbay 4 mounting, but not my other non-OWC SoftRaid 2-bay enclosure.
What is up with the other one? If you expand the volume tile, it has a "disk Identifier". correct?
Does it just not mount, or also crash the machine?
@softraid-support yes the USB3 volume is crashing. It appears for about 3-5s in SoftRAID before the panic occurs. I see the volume name and information present, and it shows as "unmounted".