[Sticky] Big Sur on M1 machines and SoftRAID issues

@softraid-support

It's about half full

@nwind

Makes sense then, between the physical limitations and fragmentation of files on the volume, you lose speed from the theoretical maximum performance.

@softraid-support

I’ll measure again on an empty one. 
Did I understand correctly that empty 4 disk RAID 5 should have read/write speed ~3x of a single disk or a bit slower?

@nwind

Yes it should be close. CPU is important, as parity has to be calculated, so older computers will not do as well on writes, but newer computers should do very well on writes.

@softraid-support

Thank you!

 I guess M1 should be fine. 

@nwind

It will be, but currently, there is some optimization to do, it is not super fast at present.

@softraid-support

Thank you, but are you sure, because this is the response I get:

sudo kextstat -b com.softraid.driver.SoftRAID

Executing: /usr/bin/kmutil showloaded --bundle-identifier com.softraid.driver.SoftRAID

No variant specified, falling back to release

Index Refs Address            Size       Wired      Name (Version) UUID <Linked Against>

@mountaincat

Yes. We are not correctly setting the version number in the beta, that is a minor bug.

But it shows the driver is indeed loading.

I am back again still trying to get SoftRAID to work on my M1 Mac mini. I have created an absolute pristine installation of macOS 11.2 on an external SSD and to keep it pristine I haven't even logged into my Apple ID.

• I downloaded the b48 version and launched SoftRAID. 
• Approved installing the driver
• I was NOT prompted to change the volume security settings
• I was NOT required to authorize SoftRAID
• The extensions cache was NOT rebuilt
• The SoftRAID app restarted and indicated it is using the 6.0.1b48 driver
• The array displays correctly in the app
• The array will not mount

I have included the log and the support document. This is getting frustrating!

@joemikeb

As a pristine system, you must go through this step again:

(security on mac OS M1 is on a volume by volume basis)

https://support.apple.com/en-lk/guide/mac-help/mchl768f7291/mac
Select reduced security and enable this:
Select the “Allow user management of kernel extensions from identified developers” checkbox to allow installation of software that uses legacy kernel extensions.

Then you must reinstall the SoftRAID driver. Click on the Startup volume in SoftRAID, and "reinstall SoftRAID driver".

Watch for the dialog box for System preferences, you must "Allow" OWC as an identified developer so the driver can load.

Hopefully this will solve the problem!

I've got this today. Why did it happen?

SoftRaid app window doesn't shoe any problems.

@nwind

Its a bug.

If you can write a small file to the volume, it would be ours.

If the volume is "write protected", then it is a macOS kernel bug.

What happens is disks are assigned bsd numbers when connected. This is "permanent" until you restart, or disconnect the disks. There is a bug where the disk number can be changed, generally during sleep. But the IO registry is not notified of the change, so if SoftRAID wrote to that volume, big problem. Whenever the system wakes from sleep, SoftRAID first checks the BSD identifier. if the same, all is good with the world. if the numbers do not match, then the volume is write protected.

This has been reported to Apple, no fix yet.

The first bug, if you can still write to the volume, appears to be a bug in SoftRAID Monitor we are tracking down to fix. It is like a false alarm.

@softraid-support

I wasn’t able to write until I restarted computer 

@nwind

That is an example of the kernel bug. Before we figured this out, we would get a user a month with data corruption. That is why we built in the protection.

@softraid-support

Better this was than data loss. 
I hope Apple will fix it.