M1 mini, Big Sur 11...
 
Notifications
Clear all

M1 mini, Big Sur 11.6.2, 2xThunderBay8, moved install, kernel panic

46 Posts
7 Users
2 Likes
1,639 Views
(@softraid-support)
Posts: 8008
Member Admin
 

@maxmartini 

Did you recreate the volume with 64k stripe unit size and see if that solves this issue for you?

 
Posted : 16/05/2022 11:39 am
(@mcbam)
Posts: 7
Active Member
Topic starter
 

What solved it at the time was to upg to 6.2b7 or b9. We're now running 6.2.1 driver and software, still latest of MacOS BigSur v11

we recently had another puzzling issue with the ThunderBays and the driver. I contemplated moving it to 64k stripe size, but we have over 60TB online and have no easy path to move and copy all that data.

 
Posted : 16/05/2022 12:12 pm
(@softraid-support)
Posts: 8008
Member Admin
 

@mcbam 

Please describe the other issue you are seeing. I may be able to assist.

 
Posted : 17/05/2022 12:14 am
(@mcbam)
Posts: 7
Active Member
Topic starter
 

Thanks but we worked wth Josh in support (among others) and found there may have been a hardware issue with one of the chassis', and there was definitely an issue with parity on one (or several) of the volume drives.

It started with a short brown out and even though these Thunderbays are attached to a UPS, they dropped from the system, and would not allow the host computer to reboot without timing out waiting for the Thunderbolt bus. After several uninstalls and re-installs of the SoftRAID software, a new chassis, and building a new and separate host system on which to test, we were able to get the drives back and to report that 1 of 3 volumes had parity issues.

The Thunderbays and/or SoftRAID simply would not allow the host computer to re-boot or shutdown because the driver and/or volumes would lock up the Thunderbolt bus. Each Thunderbay was on a separate bus, not chained.

All of this was detailed in Case: 01474575

One thing I'd like to understand better is how the driver is installed/uninstalled. When it was uninstalled, the system still had these remnants on the system, even after a restart:

/private/var/db/KernelExtensionManagement/AuxKC/CurrentAuxKC/StashedExtensions/D358FCB6-0214-467F-940F-A0A07966D591/SoftRAID.kext/Contents/MacOS/SoftRAID

/System/Volumes/Data/private/var/db/KernelExtensionManagement/AuxKC/CurrentAuxKC/StashedExtensions/D358FCB6-0214-467F-940F-A0A07966D591/SoftRAID.kext/Contents/MacOS/SoftRAID

/System/Volumes/Data/System/Library/Templates/Data/Library/Extensions/SoftRAID.kext/Contents/MacOS/SoftRAID

/System/Volumes/Data/Library/StagedExtensions/Library/Extensions/SoftRAID.kext/Contents/MacOS/SoftRAID

/System/Library/Templates/Data/Library/Extensions/SoftRAID.kext/Contents/MacOS/SoftRAID

/Library/StagedExtensions/Library/Extensions/SoftRAID.kext/Contents/MacOS/SoftRAID

When I saw all the above, we were trying to determine whether some out of version element of the driver was preventing the SoftRAID volumes from appearing.

--

Also, even now that it seems to work as expected, we still see this in the log, implying there still is an error somewhere:

May 12 15:05:25 - softraidd: Manually loading SoftRAID driver; result = 71 (0 = no error).

No one in support was able to tell us what a result= 71 means.

--

 

 
Posted : 17/05/2022 9:39 am
(@softraid-support)
Posts: 8008
Member Admin
 

@mcbam 

/Library/StagedExtensions/Library/Extensions/SoftRAID.kext/Contents/MacOS/SoftRAID

 

This is where macOS loads drivers from now. Not the /Extensions folder. If you do an "uninstall SoftRAID" and restart and still see this here, it means macOS did not flush the extensions cache correctly.

the other files are various cache files that are not important.

The result=71 code means the driver was already loaded.

 
Posted : 17/05/2022 11:45 am
(@mcbam)
Posts: 7
Active Member
Topic starter
 

This is where macOS loads drivers from now. Not the /Extensions folder. If you do an "uninstall SoftRAID" and restart and still see this here, it means macOS did not flush the extensions cache correctly.

Is there a way to manually flush the extensions cache? Or make sure the detritus is removed?

 

 
Posted : 17/05/2022 12:03 pm
(@softraid-support)
Posts: 8008
Member Admin
 

@mcbam If you use uninstall SoftRAID - all components, that will get rid of nearly everything.

 
Posted : 17/05/2022 1:56 pm
(@mcbam)
Posts: 7
Active Member
Topic starter
 

I did use that, several times, rebooting in between. Josh can confirm (though he was just on the phone), it did not remove all the components. The list above were all still on the system after the SoftRAID uninstall and reboot several times.

 

 
Posted : 17/05/2022 3:17 pm
(@softraid-support)
Posts: 8008
Member Admin
 

@mcbam 

If you do "Uninstall SoftRAID-All components", it will remove the driver and all key components.

Things left in the various macOS cache's don't really matter.

If the driver was still in the /StagedExtensions folder, after uninstalling, then that is a macOS bug. macOS is supposed to clear that. Developers are specifically not supposed to touch those parts of the OS.

 
Posted : 17/05/2022 4:56 pm
(@crubel)
Posts: 2
New Member
 

I am now running Monterey 12.4, updated it last night and this morning I powered up while leaving my Thunderbay 4 alone.  Normally, I have been removing the last drive at the suggestion of Softraid so that my M1 iMac would boot up OK with the Thunderbay connected and not panic into a reboot cycle.  So far today everything is working normally on the 6.2.1 standard release driver setup without having to pull that 4th drive out during bootup.

So maybe Apple has finally gotten around to fixing whatever the issue was....   I just wanted to pass this along ASAP since I am sure a lot of Thunderbay users were waiting for this to get fixed like me...

 

Curtis

 
Posted : 04/06/2022 2:22 pm
(@softraid-support)
Posts: 8008
Member Admin
 

@crubel 

We have to test this, as we have not been informed of any fix in macOS yet, it would be great news is true, but I would expect we would have been told a fix was incorporated in 12.4, but maybe not.

 
Posted : 04/06/2022 6:03 pm
(@crubel)
Posts: 2
New Member
 

@softraid-support 

yes of course, please test it.  I was not able to change my stripe unit size by changing the type from desktop to photo or any of the other settings for some reason so I’ve been removing my 4th disk and I’m afraid to wear out the connectors eventually. 

Today was the first time it’s booted without a panic since moving to this m1 iMac.  Hopefully it’s really fixed and not just a fluke…

Please keep me in the loop on the fix please…

Thank you for your reply,

Curtis

 
Posted : 04/06/2022 8:00 pm
(@kurt765)
Posts: 3
Active Member
 

With my Mac Mini M1 and Thunderbay with 6.2, I don't get constant kernel panic crashes. If I install 6.2.1 then it's kernel panic town. The 6.2.1 beta version seemed to avoid the issue. I see the workaround is to use a different stripe size with 6.2.1  (involving a huge hassle of re-copying 40 terabytes of data). And there's no fix yet because it's a MacOS issue and we're waiting on Apple? I just want to make sure I understand the state of where we are at right now.

 
Posted : 30/06/2022 4:50 pm
(@softraid-support)
Posts: 8008
Member Admin
 

@kurt765 

There should be no difference i behavior with 6.2 and 6.2.1.

 

For many users, connecting 5 minutes after startup avoids the issue.

(or removing a drive for 5 minutes)

 

 
Posted : 30/06/2022 6:04 pm
(@kurt765)
Posts: 3
Active Member
 

@softraid-support Well, in my experience 6.2 doesn't crash randomly and 6.2.1 does, regardless of how that works for other people. Monterey 12.3.1

 
Posted : 30/06/2022 6:40 pm
Page 3 / 4
Share:
close
open