I have to plug in a keyboard and manually boot to 25.10.4 if I reboot the system.
Tried biding my time during 25.10.5 hoping it was a fluke, but 25.10.6 is suffering the same fate.
Boot output:
Loading Linux 6.12.99-production+truenas …
Loading initial ramdisk …
error: checksum verification failed.
{ … a bunch of stuff scrolls by … }
Call Trace:
panic+0x344/0x3a0
mount_root_generic+0x1ac/0x350
prepare_namespace+0x64/0x280
kernel_init_freeable+0x413/0x460
? __pfx_kernel_init+0x10/0x10
kernel_init+0x1a/0x1c0
ret_from_fork+0x34/0x50
? __pfx_kernel_init+0x10/0x10
ret_from_fork_asm+0x1a/0x30
Kernel Offset: 0x10c00000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
—[ end Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0) ]—
Differences when selecting 25.10.5:
Loading Linux 6.12.95-production+truenas …
{ … everything else same until … }
Kernel Offset: 0x37e00000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
I can’t find anything wrong when I boot 25.10.4.
Motherboard: MSI MS-7721/A78M-E45 (MS-7721), BIOS V25.3 /03/18/2014
CPU (APU): A10-5700
RAM: 16GB
6 x 2TB SSD drives. 1 boot (I know 2TB is overkill), 5 ZFS raidz1
I’m not going to say anything about running several memtest passes to rule out bad RAM because that would be a waste of time.
Since we know it isn’t the RAM, how did you do the upgrade?
EDIT: I was being sarcastic about not running memtests.
When you have the chance to plug in a keyboard and monitor to the NAS, please run several memtest passes to have the peace of mind that you don’t have bad or failing RAM. It doesn’t take long. If your RAM passes several tests, then we can rule it out. 
Via the web interface (/ui/system/update). Clicked install update. Saved the current config just in case. Let it do its thing and reboot. Did it that way for both the 25.10.5 and 25.10.6 upgrades from 25.10.4.
Found a similar problem with someone’s 25.10.3 upgrade and theirs was a degraded boot-pool. I just did a zpool list on mine and found the same problem. Going to try cleaning that up and reinstalling the upgrade. Crossing fingers.
Comment that helped me: Kernel panic after update to 25.10.3 - #5 by breikers (though further comments suggest other things at fault)
Your boot-pool is degraded? What is the model? You said 2 TB SSD, but what brand? Can you post the output of zpool status -v?
If your SSD is failing, I wouldn’t reinstall TrueNAS on it.
I was being sarcastic about not running memtests.
When you have the chance to plug in a keyboard and monitor to the NAS, please run several memtest passes to have the peace of mind that you don’t have bad or failing RAM. It doesn’t take long. If your RAM passes several tests, then we can rule it out.
/s would help point out the sarcastic bits.
Did the sudo zpool clean boot-pool and then deleted a bunch of old boot environments through the UI as well as 25.10.5 and 26.10.6. Redid the upgrade to 25.10.6 from 25.10.4 and it booted up with out the checksum error or kernel panic.
I’ll do the memtest runs shortly.
Probably need to replace that cheap PNY SSD (bought when I was just toying with the idea and didn’t want to spend much on it). Already had several fail and get replaced in the main pool by a better brand.
Good thing AI data centers aren’t driving prices up to something unreasonable for a hobbyist. /s
That was my fault. Let’s just say I had recent PTSD. 
I’ll run a memtest on my brain now. /s
If failing hardware (SSD, RAM, CPU) caused the issue in the first place, you could be setting yourself up for future problems.
It sucks. 
It does. And my financial situation doesn’t support buying new equipment. Used and refurb worked fine until everyone decided drives had to be removed and/or holes drilled in them before resale.
Glad you could solve the problem. I never had any issues after this. All upgrades went just fine until now…
