> Please test again with the current version (6.12.33-1~bpo12+1). I think > this may have been fixed upstream by: > > commit 3c44d31cb34ce4eb8311a2e73634d57702948230 > Author: Herbert Xu <[email protected]> > Date: Sat Aug 17 14:58:35 2024 +0800
I did test with 6.12.35+deb12-armmp, the bug is still there, even a bit nastier. Kind of assumed that, as the commit you mentioned went into 6.1.113 and my last known working version is 6.6.13+bpo-armmp. With 6.12.35+deb12-armmp, the kernel does not panic immediately during a boot attempt, but errors begin to appear soon. During my testing, I got many different messages mainly around the SATA subsystem (resetting the controller, ...). Some of them even indicated possible data corruption. I've got more errors (and they appeared earlier) when two SATA drives were attached, compared to having only a single drive. However, with either configuration, the system would hang sooner or later. With 6.6.13+bpo-armmp, everything is running fine and stable with the same hardware. I did save some logfiles, so I can send them if you wish. However, there is a ton of it with only small portions being relevant, so I would need some time to process them. Phil
