Issue (on some systems) with kernels (RL 9.8) starting at kernel-5.14.0-687.36.1.el9_8.x86_64

So, I have a few RL9.8 boxes here, and I now have a similar issue on 2 of them starting with kernel

kernel-5.14.0-687.36.1.el9_8.x86_64

Both are having problems at boot time, ostensibly after filesystem decryption is done on boot. I’m about to launch into it, and I’m going to have to write a log file from the machine I can work with easily in a state where the boot is just starting. It may be that the decryption failed to work on both of these, but the wierd bit here is I have another system (my main system) that is having no problem booting at all, with that kernel or the next one.

I can boot at least one of the systems into the previous kernel

kernel-5.14.0-687.33.1.el9_8.x86_64

and it boots fine, but it is now at least two kernels off the leading edge. I attempted to save the file to whatever the root filesystem was in that degraded state, and not surprisingly, I can’t find it at all. The system was saying that I should save a log file on a usb drive, so I’m about to do that, but I’m wondering if anyone else is having similar fun. With one machine (which had a motherboard old enough it will not be able to make it to RL 10) I figured it could be the aging hardware, either with outright failure, or with no longer playing well with something in the newest kernels since it is not going to be able to turned into RL10. When the 2nd machine did the same thing (and it’s actually my newest motherboard, but one that does seem to have more compatibility issues than any of my other MB’s) I’m figuring there is something in common, and soon my good working system may join the “fail to boot” crowd.

I’m going to have to figure out how to write to a USB drive with a system that doesn’t even have a “mount” command that works, but does have a shell, and does have a working vim. I would love to hear that there is some fix I need to do that I can avoid all that and jump to the solution…

Much thanks, whether this is successful or not.
Mike

PS, it may be that the issue is at the filesystem/device decryption stage, as I see only /dev/sda and /dev/sdb, and this non-functional machine has at least 6 disks when it is fully booted… But I don’t have a clue as to what is doing this, or even if the symptom is identical with the other machine. But both booted fine with earlier kernels, and borked on this kernel…

Mike

And, this turned out to be a classic correlation does not imply causation. The new machine’s cooler died, and it would run just long enough to look like it was starting to boot before the thermal stops took hold. Waiting on Corsair for a warranty replacement (hopefully). The lack of others finding the same problem was a definite clue, so now I have the old machine who I was more inclined to think had hardware issues.

Cheers all,

Mike