1 |
Daniel Frey <djqfrey <at> gmail.com> writes: |
2 |
|
3 |
|
4 |
> It appears to be udev. Somewhere along in its stupid detection it |
5 |
> decides to process USB devices before sata ports, thusly randomly |
6 |
> renaming the boot drive to something else in the process. |
7 |
|
8 |
> It took me forever to figure this out, I eventually had a lightbulb |
9 |
> moment and used my phone to record video of it booting, then slowing it |
10 |
> down, as when the kernel panics you can't scroll back up to see WTF |
11 |
> happened. |
12 |
|
13 |
Kernel crash dumps might help [1] |
14 |
|
15 |
|
16 |
|
17 |
> This is an older machine, but I'm not convinced it's the motherboard |
18 |
> doing this. I've checked the boot order in the BIOS. I've also tried |
19 |
> setting and unsetting "BIOS order determines boot disk" in the kernel |
20 |
> config and it made no difference. |
21 |
|
22 |
You might want to 'emerge -1 sys-apps/hwids' |
23 |
|
24 |
|
25 |
> What eventually fixed it was building USB as modules. (Another kludge!) |
26 |
|
27 |
There are numerous 'usb sniffers' that capture data. Some clue |
28 |
might be found using a usb sniffer. |
29 |
|
30 |
|
31 |
|
32 |
> I have no custom udev rules, the only rules I could find were in |
33 |
> /lib/udev/rules.d: |
34 |
|
35 |
I use sys-fs/eudev. ymmv. |
36 |
|
37 |
|
38 |
> Does anyone have any explanation for this daft behaviour or know where I |
39 |
> should look? |
40 |
> I have multiple machines and it's only this one that has this problem, |
41 |
> which happened after a <at> world update long ago. |
42 |
|
43 |
If you have a similar setup on similar hardware, then 'diff' the (dmesg) |
44 |
boot log files for any differences and analyze. |
45 |
|
46 |
|
47 |
hth, |
48 |
James |
49 |
|
50 |
[1] https://wiki.gentoo.org/wiki/Kernel_Crash_Dumps |