Hey, Danny Milosavljevic <dan...@friendly-machines.com> writes:
> If you want to debug that live, you can log to the console: switch to the > console (Ctrl-Alt-F1), set the kernel console loglevel to maximum (alt > print 9), and then watch what it says there. > > I currently use the following shell script to halt without poweroff > (because if it powered off you wouldn't be able to read the screen anymore): > > $ cat debug-halt > #!/bin/sh > set -e > echo "file drivers/scsi/sd.c +p" > /sys/kernel/debug/dynamic_debug/control > echo "file drivers/ata/libata-core.c +p" > > /sys/kernel/debug/dynamic_debug/control > echo "file drivers/nvme/host/core.c +p" > > /sys/kernel/debug/dynamic_debug/control > echo "file drivers/nvme/host/pci.c +p" > > /sys/kernel/debug/dynamic_debug/control > exec herd eval root '((@ (shepherd system) halt))' Did that allow you to get more debugging info? > Because, otherwise, if one invoked "sudo halt", it would call > '((@ (shepherd system) power-off))' (see modules/shepherd/scripts/halt.scm > )--which > is not what we want in order to find the cause of this problem. > (That's maybe not so great either. Why don't we have a poweroff script, > and a (different) halt script in sbin ?) I guess we could do that. Thanks, Ludo’.
