Martin Husemann a écrit :
> On Wed, Nov 27, 2024 at 07:43:00AM +0100, BERTRAND Joël wrote:
>> Nov 27 00:48:55 legendre /netbsd: [ 135475.0035570] panic: cpu0:
>> softints stuck for 16 seconds
>
> We need to find out what is stuck there. Easiest way is to get crashdumps
> working, alternative is to enable ddb and do live debugging once the
> kernel crashed.
>
> Is the kernel you are using in /netbsd ?
Nope. In /netbsd-current. /netbsd contains a -10.0 kernel.
> Do ksyms work? nm /dev/ksyms should show you lots of kernel symbols.
It runs as expected.
> You could try the -f option to savecore.
I have added -f option.
> Is there something special about your swap/dump setup?
legendre# cat /etc/fstab | grep dp
/dev/raid0b none swap sw,dp 0 0
legendre# swapctl -l
Device 1K-blocks Used Avail Capacity Priority
/dev/raid0b 16777216 0 16777216 0% 0
legendre# cat /proc/meminfo
total: used: free: shared: buffers: cached:
Mem: 16502280192 16478519296 23760896 0 8975888384 11559067648
Swap: 17179865088 0 17179865088
MemTotal: 16115508 kB
MemFree: 23204 kB
MemShared: 0 kB
Buffers: 23204 kB
Cached: 11288152 kB
SwapTotal: 16777212 kB
SwapFree: 16777212 kB
In /var/crash, I only have bounds file (that contains index of the next
core dump, in my case : 6).
JB
Attachment:
signature.asc
Description: OpenPGP digital signature