Re: Shutdown -r under -current hangs on RPi3

On Oct 1, 2023, at 09:25, bob prohaska <fbsd@www.zefox.net> wrote:

> On Sun, Sep 24, 2023 at 08:17:54AM -0700, bob prohaska wrote:
>> 
>> It appears that setting 
>> debug.bootverbose=1 
>> suppresses the otherwise-persistent
>> Khelp module "ertt" can't unload until its refcount drops from 1 to 0.
>> message during shutdown. No new output either, apart from the swap
>> removal notice.
>> 
>> So far, the few shutdown -r experiments tried have not gotten stuck,
>> even after an OS build/install cycle.
> 
> Another shutdown hang has been observed with -current on a Pi3.
> This time bootverbose was on, as was ucom debug. This machine
> has an FTDI usb-serial adaper. 
> 
> The serial console reported:
> 
> *** FINAL System shutdown message from bob@pelorus.zefox.org ***             
> 
> System going down IMMEDIATELY                                                  
> 
> 
> Oct  1 08:21:26 pelorus shutdown[78571]: reboot by bob: 
> Stopping sshd.
> Waiting for PIDS: 1063.
> Stopping cron.
> Waiting for PIDS: 1073.
> Stopping powerd.
> Waiting for PIDS: 1002.
> Stopping devd.
> Waiting for PIDS: 752.
> Writing entropy file: .
> Writing early boot entropy file: .
> .
> Terminated
> ucom_inwakeup: tp=0xffffa00021cc1c00
> ucom_ioctl: cmd = 0x2000740e
> ucom_inwakeup: tp=0xffffa00021cc1c00
> ucom_inwakeup: tp=0xffffa00021cc1c00
> ucom_close: tp=0xffffa00021cc1c00
> ucom_shutdown: 
> ucom_dtr: onoff = 0
> ucom_line_state: on=0x00, off=0x01
> ucom_rts: onoff = 1
> ucom_line_state: on=0x02, off=0x00
> Oct  1 08:21:32 pelorus ucom_cfg_close: 
> syslogd: exiting on signal 15
> Waiting (max 60 seconds) for system process `vnlru' to stop... done
> Waiting (max 60 seconds) for system process `syncer' to stop... 
> Syncing disks, vnodes remaining... 0 0 0 done
> All buffers synced.
> Swap device da0s2b removed.
> Uptime: 3d0h14m47s
> Resetting system ... 
> 
> Both Ethernet LEDs turned off, if that's significant.
> After about half an hour, I pulled the plug. The machine
> came up hands-off.
> 

Was this a debug world/kernel? non-debug? May be
output I was expecting only is generated by the
debug variants?

Was there any lights sometimes on anywhere during that
about half hour? Might it have powered off instead of
rebooting?

Is there anything about the above message sequence
(before "Resetting system ...") that is different
from when the reboot works, also with "bootverbose was
on, as was ucom debug"?

It is an example with no "ertt" message.


===
Mark Millard
marklmi at yahoo.com