[systemd-devel] watchdog feature highly ineffecient / using too much CPU

[Michael Biebl]

2014-09-13 18:57 GMT+02:00 Marcel Holtmann <marcel at holtmann.org>:
>> break bluetooth completely by setting e.g. WatchdogSec=0..
>
> setting WatchdogSec=0 should disable that feature (as described in the manpage). Which means that systemd should

Na, this was just a typo.
I intended to write WatchdogSec=0.5

Doing that, bluetooth.service ends up in failed state here:

# systemctl status bluetooth.service
● bluetooth.service - Bluetooth service
   Loaded: loaded (/lib/systemd/system/bluetooth.service; enabled)
   Active: failed (Result: start-limit) since Sa 2014-09-13 13:53:09
CEST; 6h ago
     Docs: man:bluetoothd(8)
  Process: 4684 ExecStart=/usr/lib/bluetooth/bluetoothd (code=exited,
status=0/SUCCESS)
 Main PID: 4684 (code=exited, status=0/SUCCESS)
   Status: "Quitting"

Sep 13 13:53:09 pluto systemd[1]: Unit bluetooth.service entered failed state.
Sep 13 13:53:09 pluto systemd[1]: bluetooth.service start request
repeated too quickly, refusing to start.
Sep 13 13:53:09 pluto systemd[1]: Failed to start Bluetooth service.
Sep 13 13:53:09 pluto systemd[1]: Unit bluetooth.service entered failed state.
Sep 13 13:53:11 pluto systemd[1]: bluetooth.service start request
repeated too quickly, refusing to start.
Sep 13 13:53:11 pluto systemd[1]: Failed to start Bluetooth service.


Fwiw, I wasn't trying to imply that using something like WatchdogSec=1
is necessarily a reasonable thing to do.
But what would be a reasonable value?
Can we improve the documentation here, e.g. the system.service or
sd_notify man page.

Michael
-- 
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?