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

2014-09-13 Thread David Herrmann
Hi On Sat, Sep 13, 2014 at 8:45 PM, Michael Biebl wrote: > 2014-09-13 20:34 GMT+02:00 Reindl Harald : >> >> Am 13.09.2014 um 20:20 schrieb Michael Biebl: >>> 2014-09-13 18:57 GMT+02:00 Marcel Holtmann : > break bluetooth completely by setting e.g. WatchdogSec=0.. setting WatchdogSec

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

2014-09-13 Thread Michael Biebl
2014-09-13 20:34 GMT+02:00 Reindl Harald : > > Am 13.09.2014 um 20:20 schrieb Michael Biebl: >> 2014-09-13 18:57 GMT+02:00 Marcel Holtmann : break bluetooth completely by setting e.g. WatchdogSec=0.. >>> >>> setting WatchdogSec=0 should disable that feature (as described in the >>> manpage).

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

2014-09-13 Thread Reindl Harald
Am 13.09.2014 um 20:20 schrieb Michael Biebl: > 2014-09-13 18:57 GMT+02:00 Marcel Holtmann : >>> 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 j

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

2014-09-13 Thread Michael Biebl
2014-09-13 18:57 GMT+02:00 Marcel Holtmann : >> 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 th

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

2014-09-13 Thread Marcel Holtmann
Hi Michael, >> So, lets look at the bluetoothd code: >> >>watchdog_usec = getenv("WATCHDOG_USEC"); >>seconds = atoi(watchdog_usec) / (1000 * 1000); >> >>watchdog = g_timeout_add_seconds_full(G_PRIORITY_HIGH, >>seconds / 2, >>

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

2014-09-13 Thread David Herrmann
Hi On Sat, Sep 13, 2014 at 2:52 PM, Michael Biebl wrote: > 2014-09-13 14:48 GMT+02:00 Michael Biebl : >> Will report this to the bluez upstream maintainer > > If only I could find the upstream bluez bug tracker... I have just sent a patch to linux-bluetooth ML, you're on CC. Thanks David __

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

2014-09-13 Thread Michael Biebl
2014-09-13 14:48 GMT+02:00 Michael Biebl : > Will report this to the bluez upstream maintainer If only I could find the upstream bluez bug tracker... -- Why is it that all of the instruments seeking intelligent life in the universe are pointed away from Earth? __

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

2014-09-13 Thread Michael Biebl
2014-09-13 13:10 GMT+02:00 David Herrmann : > So, lets look at the bluetoothd code: > > watchdog_usec = getenv("WATCHDOG_USEC"); > seconds = atoi(watchdog_usec) / (1000 * 1000); > > watchdog = g_timeout_add_seconds_full(G_PRIORITY_HIGH, >

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

2014-09-13 Thread David Herrmann
Hi On Sat, Sep 13, 2014 at 1:44 AM, Michael Biebl wrote: > Hi, > > we are currently trying to debug an upgrade failure from 208 to 215 > [1] in Debian related to the sd_notify/watchdog feature. > > This bug is not reliably reproducibly, we suspect a race somewhere > when systemd is re-exec'ed and

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

2014-09-12 Thread Michael Biebl
Hi, we are currently trying to debug an upgrade failure from 208 to 215 [1] in Debian related to the sd_notify/watchdog feature. This bug is not reliably reproducibly, we suspect a race somewhere when systemd is re-exec'ed and a daemon currently tries to talk to systemd via sd_notify. In my atte