reassign 760976 udev forcemerge 754987 760976 thanks Hi
Am 09.09.2014 23:32, schrieb Steve Langasek: > On Tue, Sep 09, 2014 at 05:57:00PM +0200, Vincent Lefevre wrote: >> Package: sysvinit-core >> Version: 2.88dsf-53.4 >> Severity: normal > >> "waiting for /dev to be fully populated." takes about 7 seconds with >> udev 204-14, but 32 seconds with udev 208-8. According to > >> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=758480#24 > >> this doesn't occur with systemd-sysv, and according to Marco d'Itri, >> this is not a udev bug. > > Are you using openrc? The other bug report you cite is about a boot delay > when using openrc. If you're *not* using openrc, I don't think you can draw > any conclusions that *your* bug is not a udev bug just because that other > bug is not a udev bug. > > The message in question comes from /etc/init.d/udev, which is part of the > udev package; and it just calls 'udevadm settle', which is a udev-internal > interface. I don't think a bug here could be sysvinit's fault. The bug you are looking for is https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=754987 (therefore re-assigning and merging it) It is caused when running udev under sysvinit (well, !systemd) and you have an allow-hotplug interface in /e/n/i and /lib/udev/net.agent is triggered. You'll then run into the 30 second timeout that udev applies to processes spawned from udev rules. I haven't fully debugged yet, why this happens under sysvinit only. Running "do_everything > /dev/null 2> /dev/null &" in a subshell is sufficient to work around this problem, but I'd like to fully understand first, what's the root cause. I didn't have time yet to investigate this further, so help here is appreciated. Michael -- Why is it that all of the instruments seeking intelligent life in the universe are pointed away from Earth?
signature.asc
Description: OpenPGP digital signature
