Bug#520607: hwclock.sh should require $local_fs for start

2009-08-06 Thread Petter Reinholdtsen
retitle 520607 hwclock.sh should require checkroot for start tags 520607 + patch thanks Here is a more correct and tested patch to change the dependency of hwclock.sh to run just after checkroot.sh, and not in parallel with hwclockfirst.sh when concurrent booting is enabled. Here is a proposed ch

Bug#520607: hwclock.sh should require $local_fs for start

2009-08-03 Thread Petter Reinholdtsen
I just ran into a complete system hang during boot with my test machine, and believe it is caused by this bug. This was with full parallel booting (CONCURRENTY=makefile in /etc/default/rcS), and dependency based boot sequencing using Debian/testing. The system got stuck just after setting the sys

Bug#520607: hwclock.sh should Require $local_fs for start

2009-07-25 Thread Sven Joachim
On 2009-07-25 15:20 +0200, Petter Reinholdtsen wrote: > retitle 520607 hwclock.sh should require $local_fs for start > tag 520607 - patch > thanks > > My proposed patch were not correct, as the hwclock script should not > wait until mountnfs before running. Glad to see that you agree with me. :-)

Bug#520607: hwclock.sh should Require $local_fs for start

2009-07-25 Thread Petter Reinholdtsen
retitle 520607 hwclock.sh should require $local_fs for start tag 520607 - patch thanks My proposed patch were not correct, as the hwclock script should not wait until mountnfs before running. To avoid them starting at the same time, one could let hwclockfirst have 'X-Start-Before: hwclock' to mak

Bug#520607: hwclock.sh should Require $local_fs for start

2009-07-15 Thread Petter Reinholdtsen
[Sven Joachim] > Is this a supported setup? It seems the tzdata postinst script will > turn /etc/localtime into a real file. This has been the case since > glibc version 2.3.6-6, see #346342¹. Right. Perhaps this is the old and obsolete requirements. > I'm afraid that running hwclock.sh so lat

Bug#520607: hwclock.sh should Require $local_fs for start

2009-07-15 Thread Sven Joachim
On 2009-07-15 11:26 +0200, Petter Reinholdtsen wrote: > retitle 520607 hwclock.sh should Require mountnfs for start > tag 520607 + patch > thanks > > Acutally, hwclock.sh should rather require $remote_fs, because it > exist to set the clock after /usr/ is available when /etc/localtime is > a symli

Bug#520607: hwclock.sh should Require $local_fs for start

2009-07-15 Thread Petter Reinholdtsen
retitle 520607 hwclock.sh should Require mountnfs for start tag 520607 + patch thanks Acutally, hwclock.sh should rather require $remote_fs, because it exist to set the clock after /usr/ is available when /etc/localtime is a symlink to /usr/share/zoneinfo/. /usr/ is guaranteed to be available afte

Bug#520607: hwclock.sh should Require $local_fs for start

2009-07-13 Thread Sven Joachim
severity 520607 important user initscripts-ng-de...@lists.alioth.debian.org usertags 520607 incorrect-dependency thanks Am 21.03.2009 um 10:40 schrieb Jörg Sommer: > Package: util-linux > Version: 2.13.1.1-1 > Severity: normal > File: /etc/init.d/hwclock.sh > > Hi, > > the Required-Start conditio

Bug#520607: hwclock.sh should Require $local_fs for start

2009-03-21 Thread Jörg Sommer
Package: util-linux Version: 2.13.1.1-1 Severity: normal File: /etc/init.d/hwclock.sh Hi, the Required-Start conditions of hwclock.sh and hwclockfirst.sh are the same, but hwclock.sh should require the $local_fs. Otherwise insserv would place both scripts at the same point in the list which doesn