On Wed, Jun 22, 2011 at 6:47 PM, Thomas Bächler <[email protected]> wrote: > Am 22.06.2011 18:17, schrieb Dan McGee: >> On Tue, Jun 21, 2011 at 11:45 PM, Dave Reisner <[email protected]> wrote: >>> On Wed, Jun 22, 2011 at 01:27:54AM -0300, Gerardo Exequiel Pozzi wrote: >>>> By default, there is no network is configured in rc.conf. So if you boot >>>> the machine with an unmodified rc.conf you will see a warning >>>> about network deprecation settings. >>>> >>>> Signed-off-by: Gerardo Exequiel Pozzi <[email protected]> >>>> --- >>>> rc.conf | 2 +- >>>> 1 files changed, 1 insertions(+), 1 deletions(-) >>>> >>>> diff --git a/rc.conf b/rc.conf >>>> index 5984171..bd13fce 100644 >>>> --- a/rc.conf >>>> +++ b/rc.conf >>>> @@ -114,4 +114,4 @@ NETWORK_PERSIST="no" >>>> # If something other takes care of your hardware clock (ntpd, >>>> dual-boot...) >>>> # you should disable 'hwclock' here. >>>> # >>>> -DAEMONS=(hwclock syslog-ng network netfs crond) >>>> +DAEMONS=(hwclock syslog-ng !network !netfs crond) >>>> -- >>>> 1.7.5.2 >>>> >>> >>> I noticed this the other day on one of my VMs. I'm not sure what the >>> best course of action is. There's a few choices... >>> >>> 1) This. >>> 2) Refactor the need_legacy check. It's fairly naive at the moment. We >>> could easily cover the 'null' edge case by checking: >>> >>> [[ -z $INTERFACES && -z $interface ]] >>> >>> 3) Change where deprecation messages are printed -- i.e., only once >>> we're about to execute some net-tools based logic do we print a message. >>> This is perhaps the least desireable, imo, because it's also prone to >>> creating the most amount of spammage. I wouldn't want to see the warning >>> printed more than once per invocation. >> >> I'd be wary of disabling network by default. I agree with Dave's #2 >> suggestion though. > > ACK. > >
Yup, number two makes most sense to me too. -t
