=== harlowja is now known as harlowja_away
=== harlowja_away is now known as harlowja
[04:38] <harlowja> smoser yt, are u ok with https://code.launchpad.net/~vlastimil-holer/cloud-init/net-reconfigure/+merge/186352 ?
[04:39] <smoser> looking
[04:39] <smoser> i shouldn't be here.
[04:39] <harlowja> lol
[04:39] <harlowja> np
[04:39] <smoser> what do i get wit this ?
[04:39]  * harlowja not sure whose been testing snapshots at y!
[04:39] <harlowja> lol
[04:39] <harlowja> but the ips haven't been right, lol
[04:39] <harlowja> so smoser  i think what u'll get is the ifdown part before networking modification
[04:40] <harlowja> another idea, i think http://bazaar.launchpad.net/~cloud-init-dev/cloud-init/trunk/view/head:/sysvinit/redhat/cloud-init-local#L30 would probably like to have Should-Stop: $network
[04:40] <harlowja> although not sure
[04:42] <harlowja> i think the issue we have just noticed (don't ask why nobody noticed or told me before) is that snapshot saves image, with old netconfig, starts up off of snapshot, but init-local happens after networking starts (so config drive network writing is to late at this point)
[04:42] <harlowja> anyways
[04:43] <harlowja> let me know what u think when u get some time
[04:51] <harlowja> the /etc/rc.d/rc5.d/ that this is part of http://paste.openstack.org/show/73103/
[04:56] <smoser> harlowja, yeah, i knew it was broken in that way.
[04:56] <smoser> well, i knew it was broken for inserting statick eth0
[04:56] <smoser> i hadn't thought about other nics that were previosly configured.
[04:59] <smoser> it seems a bit broken to bring dwn configured network interfaces though.
[04:59] <harlowja> agreed
[05:01] <harlowja> what to do i wonder :-P
[05:04] <harlowja> b b
=== harlowja is now known as harlowja_away
=== harlowja_away is now known as harlowja
[17:52] <harmw> and another day goes by without new cirros release :p
[18:09] <kwadronaut> be nice ;-)
[18:11] <harmw> I'm always nice :>
[18:12] <kwadronaut> (-:
[18:31] <smoser> harmw, http://paste.ubuntu.com/7075003/
[18:31] <smoser> can you test that ?
[18:31] <smoser> adds and uses 'asroot' and adds a container output.
[18:32] <harmw> ahyea, lets see
[18:33] <smoser> then i'll pull your patch if that works.
[18:33] <smoser> and then we'll be closer 
[18:40] <harmw> awk: /proc/1/environ: Permission denied
[18:40] <harmw> I'm running without sudo now, manually invoking cirros-status
[18:40] <smoser> bah. right.
[18:40] <harmw> it does print ssh keys nicely now
[18:45] <smoser> k. i shoudl have just used 'lxc-is-container', but i was hoping to avoid the fork :)
[18:47] <smoser> http://paste.ubuntu.com/7075099/
[18:58] <harlowja> smoser u around
[18:58] <smoser> here
[18:59] <harlowja> so about that stupid ifconfig crap
[18:59] <harlowja> lol
[18:59] <smoser> k
[18:59] <harlowja> any thoughts on a good longer term solution, not that hard to adjust the sysvinit file
[19:02] <harlowja> or other way is to say users should delete there ifconfig before snapshotting as part of cleanup
[19:04] <harlowja> bb
[19:32] <smoser> harmw, i really dont want to tell pepole they have to clean up
[19:32] <smoser> i'd love to have a way around that.
[19:32] <smoser> err... harlowja 
[19:32] <harlowja> ya, smoser, hmmm
[20:21] <harmw> smoser: iamroot() doesn't work
[20:21] <harmw> iamroot() { return $(id -u)
[20:21] <harmw> }
[20:22] <harmw> anything wrong with doing it that way?
[20:25] <smoser> harmw, bah.
[20:25] <smoser> add 
[20:25] <smoser>  [ "$UID" = "0" ]
[20:25] <smoser> after it sets uid.
[20:25] <smoser> ie
[20:25] <smoser> iamroot() {
[20:25] <smoser>     [ -z "$UID" ] || UID=$(id -u)
[20:25] <smoser>     [ "$UID" = "0" ]
[20:25] <smoser> }
[20:25] <smoser> sorry. 
[20:25] <smoser> yeah, the way i had it it would always say it was root
[20:27] <harmw> uhm, how's this any different :P
[20:28] <harmw> ie. still doesn't work
[20:29] <harmw> it's now always using sudo
[23:54] <smoser> bah. harmw i'ms orry.