[00:00] <stupidnic> That's about right
[00:00] <smoser> the bug report is a bit wrong. in the description.
[00:00] <smoser> i thought it was all user-space
[00:00] <smoser> but really the user space says "hey kernel, do magic!"
[00:03] <stupidnic> and of course... now cinder decides it is going to act up
[00:09] <stupidnic> smoser: yeah, it's clearly dependent on the size of the backing volume upped the size to 1TB and it's doing the same thing. At least this image has the logging enabled so I can see exactly what is taking so long
[00:10] <smoser> yeah, and the log should even explicitly say "took X seconds" for that somewhere.
[00:10] <smoser> one interesting thing... i made that backgroundable.
[00:10] <stupidnic> Oh?
[00:10] <smoser> but it turns out it doesn't matter. its so IO heavy, that nothing else is happing. 
[00:10] <stupidnic> hahah
[00:11] <stupidnic> That's the way it goes I guess
[00:11] <smoser> http://bazaar.launchpad.net/~cloud-init-dev/cloud-init/trunk/view/head:/doc/examples/cloud-config.txt
[00:11] <smoser> you can give it a try:
[00:11] <stupidnic> the SAN might have the IOPS to handle that though
[00:11] <smoser> resize_rootfs: noblock
[00:11] <smoser> you can put that in user-data or in a /etc/cloud/cloud.cfg.d/foo.cfg file
[00:12] <stupidnic> seems like cloud-config would be the more lower impact way to test it
[00:14] <stupidnic> I have to run, but I should be back in about an hour to test it out and see how it does.
[00:14] <stupidnic> smoser: thanks for the help and pointers
[00:15] <stupidnic> I'll let you know how I make out with noblock
[00:19] <stupidnic> smoser: 2015-03-18 20:16:33,573 - util.py[DEBUG]: Resizing took 482.659 seconds
=== harlowja_ is now known as harlowja_away
=== shardy_z is now known as shardy
[16:57] <SpamapS> if I want to use configdrive on OpenStack, does cloud-init automatically try to mount and read it?
[16:57] <SpamapS> smoser: ^ ?
[16:57] <SpamapS> harlowja_away: ^
=== harlowja_away is now known as harlowja_
[17:10] <harlowja_> SpamapS yes
[17:11] <SpamapS> harlowja_: ok, mordred is suggesting that is not true.
[17:11] <SpamapS> harlowja_: our Ubuntu trusty nodes don't seem to do it anyway.
[17:11] <harlowja_> so then it starts to depend on what datasources are enabled
[17:12] <harlowja_> $ cat /etc/cloud/cloud.cfg
[17:12] <harlowja_> ie datasource_list:
[17:12] <harlowja_>  - ConfigDrive
[17:12] <harlowja_>  - Ec2
[17:12] <harlowja_>  - None
[17:13] <harlowja_> if configdrive not there, then its not gonna do anything :-P
[17:32] <harlowja_> SpamapS go give a monty a hug, or something, lol, u both at HP, go find him or something (not sure which office he is at, lol)
[17:34] <SpamapS> HAHA
[17:34] <SpamapS> He's at his home in NY
[17:34] <SpamapS> I'm at my home in LA
[17:34] <SpamapS> so.. yeah.
[17:36] <harlowja_> start walking then
[17:36] <harlowja_> lol
[17:36] <harlowja_> chop chop
[18:25] <suro-patz> smoser: Would appreciate, if you can get some time and provide some feedback on https://code.launchpad.net/~suro-patz/cloud-init/vm-clone-ip-reusage-issue/+merge/252961
[20:59] <smoser> SpamapS, i'm 99% certain it does. 
[21:00] <smoser> just the other day suggsted to someone to use it and it solved their problem  
[21:00] <smoser> if it doesnt, plase open bug
[22:59] <SpamapS> smoser: thanks, we figured it out. What mordred was on about was that it doesn't _leave_ it mounted.
[22:59] <SpamapS> smoser: also OpenStack infra can't use it because it doesn't work in precise.
[23:04]  * harlowja_ my thought on that is that its 0.6.3 and openstack isn't writing the old v1 config-drive format anymore?