[17:59] <smoser> harmw, hey. just thought i'd share : http://paste.ubuntu.com/7227342/
[17:59] <smoser> thtas just a "test cirros"
[19:05] <smoser> harlowja, 
[19:05] <smoser> 2014-04-09 19:04:16,975 - url_helper.py[DEBUG]: Calling 'http://169.254.169.254/openstack/2012-08-10/meta_data.json' failed [10/-1s]: request error [(<urllib3.connectionpool.HTTPConnectionPool object at 0x7f1339863a10>, 'Connection to 169.254.169.254 timed out. (connect timeout=10.0)')]
[19:05] <smoser> 2014-04-09 19:04:16,975 - url_helper.py[DEBUG]: [0/1] open 'http://169.254.169.254/openstack/2013-04-04/meta_data.json' with {'url': 'http://169.254.169.254/openstack/2013-04-04/meta_data.json', 'headers': {'User-Agent': 'Cloud-Init/0.7.5'}, 'allow_redirects': True, 'method': 'GET', 'timeout': 10.0} configuration
[19:05] <smoser> 2014-04-09 19:04:26,986 - url_helper.py[DEBUG]: Calling 'http://169.254.169.254/openstack/2013-04-04/meta_data.json' failed [20/-1s]: request error [(<urllib3.connectionpool.HTTPConnectionPool object at 0x7f1339863c50>, 'Connection to 169.254.169.254 timed out. (connect timeout=10.0)')]
[19:05] <harlowja> why u timing out :-P
[19:05] <smoser> because its not there.
[19:05] <smoser> the -1 is what i was annoyed at
[19:05] <smoser> [10/-1s]
[19:05] <harlowja> ya, thats odd, lol
[19:06] <harlowja> are u in a time machine
[19:06] <harlowja> will investigate
[19:07] <smoser> well that is maxwait
[19:07] <smoser> oh.
[19:12] <smoser> it seems like it should timup should have happened
[19:18] <harlowja> did ntpd jump?
[19:18] <harlowja> looking at code, one sec
[19:22] <smoser> i think there is no timeout set (or it is actually set to -1)
[19:22] <smoser> for that datasource
[19:22] <smoser> but for some reason its trying agiain
[19:30] <harlowja> smoser ya, its trying the provided urls, but failing after all of them don't work
[19:30] <harlowja> http://169.254.169.254/openstack/2012-08-10/meta_data.json, http://169.254.169.254/openstack/2013-04-04/meta_data.json'
[19:30] <harlowja> then after none of those work, it gives up 
[19:31] <smoser> ah. i see. you're right.
[19:31] <harlowja> https://code.launchpad.net/~harlowja/cloud-init/no-negative-max-wait/+merge/215020
[19:31] <smoser> so i guess ideally if we got a timeout
[19:31] <smoser> then we wouldn't try other entries.
[19:31] <smoser> as opposed to a 404
[19:31] <harlowja> well should we :)
[19:32] <harlowja> might be different hostname in other urls
[19:32] <harlowja> right?
[19:32] <harlowja> bb, got a friend coming over
[19:32] <smoser> well it hink its urls ar ll just the MD url
[19:32] <smoser> but with different versions
[19:33] <smoser> unlikely a.) that you have a friend coming over :) , and b.) that you'd find one metadata version on the same metadata server after timing out on another i think
[19:33] <smoser> ie:
[19:33] <harlowja> lol
[19:33] <harlowja> he's not here yet :-P
[19:33] <smoser> wget http://$IP/2012-08-10/meta_data.json
[19:33] <harlowja> but i agree
[19:33] <harlowja> in this case we can abort early
[19:34] <harlowja> *or likely can abort early
[19:34] <smoser> so i'm not really opposed to the -1
[19:34] <smoser> that seems not entirely unreasonable
[19:34] <smoser> i just didn't realize htat it was trying different urls
[19:34] <smoser> i thought it was re-trying, waiting for some future time when -1 seconds had passed :)
[19:35] <harlowja> back to the future
[19:35] <smoser> if i sit around and wait long enough, it will be -1 seconds from now.
[19:35] <harlowja> lol
[19:36] <harlowja> let me do the change to abort it if its all the same hostname
[19:36] <harlowja> bb
[21:12] <harlowja> smoser https://code.launchpad.net/~harlowja/cloud-init/early-abort/+merge/215036 
[21:12] <harlowja> think that should do it