15:00:48 <samstav> #startmeeting satori
15:00:49 <openstack> Meeting started Mon Jul 14 15:00:48 2014 UTC and is due to finish in 60 minutes.  The chair is samstav. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:50 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:52 <openstack> The meeting name has been set to 'satori'
15:01:48 <samstav> Hi, welcome to the satori meeting
15:01:53 <samstav> #link https://wiki.openstack.org/wiki/Satori
15:05:27 <samstav> Anyone here for satori?
15:05:29 <walker_> o/
15:05:49 <samstav> Cool.
15:06:08 <samstav> I'll go ahead and get started with action items, a few of which we declared last week
15:06:11 <walker_> I see zns in here, but not sure if he is active
15:06:31 <samstav> #topic action items
15:07:34 <samstav> Speaking of zns, we can continue to carry this action item until he says otherwise
15:07:36 <samstav> #action zns start ML discussion & blueprint for built-in python data plane discovery
15:07:53 <walker_> haha
15:10:10 <samstav> walker_ you had an item, for creating a blueprint regarding a configurable escalation prefix for running remote commands
15:10:34 <walker_> Yes. We probably need to push that out to next week
15:10:42 <samstav> ok, sounds good
15:10:51 <samstav> here's the minutes from last week btw
15:10:52 <walker_> I think there is still some work to be done around that
15:10:54 <samstav> #link http://eavesdrop.openstack.org/meetings/satori/2014/satori.2014-07-07-15.01.html
15:11:18 <samstav> ok, cool. anything you'd like to discuss during this meeting?
15:11:46 <samstav> I mean, do you want to have a short discussion about the escalation stuff during this meeting?
15:11:57 <walker_> Yes, I think we should
15:12:14 <walker_> Basically, there is no assurance that a system will have sudo access
15:12:33 <walker_> and it may still be desireable for people to login with an unprivileged account
15:13:05 <walker_> so, there needs to be a way to tell Satori to use something other than 'sudo' to escalate privileges
15:13:55 <walker_> What seems to work best is to allow the inclusion of root-password, that way Satori can login with an unpriviledged account and then use 'su' to assume the root account
15:14:19 <samstav> #topic running escalated commands through satori
15:14:51 <samstav> #info walker_: Basically, there is no assurance that a system will have sudo access. and it may still be desireable for people to login with an unprivileged account. so, there needs to be a way to tell Satori to use something other than 'sudo' to escalate privileges. What seems to work best is to allow the inclusion of root-password, that way Satori can login with an unpriviledged account and then use 'su' to assume the root account
15:15:34 <walker_> So, basically 'root-password' is optional and if given satori will use 'username' and 'password' to login, then use 'root-password' to escalate
15:15:38 <walker_> sound about right?
15:16:28 <samstav> that sounds great for the SSH based class. n1ck-o isn't around to comment for the windows equivalent
15:16:53 <walker_> good point, right now this is looking at Linux based hosts
15:17:40 <samstav> Cool. I'll move us back to action items, and re-apply yours for creating a blueprint?
15:17:47 <walker_> Yes please
15:17:51 <samstav> #topic action items
15:18:04 <samstav> #action walker_ Create a blueprint for choosing the escalation prefix ("sudo" OR "su -" OR ?) when running commands in ssh.py
15:18:55 <samstav> Since n1ck-o isn't here, I'm going to push his action item forward as well
15:19:05 <samstav> #action Address comments on submission 98383
15:19:17 <samstav> though, based on the patch sets I saw this week, it looks like he made som
15:19:27 <samstav> some good progress.
15:20:19 <samstav> I had an action item regarding the failing unittests on 98383
15:20:26 <samstav> #topic gates
15:22:58 <samstav> While I'm still uncertain about the root of the problem, I did track it down, and it pointed to the python-whois lib we use.
15:23:08 <samstav> This is regarding the unittests
15:23:15 <walker_> Nice @samstav
15:23:27 <walker_> er, minus the @
15:23:58 <samstav> Fixed our unit tests in test_dns.py for now, and submitted a patch upstream
15:24:00 <samstav> #link https://github.com/joepie91/python-whois/pull/29
15:24:37 <samstav> One of my only guesses it that a new version was released, and the old version was removed from pypi
15:24:53 <samstav> Just a guess.
15:25:47 <samstav> #link https://github.com/samstav/satori/commit/269da49c4a50572037ae84e3bfb8e7c3c97f32c6#diff-6
15:26:19 <samstav> I moved on from there to see why the pep8 and other gates were failing, again, not really getting to the root cause but still solved the problem.
15:26:52 <samstav> Made some updates to our requirements file
15:26:56 <samstav> #link https://github.com/samstav/satori/commit/269da49c4a50572037ae84e3bfb8e7c3c97f32c6#diff-9
15:27:33 <samstav> ( I'm linking to github instead of gerrit, I know. Sorry, but the interface is better. I promise that's exactly what's in my patch set. )
15:27:40 <zns> Hi - I'm here, but might not contribute much yet as I'm trying to get up to speed. FYI
15:29:08 <samstav> The empty files fix was integrated into the 1.5.7 release of pep8
15:29:10 <samstav> https://pypi.python.org/pypi/pep8/1.5.7
15:30:15 <samstav> 👍
15:30:51 <walker_> yay!
15:31:07 <walker_> that was annoying
15:31:30 <samstav> I then rebased our fork of hacking on hacking's latest release
15:31:48 <samstav> To deal with the dependency VersionConflict
15:32:00 <samstav> and tagged it @satori
15:32:09 <samstav> #link https://github.com/samstav/hacking/tree/satori
15:32:22 <walker_> oh good. I've been running into a lot of version conflict issues
15:32:25 <walker_> thanks for doing that
15:32:39 <samstav> That branch includes zns's commits
15:32:52 <samstav> 1. Detect egg-linked libraries as third-party
15:33:00 <samstav> 2. Better detection of stdlibs in virtualenvs
15:33:49 <samstav> We will need to make those same changes for our other projects' requirements files.
15:35:09 <samstav> Those fixes/updates, and a few others are included in my submission which is intended to extend the windows-support submission
15:35:11 <samstav> #link https://review.openstack.org/#/c/106134/
15:36:49 <samstav> There is a python2.x/3.x related fix in utils.py in there, where I included a "universal" unicode() function
15:36:51 <samstav> https://github.com/samstav/satori/commit/269da49c4a50572037ae84e3bfb8e7c3c97f32c6#diff-8
15:36:53 <samstav> #link https://github.com/samstav/satori/commit/269da49c4a50572037ae84e3bfb8e7c3c97f32c6#diff-8
15:37:35 <samstav> You will notice they have the same signatures
15:37:43 <samstav> python 2.x
15:37:47 <samstav> #link https://docs.python.org/2/library/functions.html#unicode
15:37:54 <samstav> python 3.x
15:37:56 <samstav> #link https://docs.python.org/3/library/stdtypes.html#str
15:39:24 <samstav> That extension also moves posh_ohai.py into its own file, as its own independent provider. There is an exception pattern which raises when an unsupported platform tries to use the wrong provider
15:39:42 <samstav> #link https://github.com/samstav/satori/commit/269da49c4a50572037ae84e3bfb8e7c3c97f32c6#diff-5
15:40:46 <samstav> And lastly, in the gates topic, there is only the beginnings of python 3 support in eventlet
15:40:48 <samstav> #link https://github.com/eventlet/eventlet/issues/6
15:41:21 <samstav> This is the only thing keeping submission 106134 from passing all gates
15:42:02 <samstav> #info submission 106134 depends_on/includes submission 98383
15:44:42 <samstav> #action zns, samstav, n1ck-o talk_about/research either 1) ditching eventlet, 2) finding an alternative greenthread lib w/ python 3 support, 3) remove python3 gate 4) use python native threading support
15:45:39 <samstav> #topic action items
15:46:18 <samstav> I had one more item which was to create a blueprint for removing the subprocess calls from pse.py (renamed to smb.py)
15:47:08 <samstav> It is minimal and I will expand it some
15:47:10 <samstav> #link https://blueprints.launchpad.net/satori/+spec/no-subprocess
15:47:16 <samstav> So that's done.
15:47:51 <samstav> #topic général discussion
15:48:07 <samstav> That's about it, unless anyone would like to discuss anything else?
15:49:46 <walker_> thanks samstav, nothing more from me
15:50:04 <samstav> cool. Thanks for your attendance. See y'all next week
15:50:08 <samstav> #endmeeting