17:00:31 <gundalow> #startmeeting Testing Working Group
17:00:31 <zodbot> Meeting started Thu May 11 17:00:31 2017 UTC.  The chair is gundalow. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:31 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
17:00:31 <zodbot> The meeting name has been set to 'testing_working_group'
17:00:34 * mattclay waves
17:00:38 <gundalow> #chair mattclay
17:00:38 <zodbot> Current chairs: gundalow mattclay
17:01:33 <gundalow> mattclay: `I've got ansible-test network-integration -vvvv --platform vyos/1.1.0 vyos_command` passing locally now
17:01:57 <mattclay> gundalow: Excellent. Did it require any changes to make it work?
17:03:46 <gundalow> mattclay: https://github.com/ansible/ansible/pull/24514/files
17:05:05 <gundalow> The group_vars were taking precedence over the ssh key
17:05:25 <mattclay> Seems simple enough. I think there's a minor formatting issue pep8 will complain about -- no space before the % on the last line of the diff.
17:05:50 <gundalow> ah, yes
17:05:57 <mattclay> Actually, neither of those need the '%s' you can just put the var in there.
17:06:02 <mattclay> Since no formatting is happening.
17:06:20 <gundalow> hum, I may have to remove the `aliases` from the other targets/
17:06:23 <mattclay> Actually, the first one should be '%s_command' % var
17:06:58 <sivel> yo
17:07:07 <gundalow> #chair sivel
17:07:07 <zodbot> Current chairs: gundalow mattclay sivel
17:07:07 <mattclay> If the others aren't passing, just remove the ci aliases.
17:08:18 <gundalow> Nice to see Environment Description in the ansible-test output now :)
17:08:33 <mattclay> Ah, running with -vvv ?
17:08:46 <gundalow> always
17:10:31 <gundalow> I'll continue poking this. Got another or so left of today, so hopefully I can get a green run for you
17:10:59 <gundalow> https://github.com/ansible/ansible/pull/24514/files#diff-5ed3c6596f1e495f6d0c6733e41ac3b5R2
17:11:04 <gundalow> For that I'd ideally like
17:11:10 <mattclay> Excellent. I'll try to finish up my coverage changes tomorrow. If you have any network tests passing I can make sure they keep working. :)
17:11:44 <gundalow> use ssh key if specified, otherwise set those vars
17:11:50 <gundalow> though not sure if that's possible
17:12:15 <gundalow> That can be a problem for next week
17:12:22 <mattclay> I don't see why not. We generate an SSH key for some of the other tests.
17:12:32 <mattclay> But yeah, can do that next week.
17:12:34 <gundalow> :)
17:12:38 <gundalow> That's it from me
17:14:28 <mattclay> The cloud module integration tests have been deployed to Shippable.
17:14:51 <mattclay> We now run CloudStack and AWS integration tests on PRs.
17:15:04 <mattclay> We have pretty good coverage for CloudStack. AWS is small but growing.
17:15:42 <mattclay> jtanner is actually starting to look at using the cloud plugins to be able to test vsphere using a simulator like we do for CloudStack.
17:16:01 <jtanner> maybe not exactly like that, but close
17:16:14 <jtanner> working on flask controller now
17:16:22 <mattclay> I've been overhauling the ansible-test code coverage injector to fix some issues so I can automate code coverage reporting.
17:16:37 <mattclay> I hope to have that pushed tomorrow.
17:17:12 <mattclay> If anyone wants to work on AWS module tests, let me know. As we get more tests I'll need to update our IAM policy and sweeper to support the additional permissions required by the tests.
17:17:47 <mattclay> We've been having some issues with timeouts on Windows tests due to lack of T2 launch credits on AWS. I've put in a request to get the limit increased and I'm waiting to hear back on that.
17:18:04 <mattclay> That's it from me.
17:19:45 <gundalow> sivel: Did you have anything
17:19:50 <sivel> I have nothing
17:19:59 <gundalow> Any comments about line numbers in validate-modules?
17:20:23 * gundalow missed last weeks meeting, so that might have been discussed already
17:21:23 <sivel> It was merged
17:21:34 <sivel> there is now native support for line numbers and column
17:21:49 <sivel> there are several TODOs listed in the codebase that need line/column added
17:22:08 <gundalow> oh, ace
17:23:06 <gundalow> hum, wonder if I can run validate-modules with --warnings enabled and hook it upto Vim's Syntastic framework
17:23:23 <sivel> The standard output is not similar to pep8 also
17:23:26 <sivel> is now*
17:23:54 <sivel> so the old style for printing is gone, and now prints in a manner that I guess syntastic can understand
17:24:05 <gundalow> aye, looks like it
17:24:20 <mattclay> sivel: We also have the --lint option to ansible-test for use with things like syntastic
17:24:41 <gundalow> alikins: were you looking at syntastic?
17:24:53 <sivel> mattclay: yeah, I just figured it was about time to get it in alignment
17:25:01 <sivel> instead of some random output format
17:25:04 * mattclay nods
17:25:07 <gundalow> +1
17:25:19 <alikins> gundalow: not recently
17:25:36 <gundalow> alikins: ah, maybe I miss remembered
17:26:04 <alikins> i've switch to 'ALE' in liue of syntastic, but same idea I suppose
17:26:06 <gundalow> If I get a free weekend I may see if I can get something running
17:26:46 <gundalow> ooooooooh, so I don't have to wait 5+ seconds for vim to load, nice
17:26:49 <alikins> though ALE is async, so it might be useable with ansible-test where syntastic isnt really
17:27:16 <gundalow> oh, ALE has ansible-lint support built in
17:28:21 <gundalow> alikins: thanks for the pointer
17:31:49 <mattclay> Does anyone have anything else to discuss?
17:32:07 <gundalow> Nothing else from me
17:35:30 <gundalow> Thanks everyone
17:35:33 <gundalow> #endmeeting