Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2016-09-09

| Channels | #puppet-openstack index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:00 bkero redhat-release >= 22.0
00:00 EmilienM yes I saw
00:01 bkero Build Date  : Thu 08 Sep 2016 11:07:44 AM PDT
00:01 bkero :(
00:01 EmilienM I don't even know if we need to install redhat-release
00:01 bkero http://paste.openstack.org/show/569352/
00:01 EmilienM but let's try...
00:02 bkero Hmm. sudo yum install redhat-release -> Package centos-release-7-2.1511.el7.centos.2.10.x86_64 already installed and latest version
00:03 EmilienM in your machine?
00:03 EmilienM it's not in gate
00:03 EmilienM see rpm-qa log file
00:04 bkero EmilienM: yep, no redhat-release on there, but I'm betting centos-release satisfies that
00:04 bkero but I don't know about the version restriction
00:04 EmilienM lets try
00:04 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Align the stars to bring our CI back  https://review.openstack.org/367551
00:04 EmilienM seriously puppetlabs packaging is terrible
00:04 EmilienM and I'm looking forward to switch on rdo's packaging
00:05 EmilienM Hunner, _ody: if you can ping packaging people that would be great
00:05 * bkero pings in #puppet-dev in the meantime.
00:07 EmilienM bkero: http://logs.openstack.org/51/367551/8/ch​eck/gate-puppet-openstack-integration-3-​scenario001-tempest-centos-7/fc2c291/con​sole.html#_2016-09-09_00_05_47_317332
00:07 EmilienM we're stuck
00:07 bkero EmilienM: we could use an older build of the puppetlabs-release rpm...if we could find one.
00:08 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Align the stars to bring our CI back  https://review.openstack.org/367551
00:08 EmilienM back to origin
00:08 EmilienM and i'll stop for today.
00:09 * bkero peers at http://yum.puppetlabs.com/el/7Server/product​s/x86_64/puppetlabs-release-7-12.noarch.rpm
00:09 johnmilton joined #puppet-openstack
00:09 bkero I think the original is just a symlink to http://yum.puppetlabs.com/el/7Server/products​/x86_64/puppetlabs-release-22.0-1.noarch.rpm
00:10 bkero EmilienM: it's a bad time for people to be around on the west coast -- 1710.
00:11 EmilienM I'll create a ticket and go out
00:11 bkero okay
00:12 _ody EmilienM: thanks. Will escalate.
00:13 bkero EmilienM: yep, this is totally what happened. A working log shows puppetlabs-release 7.12: http://logs.openstack.org/20/364620/2/check/g​ate-puppet-openstack-integration-3-scenario00​1-tempest-centos-7/9844a50/logs/rpm-qa.txt.gz
00:13 markvoelker joined #puppet-openstack
00:14 bkero https://github.com/puppetlabs/puppetl​abs-release/blob/master/templates/red​hat/puppetlabs-release.spec.erb#L21
00:14 EmilienM https://tickets.puppetlabs.com/browse/CPR-378
00:15 * bkero considers sending in a patch for a Requires: redhat-release | centos-release
00:18 EmilienM bkero: https://github.com/puppetlabs/puppetlabs-release​/commit/6027433ce8fb94f93782b375b353a546f24f7dce
00:18 EmilienM this one?
00:18 bkero EmilienM: that looks like it
00:19 EmilienM bkero: any workaround? can we download an older version?
00:19 bkero EmilienM: I'm betting it might be some automatoin that moved the symlink to the new version since 22 > 7.
00:19 bkero EmilienM: yes, have it grab  http://yum.puppetlabs.com/el/7Server/product​s/x86_64/puppetlabs-release-7-12.noarch.rpm instead
00:20 EmilienM ok let's try
00:21 bkero since Requires: redhat-release >= %{version}, it should only require centos-release/redhat-release >= 7.12 instead of 22.
00:21 EmilienM bkero: but we'll miss the pc1 package
00:21 EmilienM I don't see pc1 in https://yum.puppetlabs.com/​el/7Server/products/x86_64/
00:22 EmilienM oh, in https://yum.puppetlabs.com/el/7Server/PC1/x86_64/
00:22 EmilienM last one was puppetlabs-release-pc1-1.0.0-2.el7.noarch.rpm  ?
00:22 EmilienM 19-Apr-2016 sounds old to me
00:22 bkero yeah
00:22 bkero well, I guess they didn't have to change it
00:22 bkero EmilienM: the new package has a repo file that doesn't reference PC1 either.
00:23 EmilienM this would be a huge regression for us
00:23 EmilienM we had so much fixed in puppet4
00:24 EmilienM I give up, good luck
00:24 bkero EmilienM: http://paste.openstack.org/show/569354/
00:25 bkero nothing changed in the rpeo package, it should not cause a regression
00:25 bkero repo*
00:26 EmilienM bkero: ok so we should hardcode urls?
00:27 bkero EmilienM: It might be worth trying for now to get CI working again, then can change back when it's fixed. :/
00:27 bkero It's just 1 line in p-o-i, right?
00:27 EmilienM no
00:27 EmilienM because we have some logic for pc1 or old package
00:28 EmilienM i'm on it..
00:28 bkero oh, I haven't looked at that stuff :/
00:28 bkero EmilienM: until today that 7-12 package that I linked was the one that was used
00:28 bkero I looked at a job from yesterday and rpm-qa.txt referenced it
00:32 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Align the stars to bring our CI back  https://review.openstack.org/367551
00:33 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Align the stars to bring our CI back  https://review.openstack.org/367551
00:39 EmilienM bkero: puppet is running, let's see how it goes
00:40 * bkero crosses fingers
00:40 EmilienM this day will never finish
00:41 bkero :(
00:41 bkero at least it's only 1 day
00:45 EmilienM bkero: do you have a PR for puppet package?
00:45 EmilienM to fix the min version?
00:51 bkero EmilienM: Still trying to figure one out without breaking fedora due to that comment at the top of the commit you posted.
00:53 * bkero just tries altering the variable in there
00:53 bkero EmilienM: yeah, have a pr, just need to fork and submit
00:54 EmilienM bkero: ++
00:58 bkero EmilienM: https://github.com/puppetlab​s/puppetlabs-release/pull/72
00:58 bkero I'm hoping that works.
00:59 EmilienM bkero: I think you should put the bug number in commit message (headers even)
01:00 EmilienM bkero: https://github.com/puppetlab​s/puppetlabs-release/pull/71
01:00 EmilienM it overlaps with it ^
01:00 EmilienM I think his PR is more complete
01:01 EmilienM her* sorry for that
01:01 bkero EmilienM: yes
01:01 bkero also melissa wants to merge 1 commit into puppetlabs:master from melissa:maint/master/fuck
01:01 bkero gg
01:02 bkero closed
01:02 EmilienM ok
01:02 bkero at least I fixed the right thing :)
01:02 bkero or rather fixed it the same way
01:02 EmilienM :)
01:02 EmilienM thx for trying
01:03 bkero np
01:03 bkero I missed the Release number, so I'd have to do it again anyway
01:03 bkero and then someone over there would still need to merge it and trigger the build/release processes
01:03 EmilienM at least they are super reactive
01:04 bkero yup
01:04 EmilienM watching 367551 in zuul, it looks green
01:04 bkero the pc1 release seems to be fixed
01:04 EmilienM I'll merge it once it's done
01:04 EmilienM manually you mean?
01:04 bkero or rather it never had a redhat-release dep
01:04 bkero puppetlabs-release-pc1-el-7.noarch.rpm doesn't have a redhat-release dep
01:04 bkero http://yum.puppetlabs.com/pupp​etlabs-release-el-7.noarch.rpm still broken :(
01:04 EmilienM ok
01:04 EmilienM we'll pin
01:05 mtreinish joined #puppet-openstack
01:06 EmilienM I confirm the pin doesn't affect the version of puppet
01:06 EmilienM so we're good
01:07 bkero cool :)
01:07 bkero Does that mean CI is passing again?
01:08 EmilienM yes
01:08 EmilienM but not yet
01:08 EmilienM we need to merge my patch, I'll auto approve it
01:11 bkero EmilienM: I am doing a nasty $ while /bin/true; do sleep 5; rm puppetlabs-release-el-7.noarch.rpm; wget --quiet http://yum.puppetlabs.com/puppe​tlabs-release-el-7.noarch.rpm; rpm -qpiR puppetlabs-release-el-7.noarch.rpm |grep redhat-release; done
01:11 bkero and will let you know when it changes
01:11 EmilienM bkero: lol
01:11 bkero EmilienM: it's up!
01:11 EmilienM funny command
01:12 EmilienM yeah?
01:12 bkero and...err.
01:12 bkero http://paste.openstack.org/show/569539/
01:12 bkero they didn't sign it
01:12 EmilienM I'm leaving for real now
01:12 EmilienM we'll pin, no worries
01:12 bkero ok
01:12 bkero Thanks for dealing with all this today. The project wouldn't be where it is without you. :)
01:13 EmilienM selfish2 +2 +A
01:13 EmilienM bkero: well, it's a teamwork
01:13 EmilienM see you later
01:20 _ody EmilienM: bkero: Fix should be pushed.
01:21 bkero _ody: saw, is pushed
01:21 bkero but not signed :)
01:21 _ody They didn't sign the package?
01:21 _ody AAARGH!
01:21 bkero _ody: warning: puppetlabs-release-el-7.noarch.rpm: Header V4 RSA/SHA512 Signature, key ID 4bd6ec30: NOKEY
01:25 _ody bkero: odd. Double checked with release team and they double checked.
01:26 bkero huh
01:26 _ody ...isn't that rpm the old pre-PC1 release package?
01:26 bkero Yes
01:26 bkero The PC1 package never had the Depends that broke everything
01:26 _ody bkero: Yeah.  That's what they just pushed a fix for.
01:27 bkero _ody: the one in the docs references $ sudo rpm -ivh https://yum.puppetlabs.com/pupp​etlabs-release-el-7.noarch.rpm
01:27 bkero which is the file that doesn't seem to be signed
01:27 bkero can you try it and confirm? Maybe I'm just missing the key.
01:28 _ody bkero: Ok.  Third check, yep some things are unsigned.
01:28 bkero but it didn't throw the warning before.
01:28 bkero _ody: http://paste.openstack.org/show/569539/
01:28 bkero I did a little while loop to check
01:28 bkero _ody: we're pinning to an older release rpm so this isn't blocking anymore
01:48 ilbot3 joined #puppet-openstack
01:48 Topic for #puppet-openstack is now CI status: green - Puppet OpenStack modules http://docs.openstack.org/deve​loper/puppet-openstack-guide/
02:02 openstackgerrit Merged openstack/puppet-openstack-integration: Align the stars to bring our CI back  https://review.openstack.org/367551
02:02 EmilienM finaly
02:02 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Move ceph object len options to profile  https://review.openstack.org/367022
02:07 _ody Did we end up pinning the package or has the packaging been fixed?
02:11 jperry joined #puppet-openstack
02:13 michapma_alt joined #puppet-openstack
02:13 bkero We're pinning
02:13 bkero Package is fixed, but still not signed. Dunno if that matters. :)
02:25 openstackgerrit Xiaohua Yuan proposed openstack/puppet-nova: test new instance  https://review.openstack.org/367719
02:33 clayton joined #puppet-openstack
02:37 colonwq EmilienM, Can you compare 367022 to  358029 and 366791 when you get the opportunity?
03:34 skramaja joined #puppet-openstack
03:42 imcsk8_ joined #puppet-openstack
04:42 karthiks joined #puppet-openstack
04:49 snarwade joined #puppet-openstack
05:16 moravec joined #puppet-openstack
05:21 jaosorior joined #puppet-openstack
05:32 karthiks joined #puppet-openstack
05:43 openstackgerrit Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Fix value of ctlplane fqdn fact  https://review.openstack.org/367754
05:53 aimon joined #puppet-openstack
05:53 openstackgerrit Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Fix value of ctlplane fqdn fact  https://review.openstack.org/367754
06:08 aimon joined #puppet-openstack
06:40 openstackgerrit Merged openstack/puppet-ceph: Update Jewel Ceph RGW to support Keystone V3  https://review.openstack.org/347956
06:56 fxpester joined #puppet-openstack
06:57 mattb joined #puppet-openstack
07:01 beddari joined #puppet-openstack
07:06 social joined #puppet-openstack
07:07 tesseract- joined #puppet-openstack
07:49 openstackgerrit OpenStack Proposal Bot proposed openstack/puppet-openstack-integration: Promote RDO repository to latest consistent URL  https://review.openstack.org/367154
07:57 hjensas joined #puppet-openstack
07:57 samuelBartel joined #puppet-openstack
08:02 openstackgerrit Karthik S proposed openstack/puppet-vswitch: Added leading zeros to the pci address set by facter  https://review.openstack.org/367802
08:04 derekh joined #puppet-openstack
08:19 openstackgerrit Nick Jones proposed openstack/puppet-glance: Amend comments regarding configuration of $log_file  https://review.openstack.org/367815
08:25 dgurtner joined #puppet-openstack
08:25 dgurtner joined #puppet-openstack
08:38 dsutyagin joined #puppet-openstack
08:44 lutzb joined #puppet-openstack
08:56 imcsk8 joined #puppet-openstack
09:11 shardy joined #puppet-openstack
10:04 lucasagomes joined #puppet-openstack
10:08 richm joined #puppet-openstack
10:18 mgould joined #puppet-openstack
10:48 openstackgerrit Merged openstack/puppet-glance: Update log_file documentation  https://review.openstack.org/367696
10:58 cdelatte joined #puppet-openstack
11:10 ansmith joined #puppet-openstack
11:27 fultonj joined #puppet-openstack
11:40 iurygregory morning o/
11:44 colonwq iurygregory, and it is Friday too.
11:44 iurygregory colonwq, yeah but to me is a day with 4 meetings .-.
11:46 EmilienM hello
11:46 colonwq iurygregory, I have it easy. There are only 3.... wait 4 there is a makeup one.
11:47 iurygregory colonwq, and today is the end of the internal sprint \o/ haha
11:49 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: ubuntu: re-enable Ironic and Linuxbridge  https://review.openstack.org/367965
11:49 iurygregory ^ very fast the fix O.o
11:50 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Unpin Puppetlabs release package  https://review.openstack.org/367966
11:54 openstackgerrit Keith Schincke proposed openstack/puppet-openstack-integration: Introduce scenario004  https://review.openstack.org/339106
11:55 colonwq EmilienM, was your swap space fix yesterday only for 001 or is it for all testing scenarios?
11:55 EmilienM all scenarios
11:57 openstackgerrit Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Terminate Zaqar websocket endpoint in HAProxy  https://review.openstack.org/360329
11:58 colonwq cool. thanks.
12:04 jperry joined #puppet-openstack
12:27 openstackgerrit Merged openstack/puppet-openstack-integration: Promote RDO repository to latest consistent URL  https://review.openstack.org/367154
12:28 pradk joined #puppet-openstack
12:29 * yankcrime wishes he'd noticed that mwhahaha had fixed the comment-related bug he discovered last night before going through the rigmorole of raising a a bug report and then proposing a fix...
12:31 ansmith joined #puppet-openstack
12:32 iurygregory yankcrime, https://review.openstack.org/#/c/367696/ ?
12:32 yankcrime iurygregory: that's the one
12:33 skramaja joined #puppet-openstack
12:53 jaosorior joined #puppet-openstack
13:09 johnmilton joined #puppet-openstack
13:22 openstackgerrit David Moreau Simard proposed openstack/puppet-openstack-integration: Add magnum support to scenario002  https://review.openstack.org/367012
13:26 StraubTW joined #puppet-openstack
13:32 pradk joined #puppet-openstack
13:39 jperry joined #puppet-openstack
13:44 mwhahaha ceilometer beaker is broken on c7, http://logs.openstack.org/98/366098/5/check/gate​-puppet-ceilometer-puppet-beaker-rspec-centos-7/​864c690/console.html#_2016-09-09_08_15_17_585071
13:44 mwhahaha guess it requires the gnocchiclient now?
13:45 openstackgerrit Alex Ruiz Estradera proposed openstack/puppet-neutron: Fix the neutron-port resource with binding options  https://review.openstack.org/368014
13:45 colonwq iurygregory, I found my Heisenberg from yesterday. One of my images i use for cloning had bad data .
13:46 mjblack mwhahaha: maybe, depends on the config?
13:46 mjblack it shouldnt require it
13:46 mwhahaha you'd think that wouldn't you
13:46 mjblack it was that way in mitaka
13:47 mwhahaha https://github.com/openstack/ceilometer/b​lob/master/ceilometer/cmd/storage.py#L24
13:48 mwhahaha bet the packaging missed this new requirement
13:48 mjblack guarantee they did
13:48 iurygregory colonwq, awesome \o/
13:48 dfisher joined #puppet-openstack
13:49 iurygregory packaging again? omg D:
13:49 mwhahaha we'll see
13:49 * mwhahaha goes looking for specs
13:50 mwhahaha yup it's missing
13:50 mwhahaha is in thet buildrequires but not in the requires
13:50 iurygregory D:
13:51 * mwhahaha throws things at dmsimard
13:51 dmsimard wat
13:51 mwhahaha dmsimard: where do i report packaging issues :D
13:51 mwhahaha ceilometer depends on gnocchiclient now
13:51 dmsimard if it's a quick fix just poke me
13:51 dmsimard ok
13:52 dmsimard looking
13:53 mwhahaha i guess this passed because we include gnocchi in our scenario
13:53 mwhahaha but not in beaker
13:53 dmsimard mwhahaha: it's not even in reqs.txt .. https://github.com/openstack/ceilo​meter/blob/master/requirements.txt
13:53 mwhahaha we can blame sileht
13:53 iurygregory lol .-.
13:53 mwhahaha https://github.com/openstack/ceilometer/comm​it/9709608bb1bab999658475889637994cccf8bbfc
13:53 dmsimard well
13:53 mjblack mwhahaha: well
13:54 mjblack that change really is descriptive...
13:54 dmsimard gnocchi has a lot of "optional" dependencies
13:54 mjblack https://review.openstack.org/#/c/350611/
13:54 dmsimard like another optional dependency is python-cradox when using rbd as gnocchi backend
13:54 dmsimard python-cradox is not in reqs.txt but it's required when running rbd as gnocchi backend
13:54 dmsimard perhaps this is one of those other optional deps
13:54 EmilienM I'll blame sileht
13:54 * mwhahaha sighs
13:54 mjblack dmsimard: cradox is like a 3rd party library
13:55 mjblack python-rados is the library to use as of ceph jewel
13:55 mjblack speaking of which, mwhahaha, is the CI still failing for gnocchi?
13:55 mwhahaha no
13:55 mwhahaha should be fixed as of last night
13:55 mjblack ok, I'll kick off a recheck then
13:57 dmsimard mwhahaha: so, well, I mean.. I'll ask the other RDO guys but we don't typically "hard" require optional deps. If the user installs something that requires an optional dep, he should install it as well. It's like saying we should install swift, ceph and gluster along with glance because glance has different backends.
13:58 mwhahaha dmsimard: yea i get it, the problem is it's blowing up the ceilometer db sync which is kinda important to ceilometer and I don't think we've set anything optional that should trigger gnocchiclient
13:58 mwhahaha i'll double check today
13:58 dmsimard mwhahaha: we're following up with sileht about it right now
13:58 mwhahaha but it seems that the gnocchiclient snuck in somehow since that patch was from early august and it just blew up
13:59 dmsimard mwhahaha: 09:58<sileht>EmilienM, Gnocchi is the default now, so having it in requirement make sense
13:59 dmsimard they're going to add it to reqs.txt and we'll adjust packaging
13:59 EmilienM i'm preparing a patch
13:59 mjblack mwhahaha: it was merged september 6th
14:00 mwhahaha oh that makes sense
14:00 Ahharu joined #puppet-openstack
14:00 mwhahaha it's too early for this stuff :D
14:00 dmsimard yeah commit != merge
14:00 dmsimard I get caught by that often too.
14:01 Ahharu I totally forgot about backporting my change :D If you can take a look i'll appreciate it https://review.openstack.org/#/c/368014/
14:01 dmsimard easiest way not to get mistaken is to search "project:openstack/<project> status:merged" in gerrit and look through what merged recently
14:01 dmsimard instead of relying on git/github
14:01 mwhahaha i usually just grab the change-id
14:03 dmsimard the changeid is not in the merge commit though
14:03 mjblack dmsimard: it is
14:03 dmsimard huh
14:03 mjblack https://github.com/openstack/ceilometer/comm​it/9709608bb1bab999658475889637994cccf8bbfc
14:04 dmsimard mjblack: that's the commit
14:04 dmsimard merges look like this: https://github.com/openstack/ceilometer/comm​it/f78849c9660b8e73aa2a65b12775804eced2ecdb
14:04 dmsimard sometimes the merge will have just occured but the commit dates back weeks/months ago
14:04 mjblack hmmm
14:05 mjblack dont see the merge commit
14:07 openstackgerrit Sven Anderson proposed openstack/puppet-tempest: Add provider for ec2 credentials  https://review.openstack.org/364580
14:14 skramaja joined #puppet-openstack
14:20 mwhahaha no i grab the change-id and punch it into search in gerrit
14:20 mwhahaha takes you right to the review :D
14:22 dmsimard EmilienM: damn, it looks like the bot updated the hash to right up before the magnum fix built
14:22 dmsimard EmilienM: need another promotion, magnum still missing werkzeug :(
14:22 EmilienM dmsimard: no way ! -2
14:23 iurygregory promote RDO again?  omg
14:24 mwhahaha what will it break today!
14:24 mwhahaha adventures in packaging
14:24 iurygregory adventure in openstack \o/ next chapter: ...
14:25 dmsimard mwhahaha: I prefer that then UCA that updates and boom 20 issues in your face
14:25 iurygregory dmsimard, yeah
14:25 dmsimard RDO is all about that CI/CD goodness of following trunk
14:26 mwhahaha s/goodness/questionable life choices
14:26 iurygregory lol
14:26 dfisher hah
14:27 dmsimard well, I mean, it IS kinda depressing that things are always broken
14:27 dmsimard but I would not expect otherwise from testing and developing against the bleeding edge all the time
14:27 iurygregory true story
14:27 dmsimard If you would not balance all that fixing throughout the whole cycle, when would you do it ? Once the release ships ? And then you're releasing 3 months after the release.
14:28 mwhahaha don't get me wrong, i prefer the rdo method
14:29 mwhahaha it's just from puppet stand point because you don't gate on all of our stuff, something probably gets broken :D but at least it's a quick turnaround for rdo
14:29 dmsimard mwhahaha: we actually do gate on your stuff
14:29 dmsimard mwhahaha: but EmilienM doesn't want none of our gate
14:29 mwhahaha no the scenarios
14:29 dmsimard he bypasses it
14:30 mwhahaha er not
14:30 dmsimard mwhahaha: we're running all 3 scenarios in our gate
14:30 mwhahaha right but we install gnocchi with ceilometer in the scenario
14:30 dmsimard people should be using current-passed-ci which runs (amongst other things) all three puppet scenarios
14:30 dmsimard but EmilienM uses consistent which lands before current-passed-ci
14:30 mwhahaha so it wouldn't have caught this issue
14:30 dmsimard It would have
14:30 mwhahaha or might not have
14:31 mwhahaha no because we might have had the client installed already
14:31 EmilienM what don't I want?
14:31 * dmsimard shrugs
14:31 aimon joined #puppet-openstack
14:31 dmsimard EmilienM: I was explaining that you're trying to promote repositories that haven't passed RDO's CI :)
14:32 dmsimard consistent vs current-passed-ci
14:32 mwhahaha oh the complexities of CI
14:32 mwhahaha brb
14:37 EmilienM dmsimard: yes because I don't care about RDO CI in Puppet OpenStack Ci
14:37 EmilienM I do care of RDO CI of course but not in this context
14:37 EmilienM I don't mind to be more recent than RDO CI
14:41 ducttape_ joined #puppet-openstack
14:46 mwhahaha also we gate our promotion on the puppet scenarios which is why i said it didn't catch this issue
14:46 openstackgerrit Romanos Skiadas proposed openstack/puppet-nova: Make cinder a soft dependency  https://review.openstack.org/366907
14:54 mjblack mwhahaha: I just checked and yeah the gnocchi ci looks fixed
14:55 openstackgerrit Carlos Camacho proposed openstack/puppet-tripleo: WIP - Add puppet-lint 2.0.0 to check puppet-tripleo manifests  https://review.openstack.org/352847
14:56 openstackgerrit Carlos Camacho proposed openstack/puppet-openstack_spec_helper: WIP - Use puppet-lint 2.0.0 instead of 1.1.0  https://review.openstack.org/353488
15:13 dmsimard mwhahaha, EmilienM: promotion passed in https://review.openstack.org/#/c/368037/
15:14 mwhahaha yay
15:14 dmsimard We'll also hopefully land something to unblock further testing with designate in p-o-i today
15:14 EmilienM -2
15:14 dmsimard so that'll require /another/ promotion but I'll let that one be natural as I won't have time for it today anyway
15:15 dmsimard right now I'm more puzzled about https://review.openstack.org/#/c/367012/ which complains about a duplicate ::magnum::logging declaration while I can only find an include in init.pp, nothing else
15:15 EmilienM _ody: any chance to finish package signatures for puppetlabs-release?
15:16 mwhahaha dmsimard: logging has to be included prior to the inclusion of ::magnum
15:16 mjblack EmilienM,mwhahaha: please review https://review.openstack.org/#/c/367438/
15:17 EmilienM omh -2 too
15:17 dmsimard mwhahaha: how does that work ? Wouldn't oslo log try to write to the magnum config file ? If ::magnum hasn't run, the magnum config file wouldn't be there ?
15:17 openstackgerrit Merged openstack/puppet-openstack-integration: ubuntu: re-enable Ironic and Linuxbridge  https://review.openstack.org/367965
15:17 dmsimard Or is there some puppet magic involved
15:17 mwhahaha puppet magic
15:18 dmsimard damn magic
15:18 iurygregory today EmilienM is in a good mood hahah
15:18 EmilienM iurygregory: I'm joking :-)
15:18 dmsimard it's friday
15:18 EmilienM it's Friday!
15:18 iurygregory EmilienM, yeah i know XD
15:18 mwhahaha dmsimard: when puppet parses the catalog, it hits ::magnum first and because of https://github.com/openstack/puppet-ma​gnum/blob/master/manifests/init.pp#L90 when it hits the class { '::magnum::logging': } it throws an already included error
15:19 mwhahaha dmsimard: but if you define ::magnum::logging prior to ::magnum the include is a noop because it's already in the catalog
15:19 openstackgerrit David Moreau Simard proposed openstack/puppet-openstack-integration: Add magnum support to scenario002  https://review.openstack.org/367012
15:19 iurygregory i rember the friday that Emilien said "It's Friday no fear" in the ML
15:19 openstackgerrit David Moreau Simard proposed openstack/puppet-openstack-integration: Add magnum support to scenario002  https://review.openstack.org/367012
15:19 openstackgerrit Bradley Klein proposed openstack/puppet-monasca: Add configuration of monasca agent vertica plugin  https://review.openstack.org/366292
15:20 dmsimard mwhahaha: back in my day includes were used exactly to avoid that kind of issue
15:20 mwhahaha it's include vs class
15:20 mwhahaha and how puppet handles it
15:20 mwhahaha yea you'd think the include would come along if class is never parsed
15:21 mwhahaha but that's not how puppet does it
15:21 mwhahaha would be awesome if they did :D
15:21 EmilienM mjblack: +2
15:21 degorenko mjblack, +A :)
15:22 mjblack EmilienM: for the most part we can enable it for xenial with the exception of statsd I think
15:22 mjblack right now poi has gnocchi stuff wrapped around a conditional and only runs for centos
15:22 EmilienM ok now everyone leaves on PTO and come back in October. So I can release Puppet modules for final Newton
15:22 mjblack lol
15:23 mjblack thats a good joke there
15:23 mwhahaha nobody touch anything
15:23 mjblack :D
15:23 * mjblack touches the modules
15:23 mwhahaha NO TOUCHING
15:23 mjblack EmilienM: I need to push through the memcache change for authtoken
15:24 iurygregory mjblack, run
15:24 iurygregory let's create stable/newton yay!
15:24 EmilienM degorenko: please remove +A on https://review.openstack.org/#/c/367438/
15:24 EmilienM mjblack: sorry
15:24 EmilienM I'm discussing on telemetry channel, they disagree with this change
15:24 degorenko EmilienM, ?
15:25 mjblack EmilienM: whats the channel?
15:25 EmilienM degorenko: I'm waiting for details
15:25 EmilienM mjblack: openstack-telemetry
15:25 degorenko EmilienM, how this change is related to telemtry?
15:25 EmilienM degorenko: gnocchi is part of telemetry dude
15:26 degorenko ah, right
15:26 mjblack EmilienM: we're not removing it yet though
15:33 mjblack EmilienM: I guess a compromise...
15:33 mjblack keep cradox as default and have python-rados in there
15:33 EmilienM mjblack: keep discussion on telemetry channel please
15:34 mjblack ok
15:41 openstackgerrit Keith Schincke proposed openstack/puppet-ceph: Update radosgw service definition for Xenial  https://review.openstack.org/367456
15:42 openstackgerrit Matthew J Black proposed openstack/puppet-keystone: Install python memcache package  https://review.openstack.org/362270
15:45 openstackgerrit Merged openstack/puppet-monasca: Add configuration of monasca agent vertica plugin  https://review.openstack.org/366292
15:46 openstackgerrit Sven Anderson proposed openstack/puppet-tempest: Add provider for ec2 credentials  https://review.openstack.org/364580
15:46 openstackgerrit Matthew J Black proposed openstack/puppet-keystone: Install python memcache package  https://review.openstack.org/362270
15:46 mjblack dfisher: ^ that makes the memcache package install optional
15:46 openstackgerrit Merged openstack/puppet-tripleo: Add Ceph RGW profile  https://review.openstack.org/334081
15:47 iurygregory mjblack, if i do not set manage_memcache_package but i set the value for memcached? O.o
15:48 iurygregory manage_memcache_package = false but memcached_servers_real contain a valid value..
15:48 mjblack iurygregory: yeah
15:48 mjblack originally it was going to install the package but there was concerns about forcing the package install
15:49 openstackgerrit joined #puppet-openstack
15:49 iurygregory if the value is set o don't see reason why not install .-.
15:49 iurygregory I don't*
15:51 mjblack I guess we can make it default to true
15:51 mjblack and then override it at the modules authtoken class
15:52 mwhahaha i wouldn't
15:52 iurygregory well we have many options like, if memcached_servers_real is valid and the value is false it should fail or something...
15:52 mwhahaha i think it's ok to have the option, but leave it off
15:53 mwhahaha unrelated, i wonder if tempest just broke us http://logs.openstack.org/12/367012/6/ch​eck/gate-puppet-openstack-integration-3-​scenario001-tempest-centos-7/475711a/con​sole.html#_2016-09-09_15_37_05_200253
15:53 mwhahaha investigating
15:53 aimon joined #puppet-openstack
15:54 iurygregory I think we should say that we will not install even if the memcached_servers is set (warning message or something whe the value is false)
15:55 openstackstatus NOTICE: New setuptools release appears to have a circular import which is breaking many jobs - check for ImportError: cannot import name monkey.
15:55 Topic for #puppet-openstack is now New setuptools release appears to have a circular import which is breaking many jobs - check for ImportError: cannot import name monkey.
15:55 iurygregory mwhahaha, ^
15:55 iurygregory LOL
15:55 mwhahaha oh look there it is
15:56 mwhahaha damn monkeys
15:56 EmilienM I'm wondering if doing mathematics operations in manifests work
15:56 EmilienM like $workers = ($::processorcount / 4)
15:56 mwhahaha EmilienM: it does
15:56 EmilienM in puppet3?
15:56 mwhahaha yea
15:56 EmilienM pradk: ^ no need to write a fact
15:56 mwhahaha (($::processorcount + 0) / 4
15:56 EmilienM cool
15:57 mwhahaha +0 to force int conversion i think
15:57 mjblack iurygregory: The issue is that when we set memcache servers and python-memcache is not installed, most components fail
15:57 pradk cool i'll try that and see
15:57 pradk ah + 0 thats probably what i was missing
15:58 mwhahaha https://github.com/openstack/fuel-library​/blob/master/deployment/puppet/openstack_​tasks/manifests/roles/cinder.pp#L84-L85
15:58 iurygregory mjblack, yeah i know
15:58 pradk mwhahaha, perfect
15:58 EmilienM mwhahaha: nice
15:58 pradk thx mwhahaha and EmilienM
15:58 mjblack EmilienM: updated https://review.openstack.org/#/c/367438/
15:59 mwhahaha there's a larger issue with using memcache with the authtoken stuff
15:59 mjblack mwhahaha: what larger issue?
15:59 mwhahaha python-memcache is terrible and as such you should only really configure a single host
16:00 mwhahaha i still don't think we should be forcing python-memcache on people so i'd rather document the fact you need it rather than enable it's installation by default
16:00 mwhahaha like a giant note on the memcache server option
16:00 mjblack mwhahaha: in the review its set to false
16:01 mwhahaha yea but there was chat of setting it true somewhere :D
16:01 mjblack yeah thats why it was set to false :D
16:02 mjblack but I'll have to go through and update the other modules to make sure it can be enabled
16:02 mwhahaha no my point is that it shouldn't be enabled by us
16:05 EmilienM mjblack: looking
16:09 EmilienM mjblack: minor comment
16:09 EmilienM mjblack: we're really close
16:09 dfisher mjblack: looking
16:10 iurygregory EmilienM, =O
16:11 EmilienM iurygregory: ;-)
16:15 iurygregory EmilienM, you have done an awesome job as PTL! Thank you very much =D
16:15 mjblack EmilienM: for the docstring comment you just want me to do "= true"?
16:16 EmilienM iurygregory: thanks, it is a great teamwork
16:16 EmilienM mjblack: no, in the doc for the parameter
16:16 EmilienM mjblack: explain that not using true on ubuntu
16:16 EmilienM an dwhy
16:17 EmilienM so people don't fall in the warning
16:17 mjblack ok
16:17 EmilienM thanks
16:17 degorenko EmilienM, dude, i want to wish you good luck on a new project! You will do your best :)
16:18 EmilienM degorenko: there is no new project
16:18 EmilienM I'm staying here lol
16:19 EmilienM you fire me? :P
16:19 iurygregory yeah
16:19 bkero Just done herding cats :P
16:19 iurygregory hahaa
16:19 degorenko EmilienM, ah, i thought you are moving to something else :D
16:19 EmilienM degorenko: not at all
16:19 EmilienM I'll keep make you angry with my -1 :P
16:19 iurygregory EmilienM, is multi-task (TripleO and Puppet) haha
16:19 degorenko ah, you can try :P
16:21 openstackgerrit Matthew J Black proposed openstack/puppet-gnocchi: Add python-rados package in ceph backend class.  https://review.openstack.org/367438
16:21 EmilienM iurygregory: so are most of our contributors, look at degorenko, iberezovskiy|afk, mwhahaha: they also work on installers
16:22 iurygregory EmilienM, yeah i'm glad we have awesome people to be PTL =D
16:23 iurygregory EmilienM, we need to drink in Barcelona o/
16:23 degorenko EmilienM, are you playing in ping pong?
16:23 EmilienM degorenko: yes and I don't like to loose lol
16:24 degorenko EmilienM, hmm, we need to play few games :P
16:25 dmsimard mwhahaha: pretty sure python-memcache is tied to dogpile.cache.memcached. Would you rather default to a different backend ?
16:25 dmsimard See backends vs libs: http://docs.openstack.org/mitaka/con​fig-reference/identity/caching.html
16:25 dmsimard (scroll to the end)
16:25 mwhahaha the problem i have with python-memcache is that it uses memcache like a pooled backend
16:25 mwhahaha which is not how memcache should be used
16:26 mwhahaha it doesn't handle memcache being down well
16:27 mwhahaha the original intent of memcache was to have a ton of them and if they failed they failed. unfortunately the python implementation does not play nicely with this mentaility
16:28 mwhahaha so in the case of keystone authtoken caching you really want to setup a single instance on localhost because if you say configure it with your controllers and you loose one controller your performance tanks
16:28 mwhahaha it's a peeve of mine, that's all.
16:29 EmilienM mjblack: sorry wrong window
16:31 mjblack EmilienM: yes to what?
16:31 EmilienM mjblack: I was replying to an old question on telemetry channel, nevermind :-)
16:32 mjblack oh ok, thought I missed something
16:32 EmilienM mjblack: /me reviewing again
16:32 * mjblack rubs eyes
16:32 EmilienM mjblack: +2 this time
16:33 EmilienM mjblack: and thanks
16:41 openstackgerrit Pradeep Kilambi proposed openstack/puppet-gnocchi: Calculate workers based on processor count for better performance  https://review.openstack.org/368124
16:43 openstackgerrit David Moreau Simard proposed openstack/puppet-magnum: Add SSL support for magnum::api  https://review.openstack.org/366992
16:51 xarses joined #puppet-openstack
16:56 jperry joined #puppet-openstack
16:59 Topic for #puppet-openstack is now CI status: green - Puppet OpenStack modules http://docs.openstack.org/deve​loper/puppet-openstack-guide/
16:59 openstackstatus NOTICE: setuptools 27.1.2 addresses the circular import
17:05 markd_ joined #puppet-openstack
17:09 vinsh joined #puppet-openstack
17:09 mfisch I think there's a bug or bad assumption in puppet-keystone
17:09 mfisch EmilienM: ^
17:09 mfisch the bootstrap command is resetting the admin password
17:09 iurygregory lol
17:09 mfisch when admin_password != admin_token
17:10 iurygregory we had a problem when admin_password was not the same value of admin_token i think..
17:10 EmilienM dang
17:10 mfisch when its not the same you have a major issue
17:10 iurygregory the fast way to fix was let equal if i remember...
17:10 mfisch the admin user is broken until the next puppet run when puppet sets it back
17:11 mfisch We use a different admin token per box for security reasons
17:11 mfisch whenever bootstrap runs admin cannot auth until the next puppet run
17:11 mfisch 2016-09-09 17:06:58.062 1 INFO keystone.cmd.cli [req-54345e3e-aabd-4d84-aa25-dec0869e10e8 - - - - -] Reset password for user admin.
17:11 mfisch #disapproval eyes at keystone
17:11 EmilienM the the two params are in another class, right?
17:12 EmilienM one is in keystone::roles::admin and other is in keystone::init ?
17:12 mfisch right
17:12 mfisch but keystone-manage wants this
17:12 mfisch --bootstrap-password OS_BOOTSTRAP_PASSWORD
17:12 mfisch The bootstrap user password
17:12 mfisch not the admin token
17:12 mfisch the password for admin
17:12 mfisch I believe the puppet code is wrong
17:13 mwhahaha nevar!
17:13 mfisch lol
17:13 stevemar joined #puppet-openstack
17:13 mfisch hey steve
17:13 mwhahaha sounds like a version mismatch
17:13 stevemar mfisch: yo
17:14 mwhahaha it used to be the token i think but then they changed it to the password
17:14 mfisch stevemar: puppet-keystone is calling keystone-manage bootstrap --boostrap-password $ADMIN_TOKEN
17:14 mfisch and I think it should be admin password
17:14 mfisch mwhahaha: its password in M
17:14 mfisch mwhahaha: when did it change?
17:14 stevemar mfisch: right, --bootstrap-password will set the admin user's password to that value
17:15 mfisch ugh yeah thats the issue
17:15 stevemar mfisch: it's umm, always been that way
17:15 mfisch stevemar: since L? or forever
17:15 mfisch ok
17:15 stevemar mfisch: the whole point of bootstrap was to encourage folks to never have to use ADMIN_TOKEN ever again :)
17:16 mfisch for our install admin_token != admin_password so there's a bit of a fight
17:16 iurygregory admin_token is deprecated?
17:16 mfisch which means admin can't login everytime puppet runs bootstrap
17:16 mfisch iurygregory: yeah, its bad security mojo
17:16 mwhahaha oh i guess it's a usage thing
17:16 stevemar iurygregory: deprecated as of L i believe
17:16 stevemar no wait, M
17:16 iurygregory oh ok
17:16 mfisch if you keep using it you get a stern talking to from ayoung
17:17 stevemar mfisch: why is that an issue if the two are not equal?
17:17 mfisch stevemar: puppet runs bootstrap under certain conditions and sets the admin user's password to FOO
17:17 mfisch but its really BAR
17:17 mwhahaha so keystone doesn't have a user password, so i guess shouldn't you make admin_token = admin_password?
17:17 mwhahaha er keystone class
17:17 mfisch stevemar: so admin user cannot auth
17:19 stevemar mfisch: so under this condition, it runs bootstrap and sets it to some known value, instead of a user selected one?
17:19 stevemar mfisch: this is weird yo
17:19 mfisch stevemar: --bootstrap_password ${admin_password} is right, not ${admin_token}
17:19 mfisch thats what I need to fix
17:19 stevemar ++
17:20 mwhahaha probably needs a documentation fix
17:20 EmilienM so we need to rename the parameter?
17:20 mfisch EmilienM: We'll need to use ${keystone::roles::admin::password}
17:20 mfisch wonder if that class is included by default
17:20 mwhahaha no shouldn't you set those to be equal
17:21 EmilienM no
17:21 mfisch mwhahaha: you're missing thepoint
17:21 EmilienM but you can do the other way around maybe?
17:21 mfisch passing the admin_token to bootstrap is fundamentally wrong
17:21 mfisch EmilienM: how?
17:21 mwhahaha admin_token is the parameter name
17:21 mwhahaha it doesn't have to be the admin_token
17:21 mfisch mwhahaha: and thats the bug
17:21 mwhahaha because there are not admin tokens anymore
17:21 mfisch there still are
17:21 mfisch just deprecated
17:22 iurygregory we forgot to remove '-'
17:22 mwhahaha right but we don't use them anymore
17:22 mfisch we as in puppet?
17:22 mwhahaha yea
17:22 mwhahaha i thought
17:22 mfisch we have a keystone::disable_admin_token_auth class but its not enabled by default
17:23 mwhahaha so i'm not saying we shouldn't fix the class, just trying to figure out what exactly the change is :D
17:23 iurygregory omg we have many examples saying that we should have a value for admin_token haha
17:23 mfisch trying to think if we still use the admin token anywhere else
17:24 mwhahaha i guess the fix would be add $admin_password, set it to $admin_token if not provided and throw a deprecation warning for admin_token?
17:25 mfisch mwhahaha: you'd want $admin_password to match whats in roles::admin::password though so thats another warning check
17:25 mfisch actually make that an error
17:26 mwhahaha but that assumes you use role::admin
17:26 mwhahaha what if you don't
17:26 mfisch yeah
17:26 mwhahaha the other way might be better
17:26 mfisch the class may not be included or evaluated
17:26 mwhahaha where role::admin::password = $::keystone::admin_password by default
17:27 mfisch mwhahaha: that wfm
17:29 mfisch mwhahaha: you want me to do this one?
17:29 mwhahaha sure
17:37 mfisch this is not going to be an easy transition
17:37 mfisch mwhahaha: if we make admin_password required we should default it to something
17:38 mwhahaha thats why i said default to $admin_token
17:38 mwhahaha that's backwards compatible
17:38 mwhahaha and throw a warning
17:38 mfisch +1
17:39 mfisch also maybe a bug to remove admin_token from tests and usage
17:39 mwhahaha yes
17:39 mfisch do you mind filing
17:40 vinsh All because /someone/ had to get our puppet-keystone on master so /someone/ could work on ironic.
17:40 mfisch thanks bro  ^
17:40 mfisch ;)
17:41 vinsh ^_^
17:41 EmilienM vinsh: hey dude
17:41 mwhahaha always with the demands
17:41 vinsh EmilienM: Hi
17:41 mfisch stupid ironic
17:41 EmilienM vinsh: when do you rewrite puppet-swift? :P
17:42 mfisch after his hangover goes away #Broncos
17:42 vinsh ah yup!
17:42 mwhahaha how about them field goals
17:42 vinsh EmilienM: I have some time starting in 2 weeks where I will put a big push into completing some puppet-swift patches.
17:44 EmilienM vinsh: nice :D
17:44 EmilienM bkero: http://logs.openstack.org/66/367966/2/ch​eck/gate-puppet-openstack-integration-4-​scenario001-tempest-centos-7-nv/14e3a51/​console.html#_2016-09-09_16_18_53_768780
17:44 vinsh Next two weeks is a patch to add extrenal hooks to puppet-ironic
17:44 EmilienM I still see it, maybe I should one more time?
17:45 vinsh since we dockerized that here.
17:50 bkero oof, connection refused to logs.o.o now
17:50 * bkero tries again in a sec
17:58 mfisch mwhahaha: if you have a bug plz let me know
18:00 openstackgerrit Matt Fischer proposed openstack/puppet-keystone: Add admin_password parameter for use in bootstrap  https://review.openstack.org/368169
18:00 mfisch I can't get bundle to run anymore
18:00 bkero EmilienM: Yep. It's because the GPG key hasn't been installed yet. It's part of that RPM. Look at the next 2 lines.
18:00 mfisch mfischer@Matts-MacBook-Pro-4:~/code/puppet/puppet-keystone (master)$ bundle exec rake spec
18:00 mfisch /usr/local/lib/ruby/gems/2.1.0/gems/bundler-1.1​0.6/lib/bundler/rubygems_integration.rb:292:in `block in replace_gem': rake is not part of the bundle. Add it to Gemfile. (Gem::LoadError)
18:00 mfisch from /usr/local/bin/rake:22:in `<main>'
18:00 mfisch anyone seen that ^
18:02 mwhahaha mfisch: remove Gemfile.lock and rerun bundle install
18:02 mwhahaha also i'll file the bug in a bit
18:04 EmilienM bkero: we'll see how recheck works
18:04 bkero EmilienM: you think it failed because of that warning?
18:05 bkero EmilienM: that warning was there before yesterday too
18:05 EmilienM bkero: mhh ok
18:08 mwhahaha mfisch: https://bugs.launchpad.net/p​uppet-keystone/+bug/1621959
18:08 openstack Launchpad bug 1621959 in puppet-keystone "keystone incorrectly uses admin_token for the bootstrap password" [Undecided,New]
18:08 openstackgerrit Matt Fischer proposed openstack/puppet-keystone: Add admin_password parameter for use in bootstrap  https://review.openstack.org/368169
18:15 openstackgerrit Matt Fischer proposed openstack/puppet-cinder: Cinder hooks support  https://review.openstack.org/344464
18:16 EmilienM mfisch: I don't understand
18:16 EmilienM admin_token is deprecated but required?
18:22 iurygregory what?!
18:22 * iurygregory looking at the patch
18:23 EmilienM I commented
18:23 iurygregory maybe set both to $::os_service_default/undef and validate ?
18:30 mfisch its going to be a massive change to get rid of admin_token
18:30 mfisch all the tests
18:30 mfisch thats why we split this into 2 issues
18:30 mfisch https://bugs.launchpad.net/p​uppet-keystone/+bug/1621960
18:30 openstack Launchpad bug 1621960 in puppet-keystone "deprecate all usage of admin_token" [Undecided,New]
18:30 iurygregory mfisch, need help?
18:30 mfisch is the 2nd issue
18:30 mfisch I'm not signing up to deprecate admin_token
18:30 mfisch I'm signing up to fix broken calls to keystone-manage bootstrap
18:32 mfisch EmilienM: undef'ing that requires changing all the tests
18:32 mfisch you'll have an ordering issue
18:35 mfisch replied
18:44 dfisher linux-y question for you folks … if you issue a nova service-disable <host> <service> is that supposed to stop the service from running within the OS or the 'service' referenced in 'service-disable' an OpenStack construct?
18:46 mwhahaha openstack construct would be my assumption
18:46 mwhahaha no idea wtf that is
18:46 mwhahaha i thought that just marked a nova service as down
18:46 dfisher but it doesn't run something like 'service nova-whatever stop' on ubuntu, right?
18:47 mwhahaha no
18:47 dfisher k
18:47 dfisher thanks
18:47 openstackgerrit Pradeep Kilambi proposed openstack/puppet-gnocchi: Tune workers and thread defaults for better performance  https://review.openstack.org/368124
18:52 openstackgerrit Merged openstack/puppet-keystone: Allow to manage credential files contents  https://review.openstack.org/366394
18:54 openstackgerrit Merged openstack/puppet-openstack-integration: Update RDO repository to latest trunk  https://review.openstack.org/368037
18:56 EmilienM after 2432 rechecks
18:56 EmilienM it merged :)
19:07 mwhahaha finally
19:10 iurygregory recheck is our best friend
19:18 openstackgerrit David Moreau Simard proposed openstack/puppet-openstack-integration: Add magnum support to scenario002  https://review.openstack.org/367012
19:19 openstackgerrit David Moreau Simard proposed openstack/puppet-magnum: Add SSL support for magnum::api  https://review.openstack.org/366992
19:26 openstackgerrit David Moreau Simard proposed openstack/puppet-openstack-integration: Add magnum testing support to scenario002  https://review.openstack.org/367012
19:29 xb_ joined #puppet-openstack
19:41 fultonj joined #puppet-openstack
19:57 iurygregory EmilienM, your patch is all green https://review.openstack.org/#/c/367966
20:09 colonwq EmilienM, 289027 has all of its depends-on merge and is passing all green. Can we get this approved for merging?
20:10 openstackgerrit Merged openstack/puppet-tripleo: Added Contrail Composable Services  https://review.openstack.org/348905
20:32 xarses_ joined #puppet-openstack
20:42 openstackgerrit Matthew Treinish proposed openstack/puppet-openstack-integration: Manually run test runner for tempest  https://review.openstack.org/349203
20:48 _stowa joined #puppet-openstack
20:54 dmsimard doh
20:54 dmsimard there's no puppet-nova providers for keypairs
20:55 dmsimard RABBIT HOLE AGAIN
20:56 _ody While you're there, a type and provider for instance would be good too.
20:56 _ody ;)
20:57 dmsimard no thanks
20:57 dmsimard ugh, this is hard
20:59 mwhahaha life's hard and then you get the sweet release of death
21:00 openstackstatus joined #puppet-openstack
21:00 mwhahaha but only after writing some puppet
21:00 dmsimard mwhahaha: the magnum rabbit hole goes deeper
21:00 mwhahaha :D
21:00 dmsimard mwhahaha: need to provision a fedora atomic image (okay, easy) but I also need to provision a keypair
21:00 dmsimard in itself it's stupid easy
21:01 dmsimard like, I could hack an exec and be done with it
21:01 dmsimard and it's probably what I'm going to do
21:02 mwhahaha noooooes (ok it's friday)
21:03 dmsimard mwhahaha: well, https://github.com/openstack/puppet-​cinder/blob/9ef8773f8b2326ba20d7b5f2​2686da6e01278b61/manifests/type.pp was a thing until already not too long ago
21:03 dmsimard I can't write ruby to save me life
21:04 mwhahaha it's not so hard
21:05 mwhahaha you could probably use nova_flavor as an example
21:05 bkero dmsimard: types and providers? is copypasta
21:06 dmsimard okay, I'll try :(
21:06 * dmsimard goes deeper in the rabbit hole
21:06 mwhahaha hello and welcome back to puppet
21:07 bkero This is called 'shaving the yak'.
21:12 haplo37__ joined #puppet-openstack
21:25 dmsimard bkero: we have no yaks in canada
21:25 dmsimard only moose
21:25 dmsimard and geese
21:25 dmsimard and the occasional beaver
21:26 bkero heh
21:26 bkero This is what Canadians must call 'shaving the goose'
21:30 dfisher honking*
21:31 xarses joined #puppet-openstack
21:32 xarses joined #puppet-openstack
21:43 openstackgerrit Matthew Treinish proposed openstack/puppet-openstack-integration: Manually run test runner for tempest  https://review.openstack.org/349203
21:43 mtreinish dmsimard, EmilienM: ^^^ I finally think that is ready
21:46 asilenkov joined #puppet-openstack
21:52 mwhahaha joined #puppet-openstack
22:18 dmsimard mwhahaha: yeah, basing it off of nova_flavor isn't helpful
22:18 dmsimard mwhahaha: because it doesn't really make sense to create a keypair as the nova user :)
22:19 dmsimard I checked some other providers like glance_image or neutron_network ... glance_image doesn't support creating as another tenant and neutron_network has a "tenant" parameter which is exposed by the neutron network create command
22:19 dmsimard There's no parameter to pass to "openstack keypair create" that'd end up creating the keypair in another tenant and anyway the provider will only authenticate as nova
22:19 mwhahaha that's not true
22:20 dmsimard what am I missing ?
22:20 mwhahaha well it won't be authed as nova :D
22:20 mwhahaha you could add credentials as an option
22:20 mwhahaha by default we use whatever is in the openrc or the environment
22:20 mwhahaha i think the environment overrides the openrc
22:20 mwhahaha but anyway
22:21 dmsimard ah, so, the fetching from nova.conf is the last resort ?
22:21 mwhahaha i don't even think we do that
22:21 dmsimard it seems like it's expecting nova_config resources for the different credentials
22:21 dmsimard https://github.com/openstack/puppet-nova/blo​b/master/lib/puppet/type/nova_flavor.rb#L53
22:21 mwhahaha you might be looking in some older stuff
22:22 mwhahaha that's for something else
22:23 mwhahaha i think
22:23 mwhahaha https://github.com/openstack/puppet-nova/blob/​master/lib/puppet/type/nova_flavor.rb#L53-L60
22:23 mwhahaha that whole block goes together
22:23 mwhahaha basically that's for ensuring that stuff is done before the nova_flavor can be run
22:23 mwhahaha you want the CredentialsV3 stuff
22:23 mwhahaha https://github.com/openstack/puppe​t-nova/blob/master/lib/puppet/prov​ider/nova_flavor/openstack.rb#L11
22:23 mwhahaha thats where the credential fun is handled
22:23 mwhahaha and that's over in openstacklib
22:24 dmsimard bah, I'll look monday at this point the food and the alcohol are calling
22:24 mwhahaha yup
22:24 mwhahaha later :D
22:24 dmsimard have a nice weekend
22:56 aimon joined #puppet-openstack
23:36 openstackgerrit Alex Schultz proposed openstack/puppet-manila: [DONOTMERGE] ci test  https://review.openstack.org/368260

| Channels | #puppet-openstack index | Today | | Search | Google Search | Plain-Text | summary