Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2014-09-04

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

All times shown according to UTC.

Time Nick Message
00:00 rstarmer joined #puppet-openstack
00:09 imcsk8_ joined #puppet-openstack
00:09 gaudenz_ joined #puppet-openstack
00:11 mfa298___ joined #puppet-openstack
00:12 bogdando joined #puppet-openstack
00:23 rmoe joined #puppet-openstack
00:28 crinkle mmaglana: is there a way to pass a token directly as an X-Auth-Token header in aviator?
00:29 crinkle giving it a token_id in auth_credentials seems to pass it in the request body instead of a header
00:36 mmaglana crinkle: are you trying to use the service token?
00:37 crinkle mmaglana: I think so
00:38 mmaglana i think that should work. if you try to authenticate with a token, keystone will respond with another token which will then get used by aviator in the request headers for subsequent requests.
00:38 mmaglana that behavior conforms with http://developer.openstack.org/api-r​ef-identity-v2.html#identity-auth-v2
00:40 mmaglana crinkle: btw are you authenticating against keystone v2 or v3?
00:40 crinkle mmaglana: v2
00:41 thumpba joined #puppet-openstack
00:41 crinkle mmaglana: hmm
00:41 mmaglana ok. you should be able to authenticate successfully. did you come across any problems?
00:42 rstarmer joined #puppet-openstack
00:43 mmaglana crinkle: here's the test for token-based auth https://github.com/aviator/aviator/blob/m​aster/test/aviator/openstack/identity/req​uests/v2/public/create_token_test.rb#L142
00:43 mmaglana oh wait...
00:44 mmaglana i just looked at the test name and said 'oh there it is!' duh.
00:44 mmaglana oh wait (again)
00:44 crinkle mmaglana: I'm able to auth with a token I create via passwordCredentials using curl, but not with the service token
00:45 mmaglana crinkle: what does the log say?
00:45 crinkle mmaglana: it says 401, and that it tried to pass the token in the request body
00:45 crinkle whereas I am able to use the service token to auth if I pass it in X-Auth-Token using curl
00:46 mmaglana can you share a gist of your code?
00:46 crinkle sure, sec
00:47 mmaglana crinkle: btw, for that link i shared above, lines 156 to 159 is how you should auth by token. that should actually put the token in the body.
00:47 mmaglana here's the associated cassette of that test case: https://github.com/aviator/aviator/blob/mast​er/test/cassettes/openstack/identity/request​s/v2/public/create_token/leads_to_a_valid_re​sponse_when_provided_with_a_token.yml#L43
00:50 gildub joined #puppet-openstack
00:53 cjellick joined #puppet-openstack
00:54 crinkle mmaglana: here is the gist of what I'm trying to do and what I'm seeing
00:54 crinkle I will take a look at those test cases, that will probably help
00:54 crinkle oops link is http://paste.openstack.org/show/105511/
00:54 mmaglana :)
00:58 mmaglana crinkle: i'm not 100% yet but it's likely that service tokens must be used directly instead of as an intermediary token to get an actual token. if that is true, then aviator is not yet capable of doing that.
00:58 crinkle okay
00:58 crinkle experimentin with curl tells me that's probably the case
00:58 mmaglana crinkle: but if needed, it's as simple as declaring a new parameter 'system_token' and if that's present, to put that directly in the header request.
00:59 mmaglana of course 'simple' is a deceptive word :)
00:59 crinkle haha
00:59 mmaglana crinkle: because if you want to supply a system token for direct use, you would also have to have some way to get the service catalog.
00:59 mmaglana and i don't remember if there's an API endpoint for that.
01:00 mmaglana not to say it's an impossible change. there are just a couple of unknowns for now.
01:00 crinkle okay, I'll work around it for now and maybe dig into it later
01:01 crinkle thanks!
01:01 mmaglana have to log off for now. feel free to file a bug on github.
01:01 mmaglana np
01:04 bogdando joined #puppet-openstack
01:24 JayJ joined #puppet-openstack
01:41 thumpba joined #puppet-openstack
01:46 rstarmer joined #puppet-openstack
01:50 JayJ joined #puppet-openstack
02:07 aimon_ joined #puppet-openstack
02:46 ianw joined #puppet-openstack
02:50 xarses joined #puppet-openstack
03:10 mmaglana joined #puppet-openstack
03:17 openstackgerrit A change was merged to stackforge/puppet-cinder: Allow configuration of Cinder API workers  https://review.openstack.org/115745
03:20 JayJ joined #puppet-openstack
03:45 mmaglana crinkle: I replied to the bug you filed on aviator.
04:20 JayJ joined #puppet-openstack
05:04 saju_m joined #puppet-openstack
05:19 photocyte joined #puppet-openstack
05:21 JayJ joined #puppet-openstack
05:26 gcha joined #puppet-openstack
05:49 walkleyn joined #puppet-openstack
06:00 JayJ joined #puppet-openstack
06:20 saju_m joined #puppet-openstack
07:00 sanjayu joined #puppet-openstack
07:01 JayJ joined #puppet-openstack
07:11 saju_m joined #puppet-openstack
07:25 saju_m joined #puppet-openstack
07:31 dgurtner joined #puppet-openstack
07:32 mmagr joined #puppet-openstack
07:36 fvollero joined #puppet-openstack
07:44 openstackgerrit Risto Laurikainen proposed a change to stackforge/puppet-glance: Make user creation optional when creating service.  https://review.openstack.org/116263
07:44 openstackgerrit Risto Laurikainen proposed a change to stackforge/puppet-cinder: Make user creation optional when creating service.  https://review.openstack.org/116264
07:44 openstackgerrit Risto Laurikainen proposed a change to stackforge/puppet-nova: Make user creation optional when creating service.  https://review.openstack.org/116262
07:44 openstackgerrit Risto Laurikainen proposed a change to stackforge/puppet-neutron: Make user creation optional when creating service.  https://review.openstack.org/116265
07:53 tfz joined #puppet-openstack
08:02 JayJ joined #puppet-openstack
08:04 qba73 joined #puppet-openstack
08:15 derekh joined #puppet-openstack
08:20 mmaglana joined #puppet-openstack
08:22 mmaglana_ joined #puppet-openstack
08:38 gildub joined #puppet-openstack
08:38 social sbadia: could you please look at https://review.openstack.org/#/c/110913/ ?
09:02 JayJ joined #puppet-openstack
09:08 openstackgerrit A change was merged to stackforge/puppet-neutron: Fix the subnet dhcp enable/disable param  https://review.openstack.org/116212
09:23 mmaglana joined #puppet-openstack
09:32 openstackgerrit Lukas Bezdicka proposed a change to stackforge/puppet-neutron: Unescape value in parse_allocation_pool  https://review.openstack.org/110913
09:41 saju_m joined #puppet-openstack
09:43 saju_m joined #puppet-openstack
09:49 dgurtner joined #puppet-openstack
10:03 JayJ joined #puppet-openstack
10:24 mmaglana joined #puppet-openstack
10:29 dgurtner joined #puppet-openstack
10:45 ddmitriev joined #puppet-openstack
11:01 JayJ joined #puppet-openstack
11:12 JayJ_ joined #puppet-openstack
11:19 saju_m joined #puppet-openstack
11:21 mmaglana joined #puppet-openstack
11:40 dmsimard EmilienM: Good morning :D
11:49 parstac_pete_ joined #puppet-openstack
11:49 f10d4_ joined #puppet-openstack
11:49 dgurtner joined #puppet-openstack
11:50 TorLaedre1 joined #puppet-openstack
11:51 rlaurika_ joined #puppet-openstack
11:51 pabelanger_ joined #puppet-openstack
11:51 beddari joined #puppet-openstack
11:51 underyx joined #puppet-openstack
11:52 openstackgerrit David Moreau Simard proposed a change to stackforge/puppet-nova: Don't define virsh secret if already defined  https://review.openstack.org/119041
11:55 openstackgerrit David Moreau Simard proposed a change to stackforge/puppet-nova: virsh returns a list of secret uuids, not keyring names  https://review.openstack.org/119043
11:59 openstackgerrit David Moreau Simard proposed a change to stackforge/puppet-nova: Don't define virsh secret if already defined  https://review.openstack.org/119044
11:59 openstackgerrit David Moreau Simard proposed a change to stackforge/puppet-nova: virsh returns a list of secret uuids, not keyring names  https://review.openstack.org/119045
12:21 mmaglana joined #puppet-openstack
12:27 openstackgerrit Martin Mágr proposed a change to stackforge/puppet-heat: Implement Keystone domain creation  https://review.openstack.org/103127
12:32 morazi joined #puppet-openstack
12:48 dprince joined #puppet-openstack
12:54 dprince joined #puppet-openstack
12:58 JayJ_ joined #puppet-openstack
13:08 EmilienM good morning
13:21 mmaglana joined #puppet-openstack
13:23 mattymo EmilienM, did you relocate to canada?
13:23 EmilienM mattymo: yes
13:23 mattymo congrats man
13:23 EmilienM thanks!
13:40 francois joined #puppet-openstack
13:44 JayJ_ joined #puppet-openstack
13:59 xingchao joined #puppet-openstack
13:59 openstackgerrit Martin Mágr proposed a change to stackforge/puppet-heat: Implement Keystone domain creation  https://review.openstack.org/103127
14:06 daneyon_ joined #puppet-openstack
14:10 xavpaice_ joined #puppet-openstack
14:14 underyx_ joined #puppet-openstack
14:14 cherriges joined #puppet-openstack
14:17 odyi joined #puppet-openstack
14:21 mmaglana joined #puppet-openstack
14:25 francois joined #puppet-openstack
14:33 daneyon joined #puppet-openstack
14:33 xavpaice joined #puppet-openstack
14:46 JayJ_ joined #puppet-openstack
14:50 xingchao joined #puppet-openstack
14:50 s_albtraum joined #puppet-openstack
14:53 qba73_ joined #puppet-openstack
14:54 dmsimard EmilienM: I had a question for nova::compute::rbd
14:54 EmilienM shoot
14:54 dmsimard EmilienM: It is my understanding that it is mean to be used with libvirt for volumes ? If I use Cinder with that nova blows up when trying to boot an instance.
14:54 dmsimard The virsh secret thing does what it's supposed to do and that's fine but there are no nova.conf configurations required to boot from a cinder volume ?
14:55 EmilienM dmsimard: there isn't
14:55 EmilienM dmsimard: nova.conf is only if you need RBD support in Nova
14:55 EmilienM boot from volume afik does not need rbd config in nova.conf
14:55 dmsimard So should we have a switch like use_libvirt or use_cinder that makes it so the nova_config are applied or not ?
14:56 dmsimard EmilienM: I confirm that in my experience only the virsh secret needs to be set when booting with cinder-volume..
14:57 EmilienM dmsimard: right, this is a need (virsh secret)
14:57 dmsimard Also welcome to Canada
14:57 dmsimard :p
14:57 EmilienM dmsimard: thanks!
14:58 dmsimard IMO I would tend to add a use_libvirt parameter, defaulting to true for backwards compatibility in which if it is true, it does the nova_configs
14:59 dmsimard But I'm asking you since you're the one that initially came up with the libvirt use case
14:59 EmilienM dmsimard: it sounds like a good idea
14:59 dmsimard ok, will submit something. Thanks.
15:00 qba73 joined #puppet-openstack
15:04 JayJ_ joined #puppet-openstack
15:12 cjellick joined #puppet-openstack
15:24 mdorman joined #puppet-openstack
15:25 mdorman could i please get a review on https://review.openstack.org/#/c/118739/ ?  thanks.
15:33 EmilienM michchap_: can I go ahead and merge https://review.openstack.org/#/c/114155/ ?
15:33 openstackgerrit David Moreau Simard proposed a change to stackforge/puppet-nova: Add parameter to set virsh secret but not nova libvirt config  https://review.openstack.org/119093
15:33 dmsimard EmilienM ^
15:33 michchap_ EmilienM: sure. The implementation is also in the queue
15:33 EmilienM dmsimard: will look asap.
15:34 EmilienM michchap_: lol, making the code before blueprint :P
15:34 michchap_ EmilienM: I prefer to test the ideas before suggesting them.
15:34 michchap_ EmilienM: but yeah it's a bit backwards :)
15:34 EmilienM michchap_: hehe
15:35 openstackgerrit A change was merged to stackforge/puppet-openstack-specs: Add hash based repo management in extras  https://review.openstack.org/114155
15:36 EmilienM mgagne: https://review.openstack.org/#/c/99524/ merged too.
15:36 mgagne wonderful
15:36 openstackgerrit A change was merged to stackforge/puppet-openstack-specs: Common OpenStack Configuration Provider  https://review.openstack.org/99524
15:36 dmsimard mdorman: Interesting, I didn't know neutron and nova also had that bug
15:37 dmsimard mdorman: There's https://bugs.launchpad.net/p​uppet-openstack/+bug/1356083 for neutron and https://bugs.launchpad.net/pu​ppet-ceilometer/+bug/1348284 for ceilometer
15:37 dmsimard mdorman: perhaps we should link all those bugs together or something
15:37 EmilienM spredzy_: any update on https://review.openstack.org/#/c/117880/ ?
15:38 EmilienM dmsimard: looking your patch now.
15:38 mgagne pick one bug, mark all projects as affected and mark the others as duplicate of the choosen bug
15:39 EmilienM dmsimard: ok I have comments maybe, sorry to have miss that when you asked me
15:40 dmsimard mgagne: Will do.
15:41 EmilienM dmsimard: when usinb boot from volume or wanting to attach a RBD volume to a VM, you need libvirt/rbd_user in nova.conf
15:41 spredzy_ EmilienM, still testing it in our deployment, I ll update when I find the exact cause of why something needs to be done there, because independently of $manage_haproxy_package set to  true or false we face the issue
15:41 EmilienM spredzy_: great, just wanted to catch up with pending reviews.
15:42 mmaglana joined #puppet-openstack
15:42 dmsimard EmilienM: Ah, right, because it matches that rbd_user against the secret
15:42 dmsimard right ?
15:42 dmsimard So the virsh secret and rbd_user, that's it ?
15:42 EmilienM dmsimard: yeah
15:42 dmsimard I'll edit.
15:43 EmilienM dmsimard: that's the result of my investigations when we did our puppet module
15:44 EmilienM dmsimard: https://github.com/enovance/puppet-​openstack-cloud/blob/master/manifes​ts/compute/hypervisor.pp#L183-L198
15:44 EmilienM it's how we do now.
15:44 mdorman joined #puppet-openstack
15:44 dmsimard EmilienM: Ah so you don't use nova::compute::rbd if you use Cinder :)
15:45 EmilienM dmsimard: actually I don't use nova::compute::rbd correctly
15:45 EmilienM if you look at my manifest, I manage virsh secret without puppet-nova
15:45 EmilienM because I already managed it before puppet nova and I did not take time to catch up with this class.
15:45 mdorman dmsimard:   yeah, same problem on all of those.  i ddin’t bother creating a bug for puppet-nova, but i could if that’s needed.  i’d just like to get it merged asap so i can make some progress on my end… right now this is a blocker for me.
15:45 EmilienM which by the way I did lol
15:46 openstackgerrit A change was merged to stackforge/puppet-neutron: Unescape value in parse_allocation_pool  https://review.openstack.org/110913
15:46 dmsimard mdorman: I'm taking care of the bug part, I'll need you to amend the commit with a reference to the bug
15:46 dmsimard mdorman: Give you the link soon
15:47 mdorman sounds great, thanks.
15:47 openstackgerrit A change was merged to stackforge/puppet-glance: Add $package_ensure to glance::backend::rbd  https://review.openstack.org/118228
15:48 openstackgerrit A change was merged to stackforge/puppet-trove: Allow specifying Nova, Cinder and Swift endpoints  https://review.openstack.org/113147
15:51 dmsimard mdorman: https://bugs.launchpad.net​/puppet-nova/+bug/1356083
15:52 openstackgerrit Mike Dorman proposed a change to stackforge/puppet-nova: Makes kombu_ssl_* parameters optional when rabbit_use_ssl => true  https://review.openstack.org/118739
15:52 mdorman thanks.  ^^ is updated
15:53 dmsimard mdorman: :)
15:54 openstackgerrit David Moreau Simard proposed a change to stackforge/puppet-nova: Add parameter to set virsh secret but not nova libvirt config  https://review.openstack.org/119093
15:56 EmilienM dmsimard: in my code, I use the rbd_secret_uuid anyway
15:57 dmsimard EmilienM: Aye, it's used here too.
15:57 dmsimard EmilienM: Implicitly through the secret.xml template
15:57 dmsimard rbd_secret_uuid is set in the secret.xml template and then injected as a secret to virsh
15:58 EmilienM dmsimard: I think you can drop the L76
15:59 dmsimard EmilienM: Oh, I don't know *what* parameter breaks nova. But one of them does. I can check if the uuid parameter breaks anything.
15:59 dmsimard sec.
16:08 dmsimard EmilienM: Yeah, ok. As soon as libvirt image type is rbd it explodes
16:10 rmoe joined #puppet-openstack
16:11 dmsimard secret_uuid isn't necessary
16:11 dmsimard If you're just using cinder
16:11 dmsimard EmilienM: We can always add it to nova.conf but there's no value in doing it.
16:12 EmilienM dmsimard: I'm not sure, aere you?
16:12 dmsimard EmilienM: Just tested it
16:13 dmsimard EmilienM: rbd_user is necessary, not uuid
16:13 pabelanger joined #puppet-openstack
16:13 EmilienM dmsimard: great, so lgtm
16:13 EmilienM dmsimard: have you tried to attach a rbd volume to the instance though?
16:13 EmilienM it was my use case
16:13 dmsimard dmsimard: Using nova boot --boot-volume
16:14 dmsimard Want me to attach a second volume ?
16:14 EmilienM dmsimard: no I think that's good. But let's try it yeah. Just to be sure
16:17 hogepodge joined #puppet-openstack
16:21 JayJ_ joined #puppet-openstack
16:24 rstarmer joined #puppet-openstack
16:33 JayJ_ joined #puppet-openstack
16:40 pradk joined #puppet-openstack
16:45 JayJ_ joined #puppet-openstack
16:49 guimaluf joined #puppet-openstack
16:53 aimon joined #puppet-openstack
16:59 openstackgerrit Mike Dorman proposed a change to stackforge/puppet-heat: Makes kombu_ssl_* parameters optional when rabbit_use_ssl => true  https://review.openstack.org/119131
17:16 EmilienM dmsimard: hey, I'm currently investigate how to build CI for our OpenStack Puppet modules
17:17 EmilienM dmsimard: and I want to look at Puppet Ceph CI
17:17 EmilienM dmsimard: any pointer?
17:19 dachary EmilienM: https://pypi.python.org/pypi/gerritexec is all there is, almost ;-)
17:20 dachary and https://github.com/stackforge/puppet-ceph​/blob/master/README.md#integration-tests shows how we use it
17:21 dmsimard EmilienM: puppet-rspec-system (https://github.com/puppetlabs/rspec-system-puppet) is told to be deprected in favor of beaker (http://rubygems.org/gems/beaker)
17:21 EmilienM dachary: where do you have the VMs ?
17:21 dmsimard I believe Hunner is familiar with it
17:22 EmilienM dmsimard: ^^^^
17:23 sputnik13 joined #puppet-openstack
17:26 rstarmer joined #puppet-openstack
17:32 JayJ_ joined #puppet-openstack
17:35 Hunner Hi
17:36 Hunner dmsimard: I use beaker-rspec which works almost the same way as rspec-system (and rspec-system-puppet and rspec-system-serverspec)
17:36 Hunner Slightly different rspec helper syntax, but pretty close. And more flexible nodesets
17:39 sputnik13 joined #puppet-openstack
17:40 dmsimard EmilienM: dachary provides his own hypervisor somewhere
17:40 EmilienM dmsimard: dachary: I'm starting a new thread
17:40 dmsimard ?
17:40 JayJ_ joined #puppet-openstack
17:40 dmsimard On the ML ?
17:41 EmilienM dmsimard: of course
17:47 sputnik13 joined #puppet-openstack
17:50 sputnik13 joined #puppet-openstack
18:00 tfz joined #puppet-openstack
18:04 EmilienM dmsimard: sent
18:10 tfz joined #puppet-openstack
18:16 sputnik13 joined #puppet-openstack
18:24 xarses joined #puppet-openstack
18:26 JayJ_ joined #puppet-openstack
18:26 doo joined #puppet-openstack
18:31 thumpba joined #puppet-openstack
18:35 JayJ_ joined #puppet-openstack
18:43 mmagr joined #puppet-openstack
18:49 JayJ_ joined #puppet-openstack
19:04 thumpba joined #puppet-openstack
19:16 rstarmer joined #puppet-openstack
19:18 rstarmer joined #puppet-openstack
19:25 sputnik13 joined #puppet-openstack
19:26 rstarmer joined #puppet-openstack
19:32 cjellick joined #puppet-openstack
19:33 mdorman joined #puppet-openstack
19:36 doo joined #puppet-openstack
19:40 sputnik13 joined #puppet-openstack
19:44 sputnik13 joined #puppet-openstack
19:50 openstackgerrit Mike Dorman proposed a change to stackforge/puppet-ceilometer: Makes kombu_ssl_* parameters optional when rabbit_use_ssl => true  https://review.openstack.org/119187
19:56 mdorman i accidentally submitted the above review as a commit against the master branch, rather than a topic branch.   do i need to abandon and create a new one, or is there  a way to fix this?
20:02 sputnik13 joined #puppet-openstack
20:04 openstackgerrit Mike Dorman proposed a change to stackforge/puppet-ceilometer: Makes kombu_ssl_* parameters optional when rabbit_use_ssl => true  https://review.openstack.org/119190
20:17 openstackgerrit Mike Dorman proposed a change to stackforge/puppet-heat: Makes kombu_ssl_* parameters optional when rabbit_use_ssl => true  https://review.openstack.org/119131
20:19 photocyte joined #puppet-openstack
20:44 openstackgerrit Chris Hoge proposed a change to stackforge/puppet-ceilometer: Ensure nova-common is installed before ceilometer  https://review.openstack.org/119200
20:45 cjellick joined #puppet-openstack
20:46 imcsk8 joined #puppet-openstack
21:21 hogepodge EmilienM crinkle mgagne sbadia https://review.openstack.org/119200
21:21 hogepodge (for one of my last commits here at pl, fix this stupid bug that keeps biting me)
21:21 hogepodge (I also want to back port it)
21:22 nibalizer hogepodge: tomorrow your last day
21:22 nibalizer ?
21:22 hogepodge nibalizer today is
21:23 nibalizer wow
21:23 nibalizer well gl at your next gig!
21:42 hogepodge Thanks!
21:44 hogepodge nibalizer I’m sure we’re going to have lots of opportunities to work together.
21:45 hogepodge I’m going to be doing interop for the Foundation.
21:45 nibalizer ah sweet!
21:46 nibalizer i'm on the fence about wether or not i want to work on the puppet-openstack stuff in earnest
21:46 JayJ_ joined #puppet-openstack
21:46 nibalizer it kindof makes a lot of sense for me to do it
21:46 nibalizer but since i dont deploy openstack with them, it'd be very academic
21:56 hogepodge I think you should! Build a CI for us!
21:57 ianw joined #puppet-openstack
22:06 nibalizer i just want beaker to run on the openstack ci
22:06 nibalizer i think thats way better
22:57 thumpba joined #puppet-openstack
23:02 mdorman joined #puppet-openstack

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