| *** CaptTofu has joined #tripleo | 00:02 | |
| *** lblanchard has joined #tripleo | 00:02 | |
| *** vipul-away is now known as vipul | 00:02 | |
| dkehn | http://techcrunch.com/2014/01/21/github-president-becomes-ceo-president-becomes-ceo-in-executive-role-swap/?ncid=txtlnkusaolp00000591 | 00:06 |
|---|---|---|
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-incubator: Add openssh-server to install-dependencies https://review.openstack.org/68291 | 00:16 |
| *** morazi has quit IRC | 00:17 | |
| *** bauzas has quit IRC | 00:17 | |
| *** vipul is now known as vipul-away | 00:21 | |
| *** CaptTofu has quit IRC | 00:28 | |
| *** morazi has joined #tripleo | 00:28 | |
| *** CaptTofu has joined #tripleo | 00:28 | |
| *** morazi has quit IRC | 00:28 | |
| *** lblanchard has quit IRC | 00:40 | |
| *** lblanchard has joined #tripleo | 00:42 | |
| *** lblanchard has quit IRC | 00:53 | |
| jog0 | so I think nova may have a bug in its quota handling allowing an admin to set anyones quotas not just there own projects | 00:53 |
| *** CaptTofu has quit IRC | 00:54 | |
| lifeless | jog0: oh ? | 00:55 |
| lifeless | jog0: surely admins are meant to be able to set any quota ? | 00:56 |
| jog0 | define admin | 00:56 |
| jog0 | thats where it gets tricky nova's RBAC is poor | 00:56 |
| lifeless | in a keystone role called Admin by default | 00:56 |
| lifeless | because policy.json and OMG tenants are fail | 00:56 |
| lifeless | jog0: tell me about it. I have a file :) | 00:56 |
| jog0 | http://docs.openstack.org/api/openstack-compute/2/content/os-quota-sets.html | 00:58 |
| jog0 | in the last 3 methods, there is no validation on the tenant_id (not sure about user_id but I think its the same for that as well) | 00:58 |
| lifeless | jog0: so AIUI today, 'admin' is effectively global in most services | 00:59 |
| jog0 | so if you have access to those methods you can set any tenants quota | 00:59 |
| lifeless | jog0: I think it's a long road (worth doing) to change that | 00:59 |
| jog0 | lifeless: right, which is silly | 00:59 |
| jog0 | agreed | 00:59 |
| lifeless | jog0: but I don't think doing it piecemeal is a good idea | 00:59 |
| lifeless | jog0: better to be consistent today | 00:59 |
| lifeless | jog0: Can I grab your opinion on a related thing? | 01:00 |
| jog0 | lifeless: sure | 01:00 |
| lifeless | Say I have a network tripleo-bm-test, which is a provider network (and thus privileged) | 01:00 |
| jog0 | btw related bug https://bugs.launchpad.net/nova/+bug/1118066 | 01:00 |
| lifeless | would it be reasonable to have a tenant 'I own tripleo-bm-test' and then add some users to that tenant as well as their own... but don't give them any nova access as that tenant | 01:01 |
| lifeless | and have them nova boot as their own tenant, mixing their own network as eth0, and the provider network as eth1? | 01:01 |
| jog0 | lifeless: hmm, so you want a nova user to be able to use resources from two different tenants in one call? | 01:02 |
| lifeless | jog0: so that bug yes, I've hit that because maddeningly quota wants a uuid for tenant, not a name | 01:02 |
| lifeless | jog0: I do | 01:02 |
| jog0 | lifeless: nova doesn't really understand multiple tenants at once right now | 01:03 |
| lifeless | jog0: more generally, it seems reasonable for users to be able to use stuff from any of their 'tenants' at once | 01:03 |
| lifeless | to me | 01:03 |
| lifeless | otherwise having users be able to be in multiple tenants is entirely pointless | 01:03 |
| jog0 | lifeless: that sounds reasoanble to me too, but things can get hairy I think | 01:03 |
| lifeless | sure | 01:04 |
| lifeless | like I said, just polling opinions | 01:04 |
| jog0 | TBH our tenant/RBAC story is utter crap right now | 01:04 |
| jog0 | so ++ for a big discussion and overahul | 01:04 |
| lifeless | jog0: lets get deployable first :P | 01:04 |
| jog0 | lifeless: heh yeah, and get the dev process working (re: queue delays etc) | 01:05 |
| jog0 | but I do think many people know our multi tenant and RBAC is bad but not sure if keystone folks really grok it. they may treat it as a 'downstream from keystone issue' | 01:06 |
| *** blamar has joined #tripleo | 01:08 | |
| *** UtahDave has quit IRC | 01:17 | |
| *** mestery has quit IRC | 01:18 | |
| *** nosnos has joined #tripleo | 01:35 | |
| *** morazi has joined #tripleo | 01:37 | |
| StevenK | lifeless: First shot -- http://paste.openstack.org/show/61657 | 01:41 |
| lifeless | StevenK: may I suggest; git review ? | 01:41 |
| lifeless | StevenK: you can click 'work in progress' if you don't want folk to review it yet | 01:42 |
| lifeless | StevenK: and it avoids the whole i'm reviewing it but not really thing :) | 01:42 |
| lifeless | StevenK: also, cool | 01:42 |
| lifeless | StevenK: for the location, perhaps /mnt/state/srv/apt-mirror/ ? | 01:42 |
| lifeless | </bikeshed> | 01:43 |
| StevenK | It's not done yet. I'm not happy with the prose in the first paragraph of the README.md, and I'm still trying to figure out how to populate it | 01:43 |
| StevenK | And I need to write out the apache2 config | 01:43 |
| lifeless | oh hmm | 01:44 |
| lifeless | I wouldn't do apache config in this | 01:44 |
| lifeless | thats a separate concern | 01:44 |
| StevenK | Then how do you use said mirror? :-) | 01:44 |
| lifeless | file:/// | 01:44 |
| lifeless | or rsync:/// | 01:45 |
| lifeless | I think we will often want it exported | 01:45 |
| lifeless | but thats a separate service from maintaining it in the first place, no? | 01:45 |
| lifeless | oh, your configuration is all bong - we'll cover that in review (it should be os-apply-config based) | 01:46 |
| lifeless | learning curve stuff though - totally to be expected | 01:46 |
| lifeless | argh neutron dhcp agent races. ARG ARGH ARGH | 02:06 |
| *** ccrouch has joined #tripleo | 02:09 | |
| *** taps has quit IRC | 02:11 | |
| *** CaptTofu has joined #tripleo | 02:19 | |
| *** ftcjeff has joined #tripleo | 02:20 | |
| *** michchap has quit IRC | 02:24 | |
| *** michchap has joined #tripleo | 02:25 | |
| *** ftcjeff has quit IRC | 02:27 | |
| *** vkozhukalov has joined #tripleo | 02:30 | |
| *** coolsvap_away has quit IRC | 02:33 | |
| StevenK | lifeless: You want to review this thing so you can school me about os-apply-config? | 02:34 |
| lifeless | StevenK: sure, have you pushed to gerrit? | 02:38 |
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-image-elements: Add a debian-mirror element https://review.openstack.org/68308 | 02:39 |
| StevenK | lifeless: https://review.openstack.org/68308 | 02:39 |
| StevenK | Bah, I keep forgetting about the helpful bot | 02:40 |
| lifeless | helpful bot is helpful | 02:45 |
| *** AaronGr is now known as aarongr_away | 02:47 | |
| *** CaptTofu has quit IRC | 02:48 | |
| *** epim has quit IRC | 02:56 | |
| lifeless | StevenK: quick review up | 02:59 |
| StevenK | lifeless: Thanks, I'll dig into it | 03:00 |
| StevenK | After some tea | 03:00 |
| *** ramishra has joined #tripleo | 03:13 | |
| *** vipul-away is now known as vipul | 03:17 | |
| StevenK | lifeless: Not sure if nested loops are supported, how can I test this? | 03:33 |
| *** CaptTofu has joined #tripleo | 03:38 | |
| *** CaptTofu has quit IRC | 03:38 | |
| *** CaptTofu has joined #tripleo | 03:39 | |
| SpamapS | StevenK: yes nested loops are supported | 04:03 |
| StevenK | SpamapS: But then how does {{.}} work? | 04:04 |
| SpamapS | StevenK: test with a json file and a template file os-apply-config --fallback-metadata file.json --templates dir/with/your/templates -o /tmp/somewhere | 04:05 |
| SpamapS | actually you don't need fallback metadta you can do -m file.json | 04:06 |
| SpamapS | StevenK: sorry {{.}} ? | 04:06 |
| SpamapS | StevenK: you can do {{#outerkey}}{{#innerkey}}{{innerkey}}{{/innerkey}}{{/outerkey}} | 04:07 |
| *** coolsvap has joined #tripleo | 04:08 | |
| SpamapS | hm | 04:11 |
| SpamapS | StevenK: so I'm playing with this right now and my testing method isn't exactly "awesome" :) | 04:11 |
| StevenK | Haha | 04:11 |
| SpamapS | as in it does not work at all | 04:11 |
| SpamapS | and logging in oac is .. sparse | 04:11 |
| StevenK | You're doing better than me, I'm still working out how to run it without resorting to pip install | 04:12 |
| SpamapS | venv | 04:12 |
| SpamapS | pip install :) | 04:12 |
| lifeless | {{.}} is the current context | 04:14 |
| SpamapS | lifeless: ah, never used it ;) | 04:15 |
| *** CaptTofu has quit IRC | 04:15 | |
| StevenK | SpamapS: I see what you mean about the logging | 04:19 |
| SpamapS | yeah I'm knee deep in pdb now | 04:19 |
| StevenK | SpamapS: And sinking fast, or slow? :-) | 04:20 |
| SpamapS | it seems to be getting thicker so neither :-/ | 04:21 |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Add SSL configuration metadata for overclouds. https://review.openstack.org/67348 | 04:21 |
| StevenK | I have some not so fond memories of pdb'ing through TAL | 04:21 |
| *** akuznetsov has joined #tripleo | 04:25 | |
| SpamapS | StevenK: new people tend to expose our Boggerts ;) | 04:25 |
| lifeless | SpamapS: what are you debugging | 04:26 |
| lifeless | ? | 04:26 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Permit deploying an overcloud with SSL endpoints. https://review.openstack.org/67356 | 04:27 |
| SpamapS | lifeless: os-apply-config | 04:27 |
| lifeless | SpamapS: whats wrong? | 04:27 |
| SpamapS | I recall testing it with this method.. | 04:27 |
| SpamapS | -t template_dir -o output_dir -m file.json .. | 04:27 |
| SpamapS | does not work | 04:27 |
| SpamapS | produces empty files | 04:27 |
| lifeless | tried the variable ? | 04:27 |
| SpamapS | no.. hrm | 04:28 |
| *** spzala has quit IRC | 04:28 | |
| SpamapS | lifeless: moustache is returning u'' for a template of "hello world" .. so .. very confused | 04:29 |
| lifeless | SpamapS: orly | 04:29 |
| SpamapS | wait no its not that was hello world {{a}} | 04:29 |
| SpamapS | hello world owrks | 04:29 |
| SpamapS | works even | 04:29 |
| lifeless | SpamapS: had your dried frog pills today? | 04:30 |
| lifeless | {{a}} will require a key 'a' in your JSON | 04:30 |
| lifeless | StevenK: is SpamapS ^ fail related you your patch ? | 04:30 |
| SpamapS | yes but it should return 'hello world ' for 'hello world {{a}}' | 04:31 |
| SpamapS | I think actually no I'm just failing on 3 levels .. 2 removed.. nested sections is producing u'' | 04:31 |
| SpamapS | or not | 04:33 |
| SpamapS | ok never mind my testing method works fine | 04:33 |
| lifeless | works perfectly for me | 04:34 |
| SpamapS | {{#/a}} | 04:34 |
| SpamapS | That should produce "explosions" | 04:34 |
| SpamapS | instead it produces u'' | 04:34 |
| lifeless | ubuntu@precise-ci-overcloud-3942:~$ cat foo.json | 04:34 |
| lifeless | {"outer": {"inner": ["value", "value2"] } } | 04:34 |
| lifeless | ubuntu@precise-ci-overcloud-3942:~$ cat t/sample | 04:34 |
| lifeless | {{#outer}} | 04:34 |
| lifeless | {{#inner}} | 04:34 |
| lifeless | {{.}} | 04:34 |
| SpamapS | me thinks there be a bug inmoustache | 04:34 |
| lifeless | {{/inner}} | 04:34 |
| lifeless | {{/outer}} | 04:35 |
| lifeless | ubuntu@precise-ci-overcloud-3942:~$ cat o/sample | 04:35 |
| lifeless | value | 04:35 |
| SpamapS | lifeless: no I had a typeo.. _4_ levels of fail resolved | 04:35 |
| lifeless | value2 | 04:35 |
| lifeless | SpamapS: pystache is a pile of fail | 04:35 |
| SpamapS | typo haha | 04:35 |
| SpamapS | I typo'd typo | 04:35 |
| lifeless | SpamapS: if you want a better template engine, may I offer pybars, by yours truely | 04:35 |
| SpamapS | lifeless: yeah we used it because handlebars is under CLA or something right? | 04:35 |
| lifeless | it is indeed a Canonical project | 04:35 |
| SpamapS | a dead-enough-to-fork-if-needed project? | 04:36 |
| lifeless | https://pypi.python.org/pypi/pybars | 04:36 |
| lifeless | LGPL | 04:36 |
| lifeless | anyhow | 04:36 |
| SpamapS | LGPLv3 tho | 04:36 |
| lifeless | or we could use node and actual handlebars | 04:36 |
| SpamapS | haha yes lets bring node into this | 04:36 |
| lifeless | its webscale | 04:36 |
| SpamapS | we will be heralded accross the land! | 04:36 |
| SpamapS | the same way the 4 horsemen will be .. ;) | 04:36 |
| SpamapS | anyway, verified.. testing os-apply-config still works with just clint@clint-HP:~$ os-apply-config -m /tmp/test.json -t /tmp/t -o /tmp/o | 04:39 |
| *** vipul is now known as vipul-away | 04:40 | |
| *** vipul-away is now known as vipul | 04:44 | |
| StevenK | I want a nested loop without nested JSON | 04:48 |
| StevenK | So far my test of {{#debian-mirror.architectures}} | 04:48 |
| lifeless | StevenK: https://etherpad.openstack.org/p/11MSDYHTPO | 04:49 |
| StevenK | lifeless: Hm, that might be the missing piece | 04:53 |
| StevenK | lifeless: Oooh, now it's writing something | 04:56 |
| StevenK | deb-["amd64"] http://archive.ubuntu.com/ubuntu main restricted universe multiverse | 04:56 |
| lifeless | StevenK: in fact | 04:58 |
| lifeless | StevenK: note the extra list | 04:58 |
| StevenK | deb-["amd64"] http://archive.ubuntu.com/ubuntu saucy main restricted universe multiverse | 04:59 |
| lifeless | yeah, sketching a template for you in the etherpad | 04:59 |
| StevenK | You're missing suite | 04:59 |
| lifeless | lets see | 05:00 |
| lifeless | try that | 05:01 |
| StevenK | {{..}} works? Blink | 05:01 |
| lifeless | Dunno, does it ? | 05:01 |
| lifeless | deb-amd64 http://archive.ubuntu.com/ubuntu main restricted universe multiverse | 05:02 |
| lifeless | missing suite | 05:02 |
| openstackgerrit | A change was merged to openstack/os-apply-config: Add new username type https://review.openstack.org/67820 | 05:03 |
| lifeless | but .. did grab the parent I *think* | 05:03 |
| lifeless | ah | 05:04 |
| lifeless | .. is handlebars specific | 05:04 |
| * lifeless thinkgs | 05:04 | |
| lifeless | easiest way is to not do a cross product | 05:05 |
| StevenK | ... but we need to? | 05:05 |
| lifeless | deb-amd64 http://archive.ubuntu.com/ubuntu saucy main restricted universe multiverse | 05:10 |
| lifeless | with what is in the etherpad | 05:10 |
| StevenK | I was trying to avoid that construct | 05:10 |
| StevenK | But yes, I see it works | 05:11 |
| lifeless | I suspect pybars would do the cross product for us | 05:11 |
| lifeless | but moustache simply won't from my quick re-read of the spec | 05:12 |
| lifeless | you need .. to be the suite being iterated over, which is isn't | 05:12 |
| *** vipul is now known as vipul-away | 05:15 | |
| StevenK | "architectures": [{"arch": "amd64"}, {"arch": "i386"}] is a bit ugly, but it works | 05:19 |
| StevenK | Mustache doesn't bend this way, news at 11 | 05:19 |
| StevenK | lifeless: Where do I look for default values? | 05:22 |
| *** coolsvap is now known as coolsvap_away | 05:23 | |
| *** vipul-away is now known as vipul | 05:32 | |
| *** vipul is now known as vipul-away | 05:47 | |
| *** e0ne has joined #tripleo | 05:53 | |
| *** e0ne has quit IRC | 05:54 | |
| *** rpodolyaka1 has joined #tripleo | 05:55 | |
| *** blamar has quit IRC | 06:08 | |
| *** vipul-away is now known as vipul | 06:10 | |
| *** rdopieralski has joined #tripleo | 06:14 | |
| *** CaptTofu has joined #tripleo | 06:16 | |
| *** coolsvap_away is now known as coolsvap | 06:19 | |
| *** CaptTofu has quit IRC | 06:21 | |
| *** blamar has joined #tripleo | 06:23 | |
| *** tzumainn has quit IRC | 06:30 | |
| *** rlandy has joined #tripleo | 06:36 | |
| *** nosnos_ has joined #tripleo | 06:43 | |
| *** nosnos has quit IRC | 06:43 | |
| *** vkozhukalov has quit IRC | 06:52 | |
| lifeless | StevenK: back :) | 06:53 |
| lifeless | StevenK: default values we generally do in the heat template | 06:53 |
| *** xuhanp has joined #tripleo | 06:56 | |
| *** vipul is now known as vipul-away | 07:21 | |
| *** vipul-away is now known as vipul | 07:25 | |
| *** ramishra has quit IRC | 07:27 | |
| *** ramishra has joined #tripleo | 07:28 | |
| *** mrunge has joined #tripleo | 07:28 | |
| *** ramishra has quit IRC | 07:32 | |
| *** rpodolyaka1 has quit IRC | 07:43 | |
| *** jistr has joined #tripleo | 07:43 | |
| *** vipul is now known as vipul-away | 07:45 | |
| GheRivero | morning TripleO! | 07:49 |
| *** vipul-away is now known as vipul | 07:54 | |
| *** e0ne has joined #tripleo | 07:54 | |
| *** ramishra has joined #tripleo | 07:56 | |
| *** jprovazn has joined #tripleo | 08:06 | |
| *** jcoufal has joined #tripleo | 08:13 | |
| *** CaptTofu has joined #tripleo | 08:17 | |
| *** ramishra has quit IRC | 08:18 | |
| rpodolyaka | morning all | 08:21 |
| *** CaptTofu has quit IRC | 08:22 | |
| GheRivero | morning | 08:22 |
| *** jtomasek has joined #tripleo | 08:26 | |
| *** vkozhukalov has joined #tripleo | 08:30 | |
| *** pblaho has joined #tripleo | 08:34 | |
| rlandy | ifarkas: hi | 08:52 |
| ifarkas | rlandy, hi | 08:52 |
| rlandy | ifarkas: I have the undercloud deployed. tuskar-api.service exists but it is not started. | 08:52 |
| rlandy | what did I miss? | 08:53 |
| ifarkas | rlandy, maybe it's because of the bug I reported. What's in the log? | 08:53 |
| rlandy | ifarkas: bug number? | 08:57 |
| rlandy | not sure which log you are after - but looks like installed http://fpaste.org/70579/03810861/ | 08:58 |
| rlandy | meeting brb | 08:59 |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Remove dependencies on pep8, pyflakes and flake8 https://review.openstack.org/68126 | 09:01 |
| ifarkas | rlandy, https://bugs.launchpad.net/tripleo/+bug/1259546 | 09:01 |
| ifarkas | rlandy, I meant the log from the undercloud machine, not the log from dib | 09:02 |
| *** derekh has joined #tripleo | 09:03 | |
| derekh | lifeless: quick question, once devstack-vm-gate-wrap.sh is run the repositories in /opt/stack/new , do you know if they are left so that the one that triggered the job is at the correct commit ? | 09:05 |
| derekh | lifeless: from looking at the logs is looks like they are, if that the case then with https://review.openstack.org/#/c/68178/1 I don't think we'll need to add any ZUUL trickery | 09:07 |
| derekh | lifeless: we could just add the tripleo projects to https://github.com/openstack-infra/devstack-gate/blob/master/devstack-vm-gate-wrap.sh and set /opt/stack/new as TRIPLEO_ROOT | 09:08 |
| derekh | pleia2: ^ | 09:10 |
| derekh | if that works, once other patches merge I think we're pretty much ready to try it out | 09:11 |
| ifarkas | rlandy, the bug linked is not about the installation of tuskar. If the config file is broken then the service just won't start | 09:17 |
| ifarkas | rlandy, eg. you can check with systemctl status on the undercloud machine | 09:17 |
| derekh | back soon | 09:18 |
| *** derekh has quit IRC | 09:18 | |
| jprovazn | SpamapS: hi, you still around? | 09:24 |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-incubator: Add openssh-server to install-dependencies https://review.openstack.org/68291 | 09:29 |
| *** ramishra has joined #tripleo | 09:29 | |
| *** ramishra has quit IRC | 09:33 | |
| *** ramishra has joined #tripleo | 09:34 | |
| *** bauzas has joined #tripleo | 09:36 | |
| *** jistr has quit IRC | 09:37 | |
| *** markmc has joined #tripleo | 09:38 | |
| *** xuhanp has quit IRC | 09:52 | |
| *** jistr has joined #tripleo | 09:57 | |
| *** akrivoka has joined #tripleo | 10:00 | |
| *** athomas has joined #tripleo | 10:00 | |
| *** boris-42 has quit IRC | 10:02 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Add Copyright and License header to debian element https://review.openstack.org/66975 | 10:02 |
| *** boris-42 has joined #tripleo | 10:04 | |
| *** martyntaylor has joined #tripleo | 10:08 | |
| *** akuznetsov has quit IRC | 10:12 | |
| *** martyntaylor has quit IRC | 10:12 | |
| *** pblaho has quit IRC | 10:12 | |
| *** akuznetsov has joined #tripleo | 10:13 | |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Add openssh-server to install-dependencies https://review.openstack.org/68291 | 10:13 |
| openstackgerrit | Jiri Tomasek proposed a change to openstack/tuskar-ui: Number picker input widget and angular directive https://review.openstack.org/68053 | 10:15 |
| *** CaptTofu has joined #tripleo | 10:18 | |
| *** CaptTofu has quit IRC | 10:22 | |
| *** xuhanp has joined #tripleo | 10:25 | |
| *** xuhanp has quit IRC | 10:26 | |
| *** martyntaylor has joined #tripleo | 10:27 | |
| *** jesusaurus has quit IRC | 10:35 | |
| *** derekh has joined #tripleo | 10:35 | |
| *** jesusaurus has joined #tripleo | 10:37 | |
| *** max_lobur_afk is now known as max_lobur | 10:39 | |
| lifeless | derekh: hi so | 10:49 |
| lifeless | derekh: yes but, what about the nova version | 10:49 |
| derekh | lifeless: I'm not sure what you mean? taking nova as an example, devstack-vm-gate-wrap.sh sets the master branch of /opt/stack/git/nova to the commit the triggered the job | 10:51 |
| derekh | lifeless: soo we can just use that | 10:51 |
| derekh | lifeless: at least I think it does | 10:52 |
| * derekh in hospital ATM waiting on a root X-ray , sorry if I disappear suddenly. | 10:53 | |
| derekh | root -> foot | 10:53 |
| lifeless | derekh: we use a nova tree from ~/.cache/image-create/.... | 10:54 |
| lifeless | anyhow gnight | 10:55 |
| derekh | lifeless: yup, but setting DIB_REPOLOCATION_nova=/opt/stack/git/new/nova will pull the current master branch of nova into ~/.cache... | 10:55 |
| lifeless | derekh: so we do need to set some setuff :) | 10:55 |
| derekh | lifeless: yup, this https://review.openstack.org/#/c/68178/ | 10:56 |
| derekh | lifeless: I think thats enough, anyways I'll dig into it a bit more and know more later | 10:57 |
| *** pblaho has joined #tripleo | 10:57 | |
| *** lucasagomes has joined #tripleo | 10:57 | |
| *** michchap has quit IRC | 11:06 | |
| *** michchap has joined #tripleo | 11:06 | |
| lifeless | derekh: oh; you might talk to fungi when he gets up, about enabling ci-overcloud in production nodepool | 11:11 |
| lifeless | derekh: since that way it could be earlier in his day :) | 11:11 |
| * lifeless is really gone now | 11:11 | |
| derekh | lifeless: willdo gnight | 11:12 |
| *** lucasagomes is now known as lucas-afk | 11:17 | |
| *** e0ne_ has joined #tripleo | 11:17 | |
| *** lucas-afk is now known as lucasagomes | 11:19 | |
| *** e0ne has quit IRC | 11:21 | |
| *** derekh has quit IRC | 11:29 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Update default RHEL guest image https://review.openstack.org/68206 | 11:37 |
| *** boris-42 has quit IRC | 11:53 | |
| *** boris-42 has joined #tripleo | 11:56 | |
| *** ramishra_ has joined #tripleo | 12:09 | |
| lsmola_ | marios: hello | 12:10 |
| *** ramishra has quit IRC | 12:11 | |
| *** jcoufal has quit IRC | 12:14 | |
| *** jcoufal has joined #tripleo | 12:15 | |
| *** ifarkas has quit IRC | 12:16 | |
| *** ifarkas has joined #tripleo | 12:17 | |
| *** CaptTofu has joined #tripleo | 12:19 | |
| *** CaptTofu has quit IRC | 12:23 | |
| openstackgerrit | Jiri Tomasek proposed a change to openstack/tuskar-ui: Number picker input widget and angular directive https://review.openstack.org/68053 | 12:28 |
| marios | lsmola_: hey | 12:31 |
| lsmola_ | marios: heya | 12:31 |
| lsmola_ | marios: just wanted to ask what do plan to include in you patch | 12:31 |
| lsmola_ | marios: are you planning to implement there also the heat stack-create/update, or that will be separate patch | 12:32 |
| lsmola_ | marios: ? | 12:32 |
| *** julim has joined #tripleo | 12:32 | |
| lsmola_ | marios: hopefully I will have the devtest ready fo testing overcloud deployment this week | 12:33 |
| marios | lsmola_: perfect that will be helpful. so i hope to have another version out today. not sure about what you ask, i guess it belongs here... but there is no controller yet (since that is where the heat client will be invoked from) | 12:33 |
| marios | lsmola_: when you are ready for testing ping me and we can work it out from there (hopefully jay will be moving along too so we have progress generally) | 12:34 |
| lsmola_ | marios: ok | 12:35 |
| lsmola_ | marios: it would be perfectly fine is you prepare your first version without jay | 12:35 |
| lsmola_ | marios: just hardcode the numbers you want to deploy and everything | 12:36 |
| marios | yes this is plan ^^ | 12:36 |
| lsmola_ | marios: ok | 12:36 |
| lsmola_ | marios: I will ping you when i am there | 12:36 |
| lsmola_ | marios: are you also looking into deploying of block and object storage on separate nodes? | 12:36 |
| lsmola_ | marios: or that is for later? | 12:37 |
| lsmola_ | marios: the templates for that has been already merged in tripleo | 12:37 |
| lsmola_ | marios: I just have no idea how to use them :-) | 12:37 |
| marios | lsmola_: right. in theory both cinder and swift have templates. am more sure about cinder but not so much about swift. i guess we'll find out. but for initial perhaps start really simple, 1 compute, 1 controller for example | 12:39 |
| lsmola_ | marios: I've talked with SpamapS about swift and he says it should be quite easy to use it on separate nodes | 12:41 |
| lsmola_ | marios: so he is probably the person you should ask how to do that :-D | 12:41 |
| marios | lsmola_: ok thanks | 12:41 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Move cinder iptables rules into scoped elements https://review.openstack.org/68118 | 12:46 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-incubator: Document stable branches in devtest https://review.openstack.org/68274 | 12:47 |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Write public key to authorized_keys https://review.openstack.org/68139 | 12:47 |
| *** CaptTofu has joined #tripleo | 12:50 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Fixes cinder element dependencies https://review.openstack.org/68204 | 12:51 |
| *** CaptTofu has quit IRC | 12:55 | |
| *** coolsvap has quit IRC | 13:03 | |
| rlandy | ifarkas: apologies ... had to attend presentation 'walk through's here. To continue - I can start the tuskar-api service from the undercloud manually | 13:04 |
| lsmola_ | marios: other important thing | 13:05 |
| lsmola_ | marios: step 15 http://docs.openstack.org/developer/tripleo-incubator/devtest_overcloud.html | 13:05 |
| lsmola_ | marios: we will have to implement it in tuskar, with some wait_for loop | 13:05 |
| lsmola_ | marios: not sure about the user_config, but other are just API calls, so it should be fine | 13:06 |
| lsmola_ | marios: this haven't been previously in tuskar right? so you had to run it in console? | 13:06 |
| ifarkas | rlandy, so you can start up manually but it doesn't start automatically when the system boots? | 13:07 |
| marios | lsmola_: we used to do config/setup with bash scripts (sent as heat metadata), to be executed after the notcompute was up | 13:07 |
| marios | lsmola_: but not any more. yes we need to work that out. | 13:07 |
| rlandy | ifarkas: yes - the service was stopped initially but 'service tuskar-api restart' worked | 13:07 |
| ifarkas | rlandy, what did you find in the log? | 13:08 |
| lsmola_ | marios: ok | 13:08 |
| rlandy | ifarkas: are we talking about /var/log/os-apply-config.log? or another log? | 13:09 |
| rlandy | before I send the wrong info | 13:09 |
| ifarkas | rlandy, journalctl | 13:09 |
| *** ramishra_ has quit IRC | 13:10 | |
| *** gilliard has joined #tripleo | 13:11 | |
| rlandy | ifarkas: + /opt/stack/venvs/tuskar/bin/tuskar-dbsync --config-file /etc/tuskar/tuskar.conf | 13:15 |
| rlandy | 2014-01-22 12:08:09.459 5512 CRITICAL tuskar [-] Could not parse rfc1738 URL from string ' | 13:15 |
| ifarkas | rlandy, hmm, that seems to be a different error message than the previously mentioned bug would produce | 13:17 |
| ifarkas | rlandy, maybe this is a new bug? | 13:17 |
| *** mrunge has quit IRC | 13:19 | |
| rlandy | not sure | 13:20 |
| Ng | Morning | 13:23 |
| *** derekh has joined #tripleo | 13:35 | |
| lsmola_ | NobodyCam: hello | 13:37 |
| *** jdob has joined #tripleo | 13:39 | |
| *** jayg|g0n3 is now known as jayg | 13:44 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-incubator: Update cinder usage https://review.openstack.org/68129 | 13:44 |
| *** akuznetsov has quit IRC | 13:49 | |
| *** ftcjeff has joined #tripleo | 13:52 | |
| *** rbrady has quit IRC | 13:57 | |
| *** mestery has joined #tripleo | 13:57 | |
| *** rbrady has joined #tripleo | 13:57 | |
| *** tzumainn has joined #tripleo | 14:01 | |
| *** dprince has joined #tripleo | 14:01 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Add support Red Hat Network (RHN) https://review.openstack.org/68207 | 14:02 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Support list of Red Hat channels and repos https://review.openstack.org/68208 | 14:02 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Add support for Red Hat Satellite https://review.openstack.org/68209 | 14:02 |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Number picker input widget and angular directive https://review.openstack.org/68053 | 14:04 |
| *** akuznetsov has joined #tripleo | 14:04 | |
| *** boris-42 has quit IRC | 14:06 | |
| *** CaptTofu has joined #tripleo | 14:06 | |
| *** boris-42 has joined #tripleo | 14:10 | |
| *** ramishra has joined #tripleo | 14:11 | |
| *** lblanchard has joined #tripleo | 14:11 | |
| openstackgerrit | Dirk Mueller proposed a change to openstack/tripleo-incubator: Properly import GPG key for repository https://review.openstack.org/68389 | 14:13 |
| openstackgerrit | Dirk Mueller proposed a change to openstack/tripleo-incubator: Also install libxslt-devel https://review.openstack.org/68390 | 14:13 |
| *** matsuhashi has joined #tripleo | 14:14 | |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Add cluster support for rabbitmq element https://review.openstack.org/68392 | 14:16 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Update openstack config files to support rabbitmq cluster https://review.openstack.org/68393 | 14:16 |
| *** ramishra has quit IRC | 14:16 | |
| rlandy | rpodolyaka: hi - if I want to include tuskar in the devtest undercloud, ifarkas let me know that I need to add the tuskar-source.yaml file to the template I am using to deploy the undercloud. How would I do that? | 14:20 |
| *** coolsvap has joined #tripleo | 14:22 | |
| rbrady | rlandy: look at the Makefile in tripleo-heat-templates. There's a target there "undercloud-vm-tuskar.yaml" | 14:22 |
| rlandy | rbrady: thanks - see it | 14:24 |
| rpodolyaka | rlandy: yeah, you need that target | 14:26 |
| *** ramishra has joined #tripleo | 14:32 | |
| *** CaptTofu has quit IRC | 14:35 | |
| *** matty_dubs|gone is now known as matty_dubs | 14:41 | |
| *** matsuhashi has quit IRC | 14:45 | |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Added initial tuskar api tests https://review.openstack.org/67940 | 14:56 |
| *** nosnos_ has quit IRC | 14:57 | |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Update openstack config files to support rabbitmq cluster https://review.openstack.org/68393 | 15:03 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Add cluster support for rabbitmq element https://review.openstack.org/68392 | 15:03 |
| *** CaptTofu has joined #tripleo | 15:05 | |
| *** rdopieralski has quit IRC | 15:10 | |
| NobodyCam | morning lsmola_ | 15:11 |
| lsmola_ | NobodyCam: hello | 15:11 |
| lsmola_ | NobodyCam: just wanted to ask what is status for this https://github.com/NoBodyCam/ironic-element | 15:12 |
| lsmola_ | NobodyCam: is there some official patch for tripleo being build? | 15:12 |
| NobodyCam | lsmola_: 7:00 am here... just making coffee | 15:12 |
| lsmola_ | NobodyCam: oh, ok :-) | 15:12 |
| NobodyCam | lsmola_: I have to update myself of the status of serveral OoO patches | 15:13 |
| lsmola_ | NobodyCam: ok | 15:13 |
| marios | bnemec: slagle: thanks for your comments on the merge.py stuff, very useful. updating in a bit | 15:25 |
| lsmola_ | NobodyCam: do you think you could rebase https://review.openstack.org/#/c/51328/9 | 15:27 |
| lsmola_ | NobodyCam: it probably has some old keystone and we are getting Not authorized error from nova then | 15:27 |
| NobodyCam | lsmola_: ya. will do. in a bit. but, where are you getting that.. | 15:29 |
| ifarkas | lsmola_, akrivoka, is it the same error you find in log when using nova-ironic: http://paste.openstack.org/show/61681/ ? | 15:29 |
| akrivoka | ifarkas: no. I got the unauthorized 401 when trying "nova list". but nova service was running. | 15:30 |
| lsmola_ | ifarkas: hmm didn't noticed this | 15:30 |
| *** lucasagomes is now known as lucas-hungry | 15:31 | |
| ifarkas | akrivoka, I also got unauthorized error from client but this issue is causing that | 15:31 |
| ifarkas | at least for me | 15:31 |
| lsmola_ | NobodyCam: is this cherrypicking the commit on nova master? https://github.com/NoBodyCam/ironic-element/blob/master/nova-ironic/install.d/89-nova-ironic-services#L13 | 15:32 |
| NobodyCam | lsmola_: can you paste your ironic conf? | 15:32 |
| akrivoka | I have trashed my machine so can't say. If I run into it again will check. | 15:32 |
| lsmola_ | NobodyCam: yep, we have trashed it :-) but that was that we have accidentaly changed DIB_REPO_nova for seed, to that patch | 15:33 |
| lsmola_ | NobodyCam: but is you are cherrypicking that, ifarkas has probably different issue | 15:34 |
| lsmola_ | NobodyCam: I should get to deploying undercloud soon, so we will see what I have | 15:34 |
| NobodyCam | lsmola_: did you set IronicPassword=${UNDERCLOUD_IRONIC_PASSWORD} on the heat stack-create line? | 15:35 |
| lsmola_ | akrivoka: ^^ | 15:36 |
| lsmola_ | NobodyCam: ok, I think we should focus on ifarkas bug :-D | 15:37 |
| openstackgerrit | Marios Andreou proposed a change to openstack/tuskar: WIP: Using merge.py from tuskar to generate overcloud.yaml https://review.openstack.org/52045 | 15:38 |
| NobodyCam | lsmola_: yes :) | 15:38 |
| ifarkas | NobodyCam, I wrote the steps how to deploy with ironic in tripleo devtest env: https://etherpad.openstack.org/p/IronicDeployDevtest | 15:40 |
| ifarkas | or at least I am trying to create a step by step guide | 15:40 |
| lsmola_ | ifarkas: me and akrivoka too | 15:41 |
| lsmola_ | NobodyCam: ifarkas https://etherpad.openstack.org/p/custom-devtest | 15:42 |
| NobodyCam | awesome. we landed a bunch of lucas's patches yesterday | 15:42 |
| NobodyCam | I'm going to run thru a test here in a bit and will (rebase) and push up new patches | 15:43 |
| lsmola_ | NobodyCam: yeah I've checked, so we are now applying only the nova patch and the two image elements | 15:43 |
| lsmola_ | NobodyCam: (also the heat patch wasn't merged yet) | 15:44 |
| *** akuznetsov has quit IRC | 15:47 | |
| NobodyCam | lsmola_: side note: who is the owner of /tftpboot and /var/lib/ironic .... should be ironic? | 15:49 |
| NobodyCam | s/?/./ | 15:50 |
| lsmola_ | NobodyCam: still building undercloud | 15:50 |
| lsmola_ | ifarkas: ^^ | 15:50 |
| openstackgerrit | Ralf Haferkamp proposed a change to openstack/diskimage-builder: Fix ramdisk element for openSUSE https://review.openstack.org/68418 | 15:54 |
| *** akuznetsov has joined #tripleo | 15:55 | |
| *** jprovazn has quit IRC | 15:58 | |
| rlandy | rpodolyaka: heat stack-create (with undercloud-vm-tuskar.yaml) is expecting OvercloudKeystoneHost parameter. (There is no default:unset) Should OvercloudKeystoneHost exist at this point? | 15:59 |
| *** sballe has joined #tripleo | 16:02 | |
| rpodolyaka | rlandy: yeah, so this is a limitation of tuskar right now | 16:03 |
| rpodolyaka | rlandy: you must pass this parameter manually | 16:03 |
| rpodolyaka | rlandy: let me check the code | 16:04 |
| rpodolyaka | rlandy: oh, we even have a bug filed for tracking of this https://bugs.launchpad.net/tuskar/+bug/1235236 | 16:06 |
| *** vkozhukalov has quit IRC | 16:07 | |
| rpodolyaka | rlandy: ideally, tuskar will get those values from Heat when creating an overcloud | 16:08 |
| *** CaptTofu has quit IRC | 16:08 | |
| rpodolyaka | rlandy: but now they must be specified in the config file | 16:08 |
| rpodolyaka | rlandy: that's why element/template require you to pass those values | 16:08 |
| rlandy | rpodolyaka: ok, thanks ... I need to look at this more carefully (element/template) | 16:09 |
| *** UtahDave has joined #tripleo | 16:10 | |
| *** CaptTofu has joined #tripleo | 16:10 | |
| rpodolyaka | rlandy: np, the element is kind of ugly right now as it requires you to pass the IP of overclud control node, which you can't know in advance :( | 16:10 |
| openstackgerrit | Dirk Mueller proposed a change to openstack-infra/tripleo-ci: Avoid checking out triple- related repositories twice https://review.openstack.org/68423 | 16:10 |
| rlandy | rpodolyaka: yeah, was thinking it would not be possible to deploy the undercloud with tuskar at this point | 16:11 |
| rpodolyaka | rlandy: I'd say it is possible, but manual tweaking of config files might be needed | 16:13 |
| rpodolyaka | rlandy: at least it was possible last time I tried tuskar element | 16:13 |
| rpodolyaka | rlandy: but it was a few months ago... | 16:13 |
| rlandy | rpodolyaka: ok - so the question is, is it worth continuing down this path right now? | 16:13 |
| rpodolyaka | rlandy: it depends on what you are trying to achieve :) | 16:14 |
| rlandy | rpodolyaka: the goal was not to bug everybody on the channel :) | 16:14 |
| rpodolyaka | hey, we are friendly, don't hesitate to ask :) | 16:15 |
| rlandy | rpodolyaka: I just picked up a bug that required looking into tripleo/tuskar - which btw; I am interested in but I's rather the time is well-spent at the point that the two pieces are ready to be worked with | 16:16 |
| *** jcoufal has quit IRC | 16:16 | |
| rlandy | it was a fun learning exercise in the mean time | 16:16 |
| * SpamapS awakens | 16:17 | |
| rpodolyaka | so we basically have a working devtest story | 16:17 |
| rpodolyaka | which currently doesn't use tuskar | 16:17 |
| rpodolyaka | at some point, we'll either incorporate tuskar into existing devtest story or provide a similar tuskar story | 16:17 |
| rpodolyaka | it's the tuskar story we are missing right now | 16:18 |
| rpodolyaka | there is tuskar element, but it has... limitations | 16:18 |
| rpodolyaka | so it's not 'out-of-box' experience right now | 16:18 |
| rbrady | Spamaps: ping. help. I offer a malty and hoppy beverage (or another of your choosing) at a date in the future in trade for your heat template guidance. | 16:19 |
| rlandy | rpodolyaka: understood... | 16:20 |
| rpodolyaka | rlandy: so it's worth continuing down this path, if you want to learn how tripleo/tuskar work together | 16:20 |
| rpodolyaka | rlandy: but it's not, if you need working undercloud/overcloud | 16:20 |
| rlandy | rpodolyaka: it's the first scenario | 16:21 |
| *** lucas-hungry is now known as lucasagomes | 16:21 | |
| rpodolyaka | rlandy: then you are welcome to ask questions, file new bugs and fix existing ones :) | 16:22 |
| openstackgerrit | Petr Blaho proposed a change to openstack/python-tuskarclient: Adds help for subcommands https://review.openstack.org/56257 | 16:23 |
| rlandy | rpodolyaka: ok - then I'd have to start by asking about "manual tweaking of config files" :) | 16:24 |
| rpodolyaka | rlandy: sure | 16:24 |
| rlandy | rpodolyaka: but ... I need to travel back - so in a few hours ( or tomorrow) depending on your time zone | 16:24 |
| SpamapS | rpodolyaka: devtest is not meant to be "out of box". Quite the contrary, it is meant to be "look inside our magic box to see the gears and levers and duct tape" :) | 16:25 |
| rpodolyaka | rlandy: feel free to ping me, I'll try to help, if I'm not afk | 16:25 |
| SpamapS | rpodolyaka: I'd very much love to have tuskar encapsulate all of the intentionally manual and explicit steps into an "OpenStack installer" experience. :) | 16:25 |
| SpamapS | rbrady: what can I do for you sir? | 16:26 |
| rlandy | rpodolyaka: great, thanks | 16:26 |
| rlandy | back later | 16:26 |
| rpodolyaka | rlandy: np | 16:26 |
| *** rlandy has quit IRC | 16:26 | |
| rpodolyaka | SpamapS: yeah and it's very helpful to look inside the magic box to learn how it's built :) | 16:26 |
| SpamapS | rpodolyaka: note that I'd also very much like for Tuskar to remain mostly about a UI, and not get into things like driving merge.py. We should be _aggressively_ pushing Heat to be better, and only working around it when we can't get what we want out of Heat. | 16:27 |
| rpodolyaka | SpamapS: can't agree more :) | 16:27 |
| SpamapS | so -> go review Heat patches ;) | 16:27 |
| rpodolyaka | hehe | 16:28 |
| SpamapS | (only half kidding) | 16:28 |
| rbrady | SpamapS: I need to make this https://github.com/openstack/tripleo-heat-templates/blob/master/block-storage.yaml spawn 3 instances instead of one. I tried updating AllowedResources and copy/pasting BlockStorage0, changing the "0" to 1 and 2 respectively. I didn't have any luck going that route and I haven't found what I'm looking for in the docs yet. Is there a better way to approach this? | 16:28 |
| SpamapS | rbrady: the copy/paste approach is sort of what tuskar does for NovaCompute0 | 16:28 |
| SpamapS | rbrady: you can use OS::Heat::InstanceGroup .. but "thar be dragons" | 16:29 |
| SpamapS | rbrady: the biggest dragon being that sizing down means arbitrary deletion of the last member. | 16:29 |
| openstackgerrit | Tzu-Mainn Chen proposed a change to openstack/tuskar-ui: Update API documentation to use proper docstrings https://review.openstack.org/68427 | 16:30 |
| SpamapS | rbrady: in my infinite amounts of available time, I have been trying to devise a way to get Heat to not arbitrarily delete the last one if one of the others has been deleted already. | 16:30 |
| SpamapS | The problem right now is it goes for i in range(size): make_a_server_named("resourceName-%d" % i) | 16:31 |
| marios | SpamapS: rbrady: fwiw, this is how tuskar uses merge.py... generate file that looks like: http://paste.openstack.org/show/61684/ then use merge.py to unite that with overcloud-source.yaml | 16:31 |
| marios | rbrady: not sure if that helps with what you're doing | 16:31 |
| SpamapS | it needs to go "while count(active_servers) < size: make_a_randomly_named_server() | 16:31 |
| SpamapS | marios: yeah, thats the thing where I am a little annoyed that Tuskar didn't just fix Heat. :) | 16:32 |
| *** ramishra has quit IRC | 16:32 | |
| SpamapS | flattered that merge.py gets so much usage, but horrified that merge.py gets so much usage :) | 16:32 |
| *** jayg has quit IRC | 16:33 | |
| marios | SpamapS: :) indeed, but it is what it is right now and there isn't enough time for I right. I mean, most of the 'this is what we need to do in heat' was worked out relatively recently, as far as i understand anyway | 16:33 |
| hewbrocca | SpamapS: heh... we're working on fixing Heat, srsly... | 16:33 |
| *** jomara_ has quit IRC | 16:33 | |
| *** bcrochet has quit IRC | 16:33 | |
| hewbrocca | agree we lost a lot of time not starting with that however | 16:33 |
| hewbrocca | sigh | 16:33 |
| *** matty_dubs has quit IRC | 16:35 | |
| lsmola_ | SpamapS: lol | 16:35 |
| SpamapS | marios: yeah we only just figured the actual path out last week I think. | 16:35 |
| SpamapS | hewbrocca: learned much we have from our mistakes. | 16:35 |
| hewbrocca | indeed | 16:36 |
| lsmola_ | SpamapS: copy/paste is becoming popular again :-D | 16:36 |
| SpamapS | lsmola_: its 1995 all over again | 16:36 |
| SpamapS | it's | 16:36 |
| lsmola_ | hehe | 16:36 |
| SpamapS | I think the Trove and Savanna people will also be happy to make use of a feature which allows selective node death btw. | 16:37 |
| lsmola_ | SpamapS: I believe that feature might land in heat in I3 | 16:37 |
| lsmola_ | SpamapS: or not? | 16:37 |
| *** martyntaylor has quit IRC | 16:39 | |
| * rbrady thinks "selective node death" sounds like a metal band name | 16:39 | |
| lsmola_ | SpamapS: ifarkas ok undercloud booted and I have nova-api running | 16:40 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-incubator: Fix SSL .crt error when not using SSL. https://review.openstack.org/68432 | 16:42 |
| *** ramishra has joined #tripleo | 16:42 | |
| *** e0ne_ has quit IRC | 16:43 | |
| *** lblanchard1 has joined #tripleo | 16:44 | |
| *** rbrady has quit IRC | 16:44 | |
| SpamapS | hah I want their t-shirt | 16:44 |
| *** lblanchard has quit IRC | 16:44 | |
| *** jomara has joined #tripleo | 16:45 | |
| *** bcrochet has joined #tripleo | 16:45 | |
| *** jayg has joined #tripleo | 16:46 | |
| *** ramishra has quit IRC | 16:46 | |
| *** funzo has quit IRC | 16:47 | |
| *** funzo has joined #tripleo | 16:47 | |
| *** matty_dubs has joined #tripleo | 16:47 | |
| *** bcrochet has quit IRC | 16:53 | |
| *** jomara has quit IRC | 16:55 | |
| *** matty_dubs has quit IRC | 16:55 | |
| *** rbrady has joined #tripleo | 16:55 | |
| *** jayg has quit IRC | 16:57 | |
| *** mrunge has joined #tripleo | 16:57 | |
| *** taps has joined #tripleo | 16:58 | |
| *** lblanchard has joined #tripleo | 17:02 | |
| openstackgerrit | Ralf Haferkamp proposed a change to openstack/diskimage-builder: Fix copy_required_libs() for newer ldd releases https://review.openstack.org/68434 | 17:04 |
| *** lblanchard2 has joined #tripleo | 17:04 | |
| *** lblanchard1 has quit IRC | 17:06 | |
| *** bcrochet has joined #tripleo | 17:06 | |
| *** jayg has joined #tripleo | 17:07 | |
| *** lblanchard has quit IRC | 17:08 | |
| *** matty_dubs has joined #tripleo | 17:09 | |
| *** jomara_ has joined #tripleo | 17:10 | |
| *** matty_dubs is now known as matty_dubs|lunch | 17:11 | |
| *** bauzas has quit IRC | 17:14 | |
| *** jayg has quit IRC | 17:14 | |
| *** bcrochet has quit IRC | 17:15 | |
| *** matty_dubs|lunch has quit IRC | 17:15 | |
| *** jomara_ has quit IRC | 17:15 | |
| *** pblaho has quit IRC | 17:17 | |
| *** aarongr_away is now known as AaronGr | 17:17 | |
| *** jomara_ has joined #tripleo | 17:20 | |
| *** bcrochet has joined #tripleo | 17:23 | |
| *** mrunge is now known as mrunge_away | 17:23 | |
| *** matty_dubs|lunch has joined #tripleo | 17:24 | |
| *** marun has joined #tripleo | 17:30 | |
| *** jistr has quit IRC | 17:32 | |
| *** gothicmindfood has joined #tripleo | 17:32 | |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/os-apply-config: Change default os_config_files location https://review.openstack.org/67327 | 17:33 |
| ccrouch | NobodyCam: hi there | 17:35 |
| ccrouch | are you expecting this will land in icehouse: https://blueprints.launchpad.net/ironic/+spec/integration-into-tripleo ? | 17:35 |
| *** lucasagomes has quit IRC | 17:40 | |
| *** blamar_ has joined #tripleo | 17:40 | |
| *** blamar has quit IRC | 17:43 | |
| *** blamar_ is now known as blamar | 17:43 | |
| * Ng trundles out to get the kids from nursery and put dinner in their little faces | 17:44 | |
| NobodyCam | ccrouch: hey htere | 17:44 |
| NobodyCam | ccrouch: not in I-2 | 17:45 |
| NobodyCam | :-p but hopping by I-3 | 17:45 |
| *** mrunge_away is now known as mrunge | 17:46 | |
| SpamapS | dprince: responded to your concerns about the TripleO update story :) | 17:48 |
| dprince | SpamapS: awesome, thanks. | 17:49 |
| greghaynes | Morning, tripleo | 17:49 |
| ccrouch | NobodyCam: thanks! | 17:49 |
| NobodyCam | :) | 17:50 |
| ccrouch | NobodyCam: I'm just wondering how stable that will be compared to nova-bm. I guess the goal would be to stabilize between I3 and release? | 17:51 |
| NobodyCam | ccrouch: that is def part of ironic;s overall goal, to provide a stable env :) | 17:52 |
| *** gothicmindfood has left #tripleo | 17:53 | |
| *** lucasagomes has joined #tripleo | 17:55 | |
| *** lucasagomes has quit IRC | 17:55 | |
| ccrouch | right, i'm just wondering if that is going meant to be all done by icehouse? | 17:55 |
| *** lucasagomes has joined #tripleo | 17:55 | |
| NobodyCam | ccrouch: we think so at this point | 17:56 |
| ccrouch | okay great | 17:56 |
| *** matty_dubs|lunch is now known as matty_dubs | 17:56 | |
| NobodyCam | ::) | 17:57 |
| devananda | ccrouch: i know we'd *like* it to be done by icehouse release :) | 17:59 |
| ccrouch | ;-) | 18:00 |
| lsmola_ | NobodyCam: so, undercloud is working for me | 18:01 |
| lsmola_ | NobodyCam: but when I deploy overcloud i still get No valid host was found | 18:01 |
| lsmola_ | NobodyCam: so scheduler doesn't find my Ironic instances | 18:01 |
| ccrouch | devananda: caveats around opensource development etc :-) | 18:01 |
| lsmola_ | NobodyCam: any idea how to debug it? | 18:01 |
| devananda | ccrouch: right ;) | 18:02 |
| devananda | lsmola_: are you seeing compute resources in the undercloud n-cpu logs? | 18:03 |
| NobodyCam | lsmola_: the element in my personal repo does not set compute driver to ironic.driver.IronicDriver | 18:03 |
| NobodyCam | also have you added a node to ironic? | 18:04 |
| *** mrunge has quit IRC | 18:05 | |
| *** vkozhukalov has joined #tripleo | 18:05 | |
| lsmola_ | NobodyCam: yes I have nodes | 18:06 |
| NobodyCam | and switched the driver in nova.conf? | 18:07 |
| lsmola_ | NobodyCam: I believe this is suppose to switch driver sed -i 's/baremetal.driver.BareMetalDriver/ironic.driver.IronicDriver/' $TRIPLEO_ROOT/tripleo-heat-templates/undercloud-vm-source.yaml | 18:07 |
| NobodyCam | :) | 18:07 |
| NobodyCam | yep | 18:08 |
| NobodyCam | lsmola_: can you paste me a ironic node-show for the node you added to ironic | 18:08 |
| lsmola_ | NobodyCam: my /etc/nova/nova.conf from undercloud http://paste.openstack.org/show/61696/ | 18:09 |
| hewbrocca | SpamapS: We were asking ourselves exactly that same question about no-reboot updates just the other day | 18:09 |
| *** CaptTofu has quit IRC | 18:10 | |
| NobodyCam | lsmola_: instance_type_extra_specs is not correct | 18:10 |
| hewbrocca | that is a decent answer, although it makes me wonder about discarding all that careful work that's realized in the RPM %post section... | 18:10 |
| NobodyCam | lsmola_: I use instance_type_extra_specs = cpu_arch:i386 | 18:10 |
| hewbrocca | dprince: ^^^ | 18:11 |
| lsmola_ | SpamapS: nova show http://paste.openstack.org/show/61698/ | 18:11 |
| NobodyCam | lsmola_: I need "ironic node-show <NodeUuId>" | 18:11 |
| SpamapS | hewbrocca: it is a different philosophy | 18:12 |
| hewbrocca | Yes... yes it is | 18:12 |
| SpamapS | hewbrocca: there's no need for careful %post's if you have a narrow scope such as "OpenStack" | 18:13 |
| lsmola_ | NobodyCam: ha, my ironic node is down | 18:13 |
| SpamapS | hewbrocca: especially when your narrow scope is designed to be extremely automated and testable, and is in fact already heavily covered by automated tests. | 18:13 |
| lsmola_ | NobodyCam: http://paste.openstack.org/show/61699/ | 18:14 |
| SpamapS | hewbrocca: instead you focus on eliminating entropy | 18:14 |
| hewbrocca | SpamapS: Yeah, that's the rationale | 18:14 |
| * hewbrocca hopes it works :) | 18:14 | |
| SpamapS | hewbrocca: which the %post does one package at a time.. we're trying to do it one server at a time. :) | 18:14 |
| hewbrocca | I guess I think it's worth a punt, or we wouldn't be here | 18:14 |
| NobodyCam | ahh | 18:14 |
| hewbrocca | but I'm not going to lie to you, it scares the bejesus out of me :) | 18:14 |
| SpamapS | hewbrocca: is it just "new thing, haven't thought long enough about it" or "specific scary thing"? | 18:15 |
| NobodyCam | lsmola_: looks like there are some missing bits of data for the node | 18:15 |
| NobodyCam | lsmola_: here is what I use to create a test node: http://paste.openstack.org/show/2LQ3PEtPC524o6jEOG05/ | 18:16 |
| *** derekh has quit IRC | 18:16 | |
| hewbrocca | SpamapS: oh, well, I am an old ancient RH-er -- I have scars from the days before we got package management and upgrade testing right | 18:17 |
| NobodyCam | lsmola_: node that example is using virtual power key NOT ironic power key | 18:17 |
| NobodyCam | s/node/note/ | 18:17 |
| *** CaptTofu has joined #tripleo | 18:17 | |
| lsmola_ | NobodyCam: ok, I just used code for starting node that is in Ironic doc | 18:18 |
| lsmola_ | NobodyCam: though it was warking for me last week | 18:18 |
| *** marun has quit IRC | 18:18 | |
| * NobodyCam looks | 18:18 | |
| hewbrocca | I'm willing to accept the argument that this is a constrained enough case that, with the right automation, maybe we can escape the horrific cost of that | 18:18 |
| lsmola_ | NobodyCam: ok, will try and will let you know tomorrow | 18:18 |
| ccrouch | SpamapS: this is the best I could find for upgrade docs: | 18:18 |
| ccrouch | https://etherpad.openstack.org/p/tripleo-image-updates | 18:18 |
| ccrouch | and | 18:18 |
| ccrouch | https://github.com/openstack/tripleo-incubator/blob/master/scripts/takeovernode | 18:18 |
| lsmola_ | NobodyCam: thank you for your help :-) | 18:18 |
| SpamapS | hewbrocca: Yeah, Debian developer and long time packager here too. | 18:19 |
| hewbrocca | SpamapS: It just makes me nervous is all | 18:19 |
| SpamapS | hewbrocca: There is a different horrific cost involved with doing things one tiny piece at a time. | 18:19 |
| athomas | SpamapS, Presumably, if the newer image contained updates to glibc, for example.. something where a restart of openstack services isn't going to be sufficient, there'd be a flag to say that a reboot was required.. | 18:19 |
| hewbrocca | SpamapS: yeah, absolutely agree | 18:19 |
| SpamapS | Which is that you are now on the hook for every piece of entropy that breaks every %post in some subtle way. | 18:19 |
| SpamapS | some user is not exactly the same.. | 18:20 |
| SpamapS | some binary has been moved out of the way.. | 18:20 |
| NobodyCam | lsmola_: actually not all of that is in the doc. most is... https://wiki.openstack.org/wiki/Ironic#Create_an_emulated_baremetal_node | 18:20 |
| SpamapS | somebody drops a script in /usr/local/bin... | 18:20 |
| SpamapS | we've all been there | 18:20 |
| SpamapS | and the chance and impact of entropy goes up with each server you are managing. | 18:20 |
| SpamapS | athomas: I've been thinking about that.. | 18:22 |
| SpamapS | athomas: My thought is that we should fail in os-refresh-config's post-configure.d if we detect any open deleted files on the root partition. | 18:23 |
| *** epim has joined #tripleo | 18:23 | |
| SpamapS | athomas: that would have a good chance at causing automated tests to fail before we try to deploy to prod and miss something. | 18:23 |
| lsmola_ | NobodyCam: yeah i used that one, seem like some mistake from my part | 18:24 |
| lsmola_ | NobodyCam: akrivoka has them up, she will paste you the result | 18:24 |
| akrivoka | http://paste.openstack.org/show/61702/ | 18:25 |
| NobodyCam | note that /me looks | 18:25 |
| lsmola_ | devananda: we don't see any n-cpu running | 18:25 |
| lsmola_ | devananda: what is the full name of the service? | 18:26 |
| NobodyCam | akrivoka: lsmola_: looks like properties options are not set: properties | {} | 18:26 |
| devananda | lsmola_: nova-compute | 18:26 |
| SpamapS | ccrouch: thanks for sharing those links. tripleo-image-updates is turning into "the spec" I guess :) | 18:26 |
| NobodyCam | see lines 4 5 and 6 from my paste | 18:26 |
| lsmola_ | devananda: systemctl | grep nova-compute shows nothing :-) | 18:27 |
| athomas | SpamapS, And when you catch an instance where there are open deleted files after the update, you've identified a case where a reboot is going to be required for the prod rollout? | 18:27 |
| devananda | lsmola_, NobodyCam - why would there be no nova-compute running in your udercloud? | 18:27 |
| lsmola_ | devananda: http://paste.openstack.org/show/61703/ | 18:28 |
| NobodyCam | ??? thats a good one not sure | 18:28 |
| lsmola_ | NobodyCam: ^^ list of services I have | 18:28 |
| SpamapS | ugh, Zimbra needs to fix their quoting. | 18:28 |
| SpamapS | They indent the whole message, leading to at least two other clients (RoundCube and sup-mail) seeing the entire reploy as folded into the previous quoted message. | 18:29 |
| lsmola_ | devananda: I have it in my seed http://paste.openstack.org/show/61704/ | 18:31 |
| lsmola_ | devananda: that is a mystery :-) | 18:32 |
| devananda | lsmola_: check /var/log/upstart to see if it failed | 18:33 |
| devananda | lsmola_: nova-compute should be running in your undercloud, configured with the ironic driver | 18:33 |
| lsmola_ | devananda: fedora :-) it should be listed in systemctl as failed, it is not | 18:34 |
| lsmola_ | devananda: it's like it's not even trying to start it | 18:34 |
| lsmola_ | devananda: http://paste.openstack.org/show/61705/ | 18:37 |
| *** noslzzp has quit IRC | 18:38 | |
| devananda | NobodyCam: ah ha ^ | 18:38 |
| NobodyCam | lsmola_: but you can do things like nova list? | 18:38 |
| lsmola_ | NobodyCam: yes | 18:39 |
| *** noslzzp has joined #tripleo | 18:39 | |
| NobodyCam | I think its running if nova list is working? no? | 18:39 |
| devananda | NobodyCam: don't we need to give nova the credentials to talk to ironic in some way? | 18:39 |
| lsmola_ | NobodyCam: not sure if that is job of compute | 18:40 |
| lifeless | o/ | 18:40 |
| devananda | NobodyCam: list is API. that traceback clearly shows nova-compute start, load the ironic driver, and fail becase it can't get list from ironic | 18:40 |
| lifeless | dprince: when is the hangout? wrong account, don't get g+ integration there :> | 18:40 |
| devananda | NobodyCam: fyi, any exception raised during n-cpu init will cause the process not to start. so things like keystone errors during list_instances .... | 18:41 |
| dprince | lifeless: hangout is in 1.5 hours, sorry about the email address. I've got your correct gmail address too | 18:41 |
| lsmola_ | devananda: oh I see, there is systemctl -a, yeah nova-compute is dead | 18:42 |
| lsmola_ | devananda: will have to run, let me know if you figure it out | 18:44 |
| lsmola_ | devananda: we are almost there :-) | 18:44 |
| lsmola_ | see you tomorrow guys | 18:44 |
| NobodyCam | lsmola_: did you correct the extra specs issue? | 18:45 |
| dprince | hewbrocca: re your comment about packages, I think they may prove to be useful in some cases with regards to updates. | 18:45 |
| NobodyCam | night lsmola_ | 18:45 |
| lsmola_ | NobodyCam: not yet, will try tomorrow | 18:45 |
| hewbrocca | dprince: saw your mail to that effect | 18:45 |
| lsmola_ | NobodyCam: btw. it should be set in the element, right? | 18:45 |
| NobodyCam | devananda: we do set that in nova.conf | 18:46 |
| hewbrocca | I agree but it also makes me nervous to have two different ways of pushing code and config onto images | 18:46 |
| NobodyCam | it... the extra specs.. yes | 18:46 |
| lsmola_ | NobodyCam: ok, will check tomorrow what is wrong with my element | 18:46 |
| hewbrocca | er onto machines | 18:46 |
| hewbrocca | sheesh | 18:47 |
| lsmola_ | night | 18:47 |
| NobodyCam | night | 18:47 |
| *** jprovazn has joined #tripleo | 18:50 | |
| dprince | lifeless: can you add me to TripleO on LP? I would love to be able to self-triage bugs.... | 18:50 |
| *** athomas has quit IRC | 18:50 | |
| *** e0ne has joined #tripleo | 18:52 | |
| *** jcoufal has joined #tripleo | 18:52 | |
| *** marun has joined #tripleo | 18:53 | |
| *** julim has quit IRC | 18:54 | |
| *** rpodolyaka1 has joined #tripleo | 18:54 | |
| *** julim has joined #tripleo | 18:57 | |
| lifeless | dprince: have you requested membership? I approve everyone I get notified about | 19:10 |
| dprince | lifeless: I have! | 19:12 |
| dprince | lifeless: it sat there so I figured I'd ask | 19:12 |
| * dprince misses those sort of emails all the time himself | 19:13 | |
| lifeless | done | 19:16 |
| dprince | lifeless: thanks sir | 19:16 |
| *** vipul is now known as vipul-away | 19:17 | |
| *** max_lobur is now known as max_lobur_afk | 19:19 | |
| *** edmund has joined #tripleo | 19:20 | |
| *** e0ne_ has joined #tripleo | 19:22 | |
| *** jprovazn has quit IRC | 19:22 | |
| *** vipul-away is now known as vipul | 19:23 | |
| *** spzala has joined #tripleo | 19:23 | |
| *** epim has quit IRC | 19:24 | |
| *** epim has joined #tripleo | 19:24 | |
| *** e0ne has quit IRC | 19:25 | |
| *** akrivoka has quit IRC | 19:29 | |
| *** marun has quit IRC | 19:36 | |
| *** jcoufal has quit IRC | 19:41 | |
| greghaynes | lifeless: Hey, got any reccomendations on an ok starter ticket/feature I could hammer at? | 19:43 |
| *** vkozhukalov has quit IRC | 19:45 | |
| greghaynes | Am poking away at some of the high prio ones for now, just curious if there was anything specific you had in mind | 19:50 |
| SpamapS | greghaynes: been added to the Trello board yet? https://trello.com/b/0jIoMrdo/tripleo | 19:54 |
| greghaynes | Nope | 19:54 |
| greghaynes | Was looking at it though :) | 19:54 |
| SpamapS | greghaynes: pm me the email you would want to use for trello and I'll ad you | 19:55 |
| SpamapS | add even | 19:55 |
| lifeless | greghaynes: hey | 19:55 |
| lifeless | starter tickets - IIRC you've a bunch of sysadmin experience right? | 19:55 |
| greghaynes | Yep, although that PMTU GRU one was scaring even me :p | 19:56 |
| greghaynes | er, GRE | 19:56 |
| lifeless | greghaynes: welcome to the dark zone | 19:58 |
| greghaynes | hah | 19:59 |
| greghaynes | SpamapS: Thank | 19:59 |
| greghaynes | thanks | 19:59 |
| lifeless | ok so | 20:00 |
| lifeless | right now we have two ig arcs | 20:00 |
| lifeless | *big* | 20:00 |
| lifeless | one is CI | 20:00 |
| lifeless | the other is stateful upgrades | 20:00 |
| lifeless | SpamapS: did you reenable tripleo-cd ? | 20:00 |
| greghaynes | ok | 20:00 |
| lifeless | dprince: is https://trello.com/c/pfY65aTl/42-openstack-elements-db-sync-on-each-deploy done ? | 20:01 |
| SpamapS | lifeless: no still pounding on my retry code over here. | 20:01 |
| dprince | lifeless: I thought so | 20:01 |
| SpamapS | lifeless: we should probably just re-enable it and set about debugging if it fails | 20:01 |
| SpamapS | s/if/when/ | 20:02 |
| dprince | lifeless: at least the first pass is. perhaps there is a second step that I missed? | 20:02 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Make nova state dirs using configure.d... https://review.openstack.org/68469 | 20:03 |
| SpamapS | lifeless: regarding the path to os-collect-config's cache.. | 20:03 |
| SpamapS | lifeless: I think we should consider removing it from os-apply-config. os-collect-config doesn't actually say "it will be here" | 20:03 |
| SpamapS | lifeless: and we can actually just have os-apply-config run 'os-collect-config --print-cachedir' if it can't find the list file via the environment/cli switches | 20:04 |
| SpamapS | lifeless: sound like a sane way forward? | 20:04 |
| * Ng returns from all the kiddie/dinner things and gets his reviewing trousers on | 20:05 | |
| lifeless | dprince: should move it to done then :) | 20:08 |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: set DIB_REPOLOCATION_<project> for projects https://review.openstack.org/68178 | 20:08 |
| lifeless | greghaynes: so https://trello.com/c/M4lQQmb8/40-images-move-persistent-state-locations-to-mnt-state-same-path-unconditionally-mnt-always-exists | 20:08 |
| lifeless | greghaynes: is a card with lots of little bits of work associated with it | 20:09 |
| *** rlandy has joined #tripleo | 20:09 | |
| lifeless | greghaynes: for instance, ng has an openssh branch in progress that you could collaborate with him on | 20:09 |
| greghaynes | sounds good | 20:11 |
| *** derekh has joined #tripleo | 20:11 | |
| * greghaynes pokes around etherpad notes | 20:11 | |
| Ng | I was thinking about that branch earlier and wondered about doing a version check on cloud-init, to decide whether to just edit sshd_config, or to also patch cloud-init | 20:11 |
| pleia2 | dprince: btw, there are a couple reviewday patches hanging out there if you could have a look in the near future :) | 20:12 |
| dprince | pleia2: cool. I'll have a look. thanks for pointing them out | 20:13 |
| SpamapS | Ng: did cloud-init release a version that is configurable yet? | 20:13 |
| SpamapS | Ng: also did you try out smoser's patch? Did you see the patch? | 20:13 |
| SpamapS | Ng: was rather nice of him to respond so quickly. | 20:13 |
| lifeless | greghaynes: have you muddled through devtest ? | 20:13 |
| lifeless | greghaynes: and have you started doing code review? | 20:14 |
| Ng | SpamapS: it was nice of him, yeah. I've seen it, but not yet tested | 20:14 |
| greghaynes | Been messing with devtest, ran into the known heat issue waiting for merge and was just trying to cherry pick out the change | 20:15 |
| dprince | pleia2/lifeless/derekh/rlandy: meeting time | 20:15 |
| SpamapS | oh.. yeah I didn't even think of that. Go through devtest once. Its like our version of the 108 wooden men of the shao lin temple :) | 20:15 |
| greghaynes | Havent reviewed, is there anything I need to be added to for that? | 20:15 |
| SpamapS | at the end you deploy an instance and it brands your arms with dragons. ;) | 20:15 |
| greghaynes | haha | 20:16 |
| greghaynes | I can tell, been learning a ton about the project just by getting it to go :p | 20:17 |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/os-collect-config: Move default cache dir to persistent location https://review.openstack.org/67323 | 20:17 |
| SpamapS | greghaynes: https://wiki.openstack.org/wiki/HowToContribute <-- | 20:18 |
| SpamapS | greghaynes: in case you hadn't seen that yet | 20:18 |
| greghaynes | Yep, got a trivial merge in yesterday to sort of figure that one out :) | 20:18 |
| *** vipul is now known as vipul-away | 20:19 | |
| *** vipul-away is now known as vipul | 20:19 | |
| SpamapS | greghaynes: then you should be ready to review. :) | 20:19 |
| greghaynes | woo! | 20:19 |
| SpamapS | greghaynes: just make sure you watch all the tripleo projects | 20:19 |
| greghaynes | ok, think clarkb had me do that before I knew the implications | 20:20 |
| Ng | the TripleO wiki page has a link at the bottom that takes you to gerrit with all the projects showing | 20:20 |
| greghaynes | aha | 20:20 |
| *** lucasagomes is now known as lucas-afk | 20:21 | |
| *** CaptTofu has quit IRC | 20:26 | |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/tripleo-image-elements: Make os-collect-config preserve state. https://review.openstack.org/67329 | 20:28 |
| SpamapS | greghaynes: btw, to cherry pick the heat change in, the way I do it is to git review -d it into my local clone of heat, and then run devtest with this: | 20:28 |
| SpamapS | IB_REPOLOCATION_heat="/home/clint/src/heat" DIB_REPOREF_heat="review/steven_hardy/bug/1271190" | 20:28 |
| SpamapS | DIB_REPOLOCATION_heat="/home/clint/src/heat" DIB_REPOREF_heat="review/steven_hardy/bug/1271190" | 20:28 |
| SpamapS | Ng: weren't you doing something to make it even easier to cherry pick gerrit fixes into a DIB run? | 20:29 |
| greghaynes | Nice, was wondering how to make it use local repo. I was doing the repolocation = https://gitrepo | 20:29 |
| greghaynes | Now, seems like I should just have to re- boot-seed-vm right? | 20:30 |
| Ng | SpamapS: yep, it's just what you said, but with the REPOLOCATIOn set to https://review.openstack.org/openstack/heat | 20:30 |
| Ng | the improvement was in the REPROREF part, it wouldn't previously have grokked review/blah/lol/12 | 20:31 |
| *** akuznetsov has quit IRC | 20:31 | |
| greghaynes | Theres a blurb in a readme in dib-elements about how to make it use a gerrit change for a repolocation :) | 20:31 |
| greghaynes | Woo for docs | 20:31 |
| Ng | greghaynes: there is indeed :) | 20:31 |
| SpamapS | Ng: OH so it is just smart enough to gerrit that now? cool. | 20:32 |
| *** markmc has quit IRC | 20:32 | |
| SpamapS | oh.. docs.. | 20:32 |
| SpamapS | what a concept | 20:32 |
| *** CaptTofu has joined #tripleo | 20:33 | |
| *** vipul is now known as vipul-away | 20:34 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Make neutron-openvswitch-agent run cleanup. https://review.openstack.org/68261 | 20:39 |
| *** vipul-away is now known as vipul | 20:51 | |
| *** ProfFalk1n has joined #tripleo | 20:54 | |
| ProfFalk1n | hey all, anyone around who can help me with a Toci install? | 20:54 |
| ProfFalk1n | hey all, anyone around who can help me with a Toci install? | 20:55 |
| openstackgerrit | Ralf Haferkamp proposed a change to openstack/diskimage-builder: Make copy_required_libs() more robust https://review.openstack.org/68434 | 20:55 |
| ProfFalk1n | I'm trying to install it on Ubuntu 12.04LTS and it's now saying that openvswitch isn't installed, which package is required? Just the "common" aspect? | 20:56 |
| Ng | SpamapS: stop sending mails on the tripleo update thread just before I've finished writing ostensibly the same email! | 20:58 |
| Ng | or maybe, do keep doing that, and I'll go back to testing smoser's branch ;) | 20:58 |
| Ng | ProfFalk1n: fwiw I think you may struggle on 12.04, I don't know if any of us are using anything older than Raring, but maybe the toci folk are being a bit more conservative and have snuck some fixes in for 12.04 compatibility | 20:59 |
| ProfFalk1n | Ng: heh, doesn't look like it, you seem to need precise backports enabled here | 20:59 |
| ProfFalk1n | I'm trying to fix it up now | 20:59 |
| ProfFalk1n | (but in such a way that it won't break anything!)# | 21:00 |
| ProfFalk1n | :D | 21:00 |
| cody-somerville | mordred: lifeless: Conference line is open. | 21:00 |
| *** CaptTofu has quit IRC | 21:01 | |
| SpamapS | Ng: haha I will step back from it so that some other voice can be heard. Seriously... I won't respond for 3 hours. | 21:01 |
| ProfFalk1n | hrm, looks like it defo requires root to run though... :( | 21:01 |
| SpamapS | in fact I think I need to shut down email for a bit so I can get some actual reviews done | 21:01 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Setup ccache symlinks on openSUSE https://review.openstack.org/67965 | 21:03 |
| lifeless | cody-somerville: ruh roh - be right there | 21:08 |
| derekh | ProfFalk1n: the only person using toci lately is dprince and he is in the process of moving over to devtest, if you setting up a local tripleo devtest would be better to look at | 21:15 |
| derekh | ProfFalk1n: toci is going to become more CI fucosed | 21:16 |
| derekh | ProfFalk1n: and a lot of the devtest type stuff is going to be taken out | 21:16 |
| ProfFalk1n | derekh: oh, ok, thanks. I'll look at devtest as well, but as an exercise it could still be useful. | 21:16 |
| dprince | ProfFalk1n: yes, soon I aim to rip out all of the old TOCI scripts to avoid confusion. Just doing a final parity sync this week and then we'll be gone with several of the older scripts. | 21:17 |
| ProfFalk1n | I'm working on a patch to the README.md on how to set it up under Ubuntu 12.04LTS for those of us stuck on an older version for support reasons etc | 21:17 |
| ProfFalk1n | should I not bother? | 21:17 |
| ProfFalk1n | or will it still be useful? | 21:17 |
| *** rlandy has quit IRC | 21:19 | |
| *** CaptTofu has joined #tripleo | 21:19 | |
| *** dprince has quit IRC | 21:19 | |
| clarkb | I thought tripleo can't run on old ubuntu because everything is too old | 21:19 |
| derekh | ProfFalk1n: to be honest you could but most of what toci.sh does is now covered by devtest so is going to get ripped out , the new entry point will be toci_devtest.sh which is designed to run CI and wont be helpfull unless you set up a baremetal cloud to run it against | 21:22 |
| *** rpodolyaka1 has quit IRC | 21:22 | |
| ProfFalk1n | clarkb: I was doing well until I hit | 21:22 |
| ProfFalk1n | /opt/toci/tripleo-incubator/scripts/boot-seed-vm -a i386 pip-cache fedora selinux-permissive neutron-dhcp-agent | 21:22 |
| ProfFalk1n | Error: TE_DATAFILE not set. | 21:22 |
| Ng | clarkb: it's not everything. it's at least one thing is too old, but none of us (afaik) are testing it, so it could now be that many things are old | 21:22 |
| Ng | it was like libvirt or similar | 21:22 |
| *** lblanchard2 has quit IRC | 21:22 | |
| ProfFalk1n | derekh: so given that I need to get this running on 12.04LTS, i'd be better off concentrating on see if I can get devtest to run on 12.04 instead? | 21:23 |
| derekh | ProfFalk1n: I think so yes, | 21:23 |
| ProfFalk1n | ok, cool, I'll give it a spin. thanks. | 21:24 |
| * ProfFalk1n deletes his bug and cancels the patchset | 21:24 | |
| *** lucas-afk has quit IRC | 21:25 | |
| derekh | ProfFalk1n: sorry | 21:25 |
| *** lblanchard has joined #tripleo | 21:25 | |
| ProfFalk1n | np | 21:26 |
| ProfFalk1n | :D | 21:26 |
| ProfFalk1n | so to configure dev-test I clone tripleo-incubator and run from there? | 21:27 |
| Ng | yup | 21:28 |
| ProfFalk1n | ok, thanks, running that now. Will report back if I hit any issues... :D | 21:29 |
| *** ProfFalk1n has quit IRC | 21:30 | |
| clarkb | Ng: ovs too iirc | 21:32 |
| SpamapS | If you aren't literally deploying the cloud _today_.. there's little point in dragging 12.04 along | 21:41 |
| SpamapS | just grab trusty isos | 21:41 |
| SpamapS | by the time you get your cloud up it will be released | 21:42 |
| ccrouch | any chance anyone could review https://review.openstack.org/#/c/68274/ | 22:02 |
| ccrouch | we would love to keep the stable branches stuff moving along | 22:02 |
| *** derekh has quit IRC | 22:03 | |
| *** ftcjeff has quit IRC | 22:06 | |
| *** ftcjeff has joined #tripleo | 22:07 | |
| *** ColinTaylor has joined #tripleo | 22:08 | |
| *** akrivoka has joined #tripleo | 22:10 | |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-incubator: Document stable branches in devtest https://review.openstack.org/68274 | 22:12 |
| openstackgerrit | Dirk Mueller proposed a change to openstack/diskimage-builder: Workaround broken udev update on openSUSE https://review.openstack.org/68508 | 22:12 |
| ccrouch | thanks lifeless! | 22:13 |
| *** matty_dubs is now known as matty_dubs|gone | 22:17 | |
| bnemec | For the stable branches, do we want to generate an environment file with the DIB_REPO* vars that will point all of the git repos at the appropriate stable branches in the other projects? | 22:22 |
| slagle | bnemec: i was thinking i would add that step in the stable branch for incubator itself | 22:23 |
| bnemec | slagle: Okay, sounds good. | 22:24 |
| slagle | so, devtest from the stable branch would tell you to set DIB_REPO* (or likely just do it) | 22:24 |
| slagle | likewise, pull-tools wouls use the stable branches | 22:24 |
| slagle | *would | 22:24 |
| *** ftcjeff has quit IRC | 22:24 | |
| SpamapS | slagle: we need a pbr plugin for devtest scripts ;) | 22:25 |
| *** lblanchard has quit IRC | 22:26 | |
| *** taps has quit IRC | 22:27 | |
| slagle | SpamapS: that would update that stuff automatically? | 22:27 |
| *** jdob has quit IRC | 22:27 | |
| openstackgerrit | Ben Nemec proposed a change to openstack/tripleo-image-elements: Enable the nova v3 api https://review.openstack.org/68514 | 22:28 |
| SpamapS | slagle: I don't know if pbr actually has plugins | 22:29 |
| SpamapS | slagle: but thats basically what pbr's job is.. to make sure you don't have to commit any version stuff | 22:30 |
| SpamapS | slagle: we may just want to expand on that idea and have some conditional logic that displays things in the docs when the current commit has a tag. | 22:31 |
| slagle | cool, makes sense | 22:31 |
| lifeless | is fedora 20 releases? | 22:34 |
| lifeless | *released* ? | 22:34 |
| slagle | yes | 22:34 |
| *** sdake is now known as sdake-ooo | 22:37 | |
| *** jtomasek has quit IRC | 22:55 | |
| *** jayg|g0n3 has joined #tripleo | 23:02 | |
| *** CaptTofu has quit IRC | 23:04 | |
| *** akrivoka has quit IRC | 23:11 | |
| *** akrivoka has joined #tripleo | 23:12 | |
| *** sballe has quit IRC | 23:22 | |
| *** sballe has joined #tripleo | 23:22 | |
| *** sballe_ has joined #tripleo | 23:34 | |
| *** sballe has quit IRC | 23:35 | |
| *** taps has joined #tripleo | 23:42 | |
| StevenK | lifeless: Can I have an example heat template for defaults for my debian-mirror branch? | 23:51 |
| lifeless | StevenK: sure. SpamapS ^ | 23:52 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!