| *** dina_belova has joined #openstack-infra | 00:04 | |
| *** dolphm has quit IRC | 00:05 | |
| *** sgviking has quit IRC | 00:05 | |
| *** sgviking has joined #openstack-infra | 00:06 | |
| clarkb | woot I have git review tests working against gerrit 2.4 | 00:06 |
|---|---|---|
| clarkb | I think they will work against 2.6 as well | 00:06 |
| clarkb | now running with --until-failure to see if it is any more reliable | 00:07 |
| *** pcrews has quit IRC | 00:08 | |
| *** dina_belova has quit IRC | 00:09 | |
| *** jfriedly has quit IRC | 00:11 | |
| *** w_ has quit IRC | 00:14 | |
| *** olaph has joined #openstack-infra | 00:14 | |
| clarkb | jeblair: I think we are seeing an odd thing happen wiht the test load. There is a long weekend coming up so the test load is falling rather than climbing despite being only a few days from feature freeze | 00:14 |
| *** UtahDave has quit IRC | 00:14 | |
| *** michchap has joined #openstack-infra | 00:18 | |
| *** jhesketh_ has quit IRC | 00:18 | |
| *** sarob_ has quit IRC | 00:18 | |
| *** sarob has joined #openstack-infra | 00:19 | |
| *** michchap has quit IRC | 00:19 | |
| *** michchap has joined #openstack-infra | 00:20 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/git-review: Use OpenStack's Gerrit war when testing git-review https://review.openstack.org/44378 | 00:23 |
| *** sarob has quit IRC | 00:23 | |
| clarkb | let's see if that helps | 00:24 |
| clarkb | nope http://logs.openstack.org/78/44378/1/check/gate-git-review-python27/1460032/console.html I really do not know why that is happening | 00:26 |
| clarkb | (I think there is additional value in running the tests against our war so not completely wasted effort) | 00:26 |
| openstackgerrit | Li Yingjun proposed a change to openstack/requirements: Bump python-cinderclient requirement to >=1.0.5 https://review.openstack.org/44379 | 00:29 |
| *** zul has joined #openstack-infra | 00:36 | |
| *** jhesketh_ has joined #openstack-infra | 00:36 | |
| echohead | pleia2: thanks for the pointer. not sure how the default_data.json is populated - if nothing else, i guess one could send a patch to update it. | 00:38 |
| jeblair | clarkb: yeah, it seems like everyone has already left. probably because the bay bridge is closed. :) | 00:41 |
| jeblair | and on that note, i'm going afk for the evening | 00:42 |
| clarkb | jeblair: good night | 00:42 |
| *** jhesketh__ has joined #openstack-infra | 00:42 | |
| clarkb | I am about to do the same. It time for a break from the git-review and gerrit bashing | 00:42 |
| *** HenryG has joined #openstack-infra | 00:43 | |
| *** gyee has quit IRC | 00:45 | |
| *** w_ has joined #openstack-infra | 00:47 | |
| *** olaph has quit IRC | 00:50 | |
| *** dkliban has quit IRC | 00:51 | |
| *** olaph has joined #openstack-infra | 00:52 | |
| *** [wchrisj_ has quit IRC | 00:54 | |
| *** w_ has quit IRC | 00:55 | |
| *** svarnau has quit IRC | 01:02 | |
| *** dkranz has joined #openstack-infra | 01:03 | |
| *** dina_belova has joined #openstack-infra | 01:05 | |
| *** zul has quit IRC | 01:06 | |
| *** w_ has joined #openstack-infra | 01:06 | |
| *** pcrews has joined #openstack-infra | 01:06 | |
| *** olaph has quit IRC | 01:09 | |
| *** dina_belova has quit IRC | 01:10 | |
| *** olaph has joined #openstack-infra | 01:10 | |
| *** w_ has quit IRC | 01:11 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make config updates with with rebased branches. https://review.openstack.org/44383 | 01:15 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make config updates work with rebased branches. https://review.openstack.org/44383 | 01:17 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Install Nova and Cinder clients on puppetmaster. https://review.openstack.org/44384 | 01:20 |
| *** dims has quit IRC | 01:21 | |
| *** portante|afk is now known as portante | 01:22 | |
| *** woodspa has quit IRC | 01:23 | |
| *** dims has joined #openstack-infra | 01:24 | |
| *** sandywalsh has quit IRC | 01:27 | |
| *** w_ has joined #openstack-infra | 01:28 | |
| *** olaph has quit IRC | 01:31 | |
| *** [wchrisj_ has joined #openstack-infra | 01:32 | |
| *** Ryan_Lane has quit IRC | 01:45 | |
| *** olaph has joined #openstack-infra | 01:45 | |
| lifeless | hmm | 01:45 |
| lifeless | err: /Stage[main]/Testcabal_project::Puppetmaster/Package[python-cinderclient]: Provider pip is not functional on this host | 01:45 |
| *** sandywalsh has joined #openstack-infra | 01:45 | |
| lifeless | ^ pointers for that? [ clarkb, pleia2 et al] | 01:46 |
| uvirtbot | lifeless: Error: "clarkb," is not a valid command. | 01:46 |
| *** w_ has quit IRC | 01:47 | |
| *** dkliban has joined #openstack-infra | 01:47 | |
| *** jhesketh_ has quit IRC | 01:49 | |
| *** jhesketh__ has quit IRC | 01:49 | |
| clarkb | lifeless: is pip installed? | 01:50 |
| *** w_ has joined #openstack-infra | 01:50 | |
| clarkb | does pip --version work? | 01:50 |
| lifeless | # pip --version | 01:52 |
| lifeless | pip 1.4.1 from /usr/local/lib/python2.7/dist-packages (python 2.7) | 01:52 |
| lifeless | clarkb: ^ | 01:52 |
| *** olaph has quit IRC | 01:52 | |
| clarkb | is pip installed in a weird place so puppet can't find it? | 01:53 |
| clarkb | I think the pip is not function means puppet can't find it or running it doesn't work | 01:53 |
| clarkb | the pip --version indicates it does work | 01:53 |
| lifeless | # which pip | 01:55 |
| lifeless | /usr/local/bin/pip | 01:55 |
| lifeless | I didn't install pip AFAIK | 01:55 |
| lifeless | ii python-pip 1.0-1build1 alternative Python package installer | 01:55 |
| lifeless | there is something odd - dpkg -L pip suggests /usr/bin/pip is there. | 01:57 |
| clarkb | lifeless: that is because pip removes it when pip upgrades... | 01:58 |
| clarkb | its a bit wrong | 01:58 |
| lifeless | anyhow | 01:58 |
| lifeless | pip is there | 01:58 |
| clarkb | I would have to go look in the pip puppet privder to see where that errors comes from | 01:59 |
| lifeless | ewww | 01:59 |
| clarkb | https://github.com/puppetlabs/puppet/blob/2.7.x/lib/puppet/provider/package/pip.rb | 02:00 |
| clarkb | that error message is not in that file | 02:00 |
| clarkb | what version of puppet? | 02:00 |
| *** jhesketh_ has joined #openstack-infra | 02:01 | |
| lifeless | https://groups.google.com/forum/#!msg/puppet-dev/R8VNyORGW3M/YkuEPKT55LsJ | 02:02 |
| lifeless | gnar 'useful error message' [which no longer shows what went wrong] | 02:02 |
| *** jhesketh__ has joined #openstack-infra | 02:02 | |
| lifeless | clarkb: # puppet --version | 02:02 |
| lifeless | 2.7.23 | 02:02 |
| *** w_ has quit IRC | 02:02 | |
| *** w_ has joined #openstack-infra | 02:04 | |
| clarkb | I'm confused that doesn't show up in the github source | 02:04 |
| *** changbl has joined #openstack-infra | 02:05 | |
| *** dina_belova has joined #openstack-infra | 02:06 | |
| clarkb | oh derp | 02:06 |
| clarkb | I was searching the actual message. If I read the diff I see I am derping | 02:06 |
| clarkb | lifeless: is /usr/local/bin in puppet/root's path? | 02:07 |
| lifeless | I'm root atm, and it's in my path | 02:08 |
| lifeless | let me log directly in | 02:08 |
| lifeless | pleia2: replied to your review | 02:08 |
| lifeless | oh, I can't. Uhm.. | 02:09 |
| lifeless | (login as root that is, PermitRootLogin no) | 02:10 |
| *** dina_belova has quit IRC | 02:10 | |
| *** yaguang has joined #openstack-infra | 02:11 | |
| lifeless | clarkb: and AFAICT in puppets too | 02:11 |
| lifeless | I'll symlink /usr/bin/pip though and try again | 02:11 |
| lifeless | clarkb: pip removing the system packaged pip is so evil though | 02:12 |
| clarkb | lifeless: it is | 02:12 |
| lifeless | clarkb: and pointless, since dpkg updates will reinstate it | 02:12 |
| lifeless | dstufft: ^ | 02:12 |
| lifeless | clarkb: once the puppetmaster is setup, do you need the full puppet apply --modulepath='/opt/config/production/modules:/etc/puppet/modules' -e 'include testcabal_project::puppetmaster' thing? | 02:13 |
| *** dkliban has quit IRC | 02:13 | |
| lifeless | clarkb: or is there a shorter command that can be used ? | 02:13 |
| clarkb | once puppet master is running you can run puppet in agent mode instead of apply mode (apply is purely local hence the need for pointing out modules and stuff) | 02:14 |
| clarkb | lifeless: puppet agent --test will run it in agent master mode once | 02:14 |
| clarkb | and give you nice debug output | 02:14 |
| lifeless | clarkb: on the master right ? | 02:15 |
| clarkb | lifeless: wherever you want to apply the puppet configs | 02:15 |
| lifeless | it's hung, I think... | 02:15 |
| lifeless | root@ci-puppetmaster:/opt/config/production# puppet agent --test | 02:15 |
| lifeless | info: Creating a new SSL key for undef | 02:15 |
| lifeless | info: Creating a new SSL certificate request for undef | 02:15 |
| lifeless | info: Certificate Request fingerprint (md5): 8B:C3:97:C3:9B:F4:A2:B8:2A:49:56:D3:6E:FB:56:BE | 02:15 |
| lifeless | Exiting; no certificate found and waitforcert is disabled | 02:15 |
| clarkb | oh Ithink you need to edit your /etc/puppet/puppet.conf | 02:16 |
| clarkb | to tell it what the certname is and which master to talk to | 02:16 |
| lifeless | clarkb: this is the master | 02:16 |
| clarkb | the puppet apply you run the first time should do that | 02:16 |
| clarkb | lifeless: right, but the master is also a client | 02:16 |
| lifeless | the apply I run the first time to setup the master doesn't do that? | 02:16 |
| *** olaph has joined #openstack-infra | 02:16 | |
| clarkb | I think it should | 02:16 |
| clarkb | though we may be doing it in a very openstack specific part of the code | 02:17 |
| lifeless | hmm, server=ci-puppetmaster.openstack.org | 02:17 |
| lifeless | certname=undef | 02:17 |
| lifeless | huh, I appear to have an unfixed file here | 02:17 |
| * lifeless digs | 02:17 | |
| clarkb | http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/manifests/base.pp#n94 | 02:18 |
| *** w_ has quit IRC | 02:18 | |
| lifeless | yes, failed to update one of the references | 02:18 |
| lifeless | in modules/testcabal_project/manifests/base.pp | 02:19 |
| lifeless | the symlink worked | 02:22 |
| clarkb | :/ | 02:22 |
| lifeless | indeed | 02:22 |
| clarkb | I wonder if puppte which is special and not using PATH? | 02:23 |
| lifeless | certname=undef | 02:23 |
| *** vipul has quit IRC | 02:23 | |
| lifeless | still; but the right server now | 02:23 |
| lifeless | and wheee boom | 02:23 |
| openstackgerrit | A change was merged to openstack/requirements: bump keystoneclient to 0.3.2 https://review.openstack.org/43919 | 02:24 |
| lifeless | clarkb: n# puppet agent --test | pastebinit | 02:24 |
| lifeless | http://paste.ubuntu.com/6042559/ | 02:24 |
| *** vipul has joined #openstack-infra | 02:24 | |
| clarkb | lifeless: looks like you may need to chmod the puppet reports dir | 02:26 |
| clarkb | line 133 in the paste | 02:26 |
| lifeless | https://bugs.launchpad.net/openstack-ci/+bug/1218696 | 02:27 |
| uvirtbot | Launchpad bug 1218696 in openstack-ci "new puppetmaster ends up with broken pip provider" [Undecided,New] | 02:27 |
| *** anteaya has quit IRC | 02:27 | |
| lifeless | https://bugs.launchpad.net/openstack-ci/+bug/1218697 | 02:28 |
| uvirtbot | Launchpad bug 1218697 in openstack-ci "bad mode on reports directory" [Undecided,New] | 02:28 |
| lifeless | clarkb: how does one express 'set a mode' in puppet ? | 02:28 |
| lifeless | clarkb: or should reports be puppet:puppet rather than root:root ? | 02:29 |
| clarkb | lifeless: with the file resource http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/manifests/jenkins.pp#n158 | 02:29 |
| lifeless | clarkb: also I love the error 'change from X to Y; failed to set mode X..' WTF | 02:30 |
| clarkb | yeah that message isn't so great | 02:30 |
| openstackgerrit | A change was merged to openstack/requirements: Bump python-cinderclient requirement to >=1.0.5 https://review.openstack.org/44379 | 02:30 |
| lifeless | looks like this is not master specific, right? | 02:31 |
| lifeless | so I should do this in base? or server? | 02:31 |
| *** [wchrisj_ has quit IRC | 02:31 | |
| clarkb | lifeless: reports is master specific | 02:31 |
| clarkb | lifeless: I think passenger/puppet is trying to make sure it can write/read into the reports dir | 02:32 |
| lifeless | clarkb: it's cutting -a off | 02:32 |
| clarkb | oh, so being defensive | 02:32 |
| clarkb | in any case that is master specific (possibly also dashboard specific because you can send reports there too) | 02:33 |
| lifeless | clarkb: ls -ld /var/lib/puppet/reports/ | 02:33 |
| lifeless | drwxr-xr-x 4 root root 4096 Aug 30 01:39 /var/lib/puppet/reports/ | 02:33 |
| lifeless | clarkb: is it 'dir' or 'directory' to assert a dir exists? | 02:33 |
| clarkb | directory | 02:33 |
| lifeless | like http://paste.ubuntu.com/6042568/ ? | 02:34 |
| clarkb | lifeless: s/directory/file/ s/present/directory/ | 02:34 |
| lifeless | http://paste.ubuntu.com/6042570/ ? | 02:35 |
| clarkb | yup | 02:35 |
| *** dkliban has joined #openstack-infra | 02:36 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Ensure reports directory is setup correctly. https://review.openstack.org/44391 | 02:37 |
| lifeless | we probably want some of this stack to land :> | 02:37 |
| lifeless | clarkb: warning: Could not retrieve fact fqdn | 02:40 |
| clarkb | lifeless: oh man this is the best thing ever | 02:40 |
| lifeless | clarkb: that will explain the undef on certname | 02:40 |
| lifeless | clarkb: what is? | 02:40 |
| clarkb | lifeless: so facter is actually pretty dumb when it comes to figuring out an fqdn | 02:40 |
| clarkb | if your domain is '.' eg your hostname is the TLD then facter undefs fqdn | 02:41 |
| clarkb | because a hostname is apparently not an fqdn when anchored under '.' | 02:41 |
| lifeless | # puppet agent --test | pastebinit | 02:41 |
| lifeless | http://paste.ubuntu.com/6042589/ | 02:41 |
| clarkb | lifeless: I think you may need to generate a server cert with the faux fqdn | 02:42 |
| clarkb | lifeless: or use the name iwthout a domain in your server specification | 02:42 |
| lifeless | I've just reloaded hostname - I'd edited the hostname file but not kicked the upstart job | 02:42 |
| clarkb | that should do it | 02:42 |
| clarkb | so annoying | 02:42 |
| lifeless | I'll re-run the apply | 02:43 |
| clarkb | facter's code explicitly does something like if not domain: fqdn = None | 02:43 |
| clarkb | which is wrong... | 02:43 |
| lifeless | ok, I now have a valid certname in puppet.conf | 02:43 |
| clarkb | but it has been argued a couple times before and it never went anywhere so I just work around it now if I have to | 02:43 |
| lifeless | and more notes to roll into docs/bugfixes. | 02:44 |
| *** melwitt has quit IRC | 02:44 | |
| lifeless | puppet agent --test | 02:44 |
| lifeless | info: Creating a new SSL key for ci-puppetmaster.testing-cabal.org | 02:44 |
| lifeless | info: Creating a new SSL certificate request for ci-puppetmaster.testing-cabal.org | 02:44 |
| lifeless | info: Certificate Request fingerprint (md5): D5:B0:14:60:AF:86:1E:37:0E:53:B8:1E:A2:54:56:AF | 02:44 |
| lifeless | Exiting; no certificate found and waitforcert is disabled | 02:44 |
| lifeless | this is for the puppetmaster on the puppetmaster | 02:44 |
| lifeless | after re-running the whole ' | 02:44 |
| lifeless | puppet apply --modulepath='/opt/config/production/modules:/etc/puppet/modules' -e 'include testcabal_project::puppetmaster' | 02:44 |
| lifeless | command | 02:44 |
| *** jfriedly has joined #openstack-infra | 02:45 | |
| clarkb | lifeless: so now I think you need to sign the cert | 02:46 |
| openstackgerrit | A change was merged to openstack-infra/config: Fix semantic error in example config. https://review.openstack.org/44000 | 02:46 |
| clarkb | puppetca --list && puppetca --sign $HOSTNAME | 02:46 |
| openstackgerrit | A change was merged to openstack-infra/config: Correct launch README. https://review.openstack.org/44154 | 02:46 |
| lifeless | so how does launch.py do this automatically? | 02:46 |
| openstackgerrit | A change was merged to openstack-infra/config: Make root-only information clear in sysadmin.rst. https://review.openstack.org/44162 | 02:46 |
| lifeless | I see the steps in puppet.rst | 02:46 |
| *** w_ has joined #openstack-infra | 02:47 | |
| clarkb | lifeless: launch.py pregenerates and signs the cert, copies it to where it needs to be on the puppet master, and copies it to where it needs ot be on the client | 02:47 |
| clarkb | sorry not launch.py | 02:47 |
| lifeless | ok | 02:47 |
| clarkb | this is a manual step in the readme | 02:47 |
| clarkb | then you pass the name of the cert to launch.py so that it can copy it to the client | 02:47 |
| lifeless | I am wondering if we can do the puppetmaster cert for the puppetmaster automatically? | 02:47 |
| clarkb | lifeless: possibly since you could run all of the commands to do that on the master itself with puppet | 02:48 |
| clarkb | lifeless: in https://review.openstack.org/#/c/44163/3 did you save your reply to my -1? | 02:48 |
| clarkb | lifeless: you mention it to pleia2 but I don't see it | 02:48 |
| lifeless | oh gerrit I hate thee | 02:49 |
| lifeless | clarkb: done | 02:49 |
| clarkb | thansk | 02:49 |
| lifeless | on patch set 1 | 02:49 |
| *** olaph has quit IRC | 02:50 | |
| clarkb | lifeless: changing that is tricky, because we'd have to detect whether or not git-http is being used | 02:50 |
| clarkb | possibly by checking headers? | 02:50 |
| clarkb | because we only want non git-http to go to /cgit/* | 02:50 |
| *** adalbas has quit IRC | 02:51 | |
| lifeless | clarkb: I believe the urls are entirely distinct | 02:52 |
| lifeless | clarkb: there is the magic url for the smart thing, and then the .git/ contents | 02:52 |
| lifeless | clarkb: all other paths are cgit | 02:52 |
| clarkb | if you point your browser at the url you give in that doc it won't do you much good | 02:53 |
| lifeless | clarkb: if you point git at it , it won't do much good. | 02:53 |
| lifeless | clarkb: the prose surrounding it says 'a clone of' | 02:53 |
| *** DennyZhang has joined #openstack-infra | 02:54 | |
| clarkb | lifeless: yeah I am fine with the content after your comment | 02:54 |
| lifeless | clarkb: I could format it differently, just put it at ``...`` instead? | 02:54 |
| *** w_ has quit IRC | 02:54 | |
| clarkb | lifeless: just explaining that fixing it is a little tricky | 02:54 |
| lifeless | clarkb: ack | 02:54 |
| lifeless | clarkb: ugh - err: /File[/var/lib/puppet/lib]: Failed to generate additional resources using 'eval_generate: Server hostname 'ci-puppetmaster.testing-cabal.org' did not match server certificate; expected ci-puppetmaster | 02:55 |
| lifeless | is back | 02:55 |
| clarkb | did something change your hostname back? | 02:55 |
| lifeless | # hostname -f | 02:55 |
| lifeless | ci-puppetmaster.testing-cabal.org | 02:55 |
| lifeless | no | 02:55 |
| clarkb | I wonder if the server cert was generated with the old name | 02:56 |
| clarkb | (there are two certs at play here for mutual auth and identification, the server cert and the client cert) | 02:56 |
| clarkb | I think | 02:56 |
| lifeless | the server cert is probably wrong | 02:56 |
| clarkb | that is what I am thinking. I believe there is a puppetca command to regenerate /me looks | 02:57 |
| clarkb | http://projects.puppetlabs.com/projects/1/wiki/certificates_and_security#Regenerate-Certificates-for-Puppet-Master | 02:57 |
| lifeless | what does 'stop puppet master' mean ? | 02:58 |
| lifeless | clarkb: any reason base doesn't set puppet to start on boot ? | 02:58 |
| clarkb | lifeless: in this case stop apache so that the passenger processes stop | 02:59 |
| clarkb | lifeless: yes, our slaves do not run puppet as a daemon and use cron instead. Also, we often want a little more control of what puppet does when first building a new host | 02:59 |
| lifeless | clarkb: the puppet.rst doc says to change e/d/p to start on boot | 02:59 |
| lifeless | clarkb: is the doc stale? | 03:00 |
| clarkb | lifeless: no, its fine we just don't want to do it automagically | 03:00 |
| lifeless | clarkb: so, it *should* be set on on the puppetmaster? | 03:00 |
| clarkb | yes | 03:01 |
| clarkb | the puppetmaster runs a client that talk to the master | 03:01 |
| lifeless | clarkb: does launch.py make it start on boot ? | 03:01 |
| clarkb | it does not | 03:01 |
| clarkb | README notes this near EOF | 03:02 |
| lifeless | what does 'Restart the puppet master.' mean here? | 03:03 |
| clarkb | lifeless: in the regenerate certificates for puppet master page? | 03:03 |
| clarkb | lifeless: start apache again | 03:04 |
| lifeless | yes, thanks | 03:04 |
| clarkb | the master is a set of processes that run as apache workers | 03:04 |
| lifeless | should puppetqd be running? | 03:04 |
| clarkb | lifeless: no I don't think so | 03:04 |
| clarkb | there should be some passenger/puppet processes | 03:05 |
| clarkb | but they aren't daemons on their own | 03:05 |
| lifeless | ok, so service puppet start, rather than rebooting, after setting up the pupet master ? | 03:05 |
| lifeless | still getting info: Retrieving plugin | 03:05 |
| lifeless | err: /File[/var/lib/puppet/lib]: Failed to generate additional resources using 'eval_generate: Server hostname 'ci-puppetmaster.testing-cabal.org' did not match server certificate; expected ci-puppetmaster | 03:05 |
| clarkb | service puppet start runs the client | 03:06 |
| lifeless | right, and you said /e/d/puppet should say yes to start on boot | 03:06 |
| clarkb | yup | 03:06 |
| lifeless | but unless one reboots thepuppetmaster | 03:06 |
| *** dina_belova has joined #openstack-infra | 03:06 | |
| lifeless | puppet won't be started | 03:06 |
| *** w_ has joined #openstack-infra | 03:06 | |
| clarkb | right so service puppet start is what you want | 03:06 |
| clarkb | I am just failing to context switch with you | 03:07 |
| clarkb | so did the new cert get generated with the old hostname? | 03:08 |
| lifeless | no, I didn't wipe the client state | 03:09 |
| lifeless | and on the puppet master thats in the same damn directory | 03:09 |
| lifeless | so I'm hitting it harder at the moment | 03:09 |
| lifeless | nuts - SSLCertificateFile: file '/var/lib/puppet/ssl/certs/ci-puppetmaster.pem' does not exist or is empty | 03:09 |
| clarkb | http://docs.puppetlabs.com/guides/troubleshooting.html | 03:09 |
| clarkb | there is another thing on the problem | 03:09 |
| clarkb | but it looks like the same content | 03:09 |
| *** dina_belova has quit IRC | 03:10 | |
| clarkb | openssl x509 -text -in /var/lib/puppet/ssl/certs/<hostname_of_puppet_master> to look at hte server cert | 03:10 |
| lifeless | thats bad filename | 03:11 |
| lifeless | is the original hostname | 03:11 |
| lifeless | bad apache2 puppetmaster site definition | 03:11 |
| clarkb | oh | 03:11 |
| lifeless | ok, muuuuch more progress. | 03:12 |
| lifeless | err: Could not retrieve catalog from remote server: Error 400 on SERVER: malformed format string - %S at /opt/config/production/manifests/site.pp:134 on node ci-puppetmaster.testing-cabal.org | 03:13 |
| lifeless | from agent --test | 03:13 |
| clarkb | cool it is actually compiling your manifests now | 03:14 |
| clarkb | site.pp has an error on line 134 | 03:14 |
| lifeless | nuts, hate all the merge commits; can gerrit be told to merge via rebase, since it has an artifical flat history... | 03:15 |
| clarkb | newer gerrit can rebase if necessary. gerrit that we have can only do cherry picks which we stopped doing because it makes a few things weird | 03:15 |
| clarkb | it actually updates your commit messages instead of using git notes | 03:15 |
| lifeless | whats the error on line 134? | 03:15 |
| lifeless | sysadmins => hiera('sysadmins'), | 03:15 |
| lifeless | ? | 03:15 |
| *** olaph has joined #openstack-infra | 03:16 | |
| lifeless | oh, my return from hiera is a list | 03:16 |
| lifeless | shouldn't it be? | 03:17 |
| clarkb | lifeless: google indicates maybe your yaml is malformed? | 03:17 |
| lifeless | clarkb: it is exactly what I pasted you :( | 03:17 |
| lifeless | clarkb: what does the sysadmins key for openstack-ci hiera look like ? | 03:17 |
| clarkb | lifeless: I am not on a machine with keys to look but I am pretty sure it uses the other list syntax (that shouldn't matter) | 03:18 |
| *** w_ has quit IRC | 03:18 | |
| lifeless | changed it to the block format | 03:19 |
| lifeless | happy now | 03:19 |
| lifeless | so yay | 03:19 |
| clarkb | wait really? | 03:19 |
| lifeless | so now in theory, I have a fully functional puppet master. Yay. | 03:19 |
| clarkb | it didn't like [] but - foo\n- bar is fine? | 03:19 |
| lifeless | that or perhaps the trailing , | 03:19 |
| lifeless | yaml confuses me :) | 03:20 |
| clarkb | also yay to functioning master | 03:20 |
| lifeless | now, write up more bootstrap docs | 03:20 |
| lifeless | then get the dashboard up | 03:20 |
| lifeless | I'm going to use a 2G instance - I'm on a rackspace free account, I don't want to abuse it - | 03:20 |
| clarkb | dashboard size depends a lot on number of lcients too | 03:21 |
| clarkb | 2G is probably fine especially if once pleia2 gets those optimizations working | 03:21 |
| *** ArxCruz has joined #openstack-infra | 03:21 | |
| *** w_ has joined #openstack-infra | 03:23 | |
| *** olaph has quit IRC | 03:26 | |
| *** olaph has joined #openstack-infra | 03:27 | |
| jhesketh__ | jeblair: ping | 03:28 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document setting the correct hostname. https://review.openstack.org/44393 | 03:29 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Ensure reports directory is setup correctly. https://review.openstack.org/44391 | 03:29 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document hiera setup a bit more comprehensively. https://review.openstack.org/44357 | 03:29 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make the adduser commands copy-pasteable. https://review.openstack.org/44270 | 03:29 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Install Nova and Cinder clients on puppetmaster. https://review.openstack.org/44384 | 03:29 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Add a pointer to launch/README for root operators. https://review.openstack.org/44163 | 03:29 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make setting up a Puppetmaster easier. https://review.openstack.org/44209 | 03:29 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make config updates work with rebased branches. https://review.openstack.org/44383 | 03:29 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document bootstrapping a new infrastructure. https://review.openstack.org/44262 | 03:29 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Start documenting running your own CI infra. https://review.openstack.org/44177 | 03:29 |
| lifeless | boom | 03:29 |
| lifeless | sorry :) | 03:29 |
| *** w_ has quit IRC | 03:29 | |
| *** w_ has joined #openstack-infra | 03:30 | |
| *** olaph has quit IRC | 03:32 | |
| *** dims has quit IRC | 03:32 | |
| clarkb | jhesketh__: I think jeblair called it a night a few hours ago | 03:34 |
| jhesketh__ | thought I might be a little late but worth a try :-) | 03:34 |
| lifeless | clarkb: openstack creds - you upload those to servers by hand ? | 03:35 |
| clarkb | lifeless: yes, and we put them in the files that are sourced according to launch/README | 03:37 |
| lifeless | would making /root/ci-launch via puppet be reasonable? | 03:37 |
| clarkb | lifeless: for launch.py and friends? | 03:38 |
| *** olaph has joined #openstack-infra | 03:38 | |
| clarkb | lifeless: or for credentials that can be sourced? | 03:38 |
| lifeless | the latter | 03:38 |
| lifeless | just make it, right owner and mode | 03:38 |
| clarkb | I suppose. You will need to put the info in hiera or some other place mostly protected | 03:39 |
| lifeless | clarkb: what info? | 03:39 |
| clarkb | the credentials | 03:39 |
| lifeless | clarkb: I'm not proposing to change how the creds are uploaded | 03:39 |
| clarkb | oh | 03:39 |
| lifeless | clarkb: just to make teh directory they end up in | 03:39 |
| lifeless | clarkb: so it's not by hand | 03:39 |
| clarkb | I see. ya that would be fine | 03:39 |
| clarkb | and help enforce that the permissions are correct | 03:40 |
| lifeless | exactly | 03:40 |
| *** w_ has quit IRC | 03:40 | |
| lifeless | can one put comments inside a definition | 03:40 |
| clarkb | lifeless: yes | 03:41 |
| clarkb | comments can start anywhere and go to end of line iirc | 03:41 |
| clarkb | you can also do C style block comments /* */ | 03:42 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Create ci-launch directory on the puppetmaster. https://review.openstack.org/44394 | 03:42 |
| *** vogxn has joined #openstack-infra | 03:42 | |
| clarkb | I am going to AFK now | 03:43 |
| lifeless | au revoir | 03:43 |
| clarkb | tomorrow I plan on reviewing this stuff so hopefully we can push a bunch in before you start work again on Monday | 03:43 |
| clarkb | (assuming you don't also hack on it over the wekeend :) ) | 03:43 |
| lifeless | presentation is on Friday I think; so yeah, am going to be fairly focused. | 03:44 |
| lifeless | I need a consistent stack up to gerrit/zuul/jenkins/nodepool/slaves. | 03:44 |
| *** jfriedly has quit IRC | 03:45 | |
| *** w_ has joined #openstack-infra | 03:46 | |
| *** olaph has quit IRC | 03:48 | |
| *** dkliban has quit IRC | 03:48 | |
| *** olaph has joined #openstack-infra | 03:49 | |
| *** w_ has quit IRC | 03:50 | |
| *** nicedice__ has joined #openstack-infra | 03:51 | |
| *** nicedice_ has quit IRC | 03:53 | |
| *** dkliban has joined #openstack-infra | 04:01 | |
| *** w_ has joined #openstack-infra | 04:01 | |
| *** ArxCruz has quit IRC | 04:02 | |
| *** olaph has quit IRC | 04:04 | |
| *** dina_belova has joined #openstack-infra | 04:07 | |
| *** w_ has quit IRC | 04:08 | |
| *** dina_belova has quit IRC | 04:11 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make third-party imports clearer in launch/. https://review.openstack.org/44395 | 04:19 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Install launch/ dependencies on puppetmaster. https://review.openstack.org/44384 | 04:19 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document setting the correct hostname. https://review.openstack.org/44393 | 04:19 |
| *** odyi has quit IRC | 04:19 | |
| lifeless | nuts, rebase mess up | 04:20 |
| *** nicedice_ has joined #openstack-infra | 04:21 | |
| *** odyi has joined #openstack-infra | 04:21 | |
| *** nicedice__ has quit IRC | 04:24 | |
| *** [wchrisj_ has joined #openstack-infra | 04:24 | |
| *** sgviking has quit IRC | 04:30 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Ensure reports directory is setup correctly. https://review.openstack.org/44391 | 04:31 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Install launch/ dependencies on puppetmaster. https://review.openstack.org/44384 | 04:31 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Create ci-launch directory on the puppetmaster. https://review.openstack.org/44394 | 04:31 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make third-party imports clearer in launch/. https://review.openstack.org/44395 | 04:31 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document setting the correct hostname. https://review.openstack.org/44393 | 04:31 |
| *** DennyZhang has quit IRC | 04:37 | |
| *** sgviking has joined #openstack-infra | 04:43 | |
| [wchrisj_ | Could someone with quantum/neutron experiencetell me if this is an issue with the CLI or the docs: https://bugs.launchpad.net/python-neutronclient/+bug/1198750 Has this bug been misclassified? | 04:43 |
| uvirtbot | Launchpad bug 1198750 in python-neutronclient "help for security-group-list calls out shell, should be csv" [Undecided,New] | 04:43 |
| *** [wchrisj_ has quit IRC | 04:46 | |
| *** [wchrisj_ has joined #openstack-infra | 04:46 | |
| *** boris-42 has joined #openstack-infra | 04:46 | |
| clarkb | [wchrisj_: that is an issue with the neutronclient args help | 04:49 |
| clarkb | which is not in the docs | 04:49 |
| [wchrisj_ | Thanks - newb here | 04:49 |
| [wchrisj_ | Would that be "python-neutronclient"? | 04:50 |
| clarkb | http://git.openstack.org/cgit/openstack/python-neutronclient/tree/ | 04:51 |
| [wchrisj_ | I've been looking at the github repo | 04:51 |
| [wchrisj_ | same thing? | 04:52 |
| *** amotoki has joined #openstack-infra | 04:52 | |
| clarkb | they are mirrors of each other | 04:52 |
| [wchrisj_ | aha - thanks! | 04:53 |
| *** jhesketh__ has quit IRC | 04:54 | |
| clarkb | [wchrisj_: https://cliff.readthedocs.org/en/latest/show_commands.html you may find that interesting. (that doc implies to me this may not be a bug) | 04:54 |
| *** jhesketh__ has joined #openstack-infra | 04:54 | |
| [wchrisj_ | This looks like a bug: | 04:57 |
| [wchrisj_ | usage: quantum security-group-rule-create [-h] [-f {shell,table} | 04:57 |
| *** nati_ueno has joined #openstack-infra | 04:57 | |
| [wchrisj_ | in this context, "shell" s/b "csv" | 04:57 |
| [wchrisj_ | appears | 04:57 |
| clarkb | [wchrisj_: I can't parse 'in this context, "shell" s/b "csv" appears' | 04:58 |
| clarkb | [wchrisj_: but "shell" comes form cliff and is documented in the page I linked | 04:59 |
| [wchrisj_ | sorry… the first part of the bug description seems to reflect that there is a bug | 04:59 |
| *** reed has quit IRC | 04:59 | |
| [wchrisj_ | I agree with you wrt the second paragraph of the bug desc | 05:00 |
| clarkb | there is an assertion that a bug exists, but I think that assertion is incorrect. This is expected behavior from cliff (I could be wrong though) | 05:00 |
| [wchrisj_ | ok | 05:00 |
| *** vogxn has quit IRC | 05:00 | |
| openstackgerrit | Joshua Hesketh proposed a change to openstack-infra/zuul: Add support for emailing results via SMTP https://review.openstack.org/42645 | 05:05 |
| *** SergeyLukjanov has joined #openstack-infra | 05:06 | |
| *** dina_belova has joined #openstack-infra | 05:07 | |
| *** [wchrisj_ has quit IRC | 05:09 | |
| *** vogxn has joined #openstack-infra | 05:10 | |
| *** dina_belova has quit IRC | 05:12 | |
| *** vogxn has quit IRC | 05:13 | |
| *** nicedice_ has quit IRC | 05:14 | |
| *** vogxn has joined #openstack-infra | 05:21 | |
| *** dkliban has quit IRC | 05:25 | |
| *** fifieldt has joined #openstack-infra | 05:34 | |
| *** SergeyLukjanov has quit IRC | 05:35 | |
| *** vogxn has quit IRC | 05:43 | |
| *** dina_belova has joined #openstack-infra | 05:57 | |
| *** DennyZhang has joined #openstack-infra | 05:59 | |
| *** DennyZhang has quit IRC | 06:00 | |
| *** DennyZhang has joined #openstack-infra | 06:00 | |
| *** dina_belova has quit IRC | 06:01 | |
| *** dina_belova has joined #openstack-infra | 06:02 | |
| *** olaph has joined #openstack-infra | 06:06 | |
| *** fifieldt has quit IRC | 06:06 | |
| *** dina_belova has quit IRC | 06:06 | |
| amotoki | hi, when I ran "tox -e pep8" in neutronclient after devstack run, i got pyflakes version conflict. http://paste.openstack.org/show/45451/ | 06:07 |
| amotoki | It seems that if test-requirements has the line "pyflakes>=0.7.2,<0.7.4" the version conflict occurs. | 06:08 |
| amotoki | after I removed pep8,pyflakes,flake8 from test-requirements, the issue solved. Is it a correct solution? | 06:09 |
| amotoki | It is same as test-requirements in nova, neutron, python-novaclient. | 06:10 |
| *** w_ has joined #openstack-infra | 06:14 | |
| *** nati_ueno has quit IRC | 06:15 | |
| *** olaph has quit IRC | 06:17 | |
| *** nati_ueno has joined #openstack-infra | 06:18 | |
| *** olaph has joined #openstack-infra | 06:23 | |
| *** w_ has quit IRC | 06:25 | |
| *** vogxn has joined #openstack-infra | 06:25 | |
| clarkb | only hacking should be listed | 06:29 |
| clarkb | it has dependencies on the others and will install the correct versions | 06:29 |
| *** SergeyLukjanov has joined #openstack-infra | 06:31 | |
| *** SergeyLukjanov has quit IRC | 06:33 | |
| *** DennyZhang has quit IRC | 06:35 | |
| *** morganfainberg is now known as morganfainberg|a | 06:41 | |
| amotoki | thanks. I will fix test-requirements in neutronclient. | 06:41 |
| *** Bada has joined #openstack-infra | 06:45 | |
| *** Bada has quit IRC | 06:45 | |
| *** boris-42 has quit IRC | 06:49 | |
| *** w_ has joined #openstack-infra | 06:50 | |
| *** odyssey4me has joined #openstack-infra | 06:51 | |
| *** olaph has quit IRC | 06:52 | |
| *** nati_ueno has quit IRC | 06:53 | |
| *** dina_belova has joined #openstack-infra | 06:58 | |
| *** odyssey4me has quit IRC | 06:58 | |
| *** odyssey4me has joined #openstack-infra | 06:59 | |
| *** thomasbiege has joined #openstack-infra | 07:04 | |
| *** dina_belova has quit IRC | 07:08 | |
| *** dina_belova has joined #openstack-infra | 07:09 | |
| *** thomasbiege2 has joined #openstack-infra | 07:13 | |
| *** dina_belova has quit IRC | 07:13 | |
| *** thomasbiege has quit IRC | 07:15 | |
| *** thomasbiege2 has quit IRC | 07:18 | |
| *** w_ has quit IRC | 07:24 | |
| *** fifieldt has joined #openstack-infra | 07:25 | |
| *** Bada has joined #openstack-infra | 07:35 | |
| *** w_ has joined #openstack-infra | 07:36 | |
| *** vogxn has quit IRC | 07:37 | |
| *** odyssey4me has quit IRC | 07:41 | |
| *** w_ has quit IRC | 07:50 | |
| *** olaph has joined #openstack-infra | 08:03 | |
| *** fbo_away is now known as fbo | 08:04 | |
| *** jhesketh__ has quit IRC | 08:07 | |
| *** w_ has joined #openstack-infra | 08:13 | |
| *** jhesketh__ has joined #openstack-infra | 08:13 | |
| *** hashar has joined #openstack-infra | 08:14 | |
| *** derekh has joined #openstack-infra | 08:15 | |
| *** olaph has quit IRC | 08:16 | |
| *** ruhe has joined #openstack-infra | 08:17 | |
| *** dina_belova has joined #openstack-infra | 08:20 | |
| dukhlov_ | Hi all | 08:21 |
| dukhlov_ | Can anybody help me with my problem that we discussed yesterday? | 08:22 |
| dukhlov_ | means "fatal: ICLA contributor agreement requires current contact information." | 08:22 |
| dukhlov_ | ? | 08:22 |
| *** dina_belova has quit IRC | 08:24 | |
| jd__ | dukhlov_: probably but you'll have to wait for the sun to rise above the US west coast | 08:25 |
| *** olaph has joined #openstack-infra | 08:25 | |
| *** w_ has quit IRC | 08:27 | |
| dukhlov_ | :). Ok, I will wait | 08:27 |
| *** vogxn has joined #openstack-infra | 08:28 | |
| *** DennyZhang has joined #openstack-infra | 08:33 | |
| *** olaph has quit IRC | 08:35 | |
| hashar | does anyone know who's in charge of building debian packages beside pabelanger ? :] | 08:39 |
| *** jhesketh__ has quit IRC | 08:40 | |
| *** olaph has joined #openstack-infra | 08:47 | |
| *** DennyZhang has quit IRC | 08:48 | |
| *** DennyZhang has joined #openstack-infra | 08:48 | |
| Mithrandir | hmm, did anybody get around to implementing the job cleanup stuff in jjb? | 08:53 |
| Mithrandir | or should I do it? | 08:53 |
| *** thomasbiege has joined #openstack-infra | 09:07 | |
| *** w_ has joined #openstack-infra | 09:11 | |
| *** olaph has quit IRC | 09:14 | |
| *** dina_belova has joined #openstack-infra | 09:20 | |
| *** dina_belova has quit IRC | 09:25 | |
| *** olaph has joined #openstack-infra | 09:27 | |
| *** w_ has quit IRC | 09:28 | |
| *** olaph has quit IRC | 09:35 | |
| *** ruhe has quit IRC | 09:38 | |
| *** boris-42 has joined #openstack-infra | 09:45 | |
| *** thomasbiege2 has joined #openstack-infra | 09:48 | |
| *** olaph has joined #openstack-infra | 09:48 | |
| *** thomasbiege has quit IRC | 09:49 | |
| *** w_ has joined #openstack-infra | 09:51 | |
| *** olaph has quit IRC | 09:54 | |
| *** hashar has quit IRC | 09:55 | |
| *** olaph has joined #openstack-infra | 09:59 | |
| *** w_ has quit IRC | 10:00 | |
| *** pcm_ has joined #openstack-infra | 10:03 | |
| *** pcm_ has joined #openstack-infra | 10:04 | |
| *** thomasbiege2 has quit IRC | 10:08 | |
| *** alexpilotti has joined #openstack-infra | 10:12 | |
| *** ruhe has joined #openstack-infra | 10:15 | |
| *** asalkeld has left #openstack-infra | 10:19 | |
| *** dina_belova has joined #openstack-infra | 10:21 | |
| *** boris-42 has quit IRC | 10:24 | |
| *** yaguang has quit IRC | 10:25 | |
| *** dina_belova has quit IRC | 10:26 | |
| *** DennyZha` has joined #openstack-infra | 10:30 | |
| openstackgerrit | Thierry Carrez proposed a change to openstack-infra/odsreg: Order topic names in the 'suggest session' dialog https://review.openstack.org/44434 | 10:39 |
| openstackgerrit | Daniel Izquierdo proposed a change to openstack-infra/config: Addition of the openstack-activity-board-dash project to the StackForge environment https://review.openstack.org/44057 | 10:42 |
| openstackgerrit | Thierry Carrez proposed a change to openstack-infra/odsreg: Empty comment box after posting one https://review.openstack.org/44435 | 10:50 |
| *** w_ has joined #openstack-infra | 10:50 | |
| *** olaph has quit IRC | 10:53 | |
| *** w_ has quit IRC | 11:00 | |
| *** olaph has joined #openstack-infra | 11:00 | |
| *** vogxn has quit IRC | 11:04 | |
| *** dukhlov_ has quit IRC | 11:04 | |
| *** w_ has joined #openstack-infra | 11:05 | |
| *** DennyZha` has quit IRC | 11:08 | |
| *** DennyZhang has quit IRC | 11:08 | |
| *** olaph has quit IRC | 11:09 | |
| *** dukhlov has joined #openstack-infra | 11:12 | |
| *** michchap has quit IRC | 11:12 | |
| *** ruhe has quit IRC | 11:19 | |
| *** dina_belova has joined #openstack-infra | 11:21 | |
| openstackgerrit | A change was merged to openstack-infra/odsreg: Order topic names in the 'suggest session' dialog https://review.openstack.org/44434 | 11:25 |
| *** dina_belova has quit IRC | 11:26 | |
| openstackgerrit | A change was merged to openstack-infra/odsreg: Empty comment box after posting one https://review.openstack.org/44435 | 11:27 |
| *** vogxn has joined #openstack-infra | 11:27 | |
| *** olaph has joined #openstack-infra | 11:27 | |
| *** ruhe has joined #openstack-infra | 11:29 | |
| *** w_ has quit IRC | 11:29 | |
| *** w_ has joined #openstack-infra | 11:30 | |
| *** DennyZhang has joined #openstack-infra | 11:31 | |
| *** jamiem has joined #openstack-infra | 11:32 | |
| *** jamiem has quit IRC | 11:33 | |
| *** olaph has quit IRC | 11:33 | |
| *** jamiem has joined #openstack-infra | 11:34 | |
| *** vogxn has quit IRC | 11:35 | |
| *** boris-42 has joined #openstack-infra | 11:37 | |
| *** jamiem has quit IRC | 11:37 | |
| *** boris-42 has quit IRC | 11:40 | |
| *** hashar has joined #openstack-infra | 11:41 | |
| *** olaph has joined #openstack-infra | 11:43 | |
| *** dims has joined #openstack-infra | 11:44 | |
| *** w_ has quit IRC | 11:46 | |
| *** rcleere has joined #openstack-infra | 11:46 | |
| *** w_ has joined #openstack-infra | 11:48 | |
| *** hashar has quit IRC | 11:48 | |
| *** olaph has quit IRC | 11:49 | |
| *** hashar has joined #openstack-infra | 11:51 | |
| *** olaph has joined #openstack-infra | 11:51 | |
| *** w_ has quit IRC | 11:53 | |
| *** weshay has joined #openstack-infra | 11:55 | |
| *** w_ has joined #openstack-infra | 11:55 | |
| *** olaph has quit IRC | 11:58 | |
| *** thomasbiege has joined #openstack-infra | 12:00 | |
| *** thomasm has joined #openstack-infra | 12:01 | |
| thomasm | hey all! | 12:03 |
| *** olaph has joined #openstack-infra | 12:06 | |
| *** w_ has quit IRC | 12:08 | |
| *** pblaho has joined #openstack-infra | 12:11 | |
| *** w_ has joined #openstack-infra | 12:12 | |
| *** DennyZhang has quit IRC | 12:12 | |
| *** DennyZhang has joined #openstack-infra | 12:13 | |
| *** Grizzlebee has joined #openstack-infra | 12:13 | |
| *** ArxCruz has joined #openstack-infra | 12:14 | |
| *** olaph has quit IRC | 12:14 | |
| *** rfolco has joined #openstack-infra | 12:19 | |
| *** dina_belova has joined #openstack-infra | 12:22 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Ensure reports directory is setup correctly. https://review.openstack.org/44391 | 12:22 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make the adduser commands copy-pasteable. https://review.openstack.org/44270 | 12:22 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Install launch/ dependencies on puppetmaster. https://review.openstack.org/44384 | 12:22 |
| *** openstackstatus has quit IRC | 12:22 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make config updates work with rebased branches. https://review.openstack.org/44383 | 12:22 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document bootstrapping a new infrastructure. https://review.openstack.org/44262 | 12:22 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Create ci-launch directory on the puppetmaster. https://review.openstack.org/44394 | 12:22 |
| *** cppcabrera has joined #openstack-infra | 12:22 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make third-party imports clearer in launch/. https://review.openstack.org/44395 | 12:22 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document setting the correct hostname. https://review.openstack.org/44393 | 12:22 |
| *** cppcabrera has left #openstack-infra | 12:23 | |
| *** dkliban has joined #openstack-infra | 12:24 | |
| *** olaph has joined #openstack-infra | 12:26 | |
| *** dina_belova has quit IRC | 12:27 | |
| *** w_ has quit IRC | 12:27 | |
| *** DennyZhang has quit IRC | 12:27 | |
| ttx | jeblair: trying to push initial commit for openstack/governance (with basic structure and .gitreview) but it does not seem to be in gerrit yet | 12:28 |
| ttx | could be a side-effect of not having "upstream" set for it in review.projects.yaml | 12:29 |
| *** ruhe has quit IRC | 12:32 | |
| jd__ | question on doc: I am under the impression that we push something to a stable branch, the documentation job is ran and the stable documentation is pushed at http://docs.openstack.org/developer/ceilometer/, and the same things goes to the same place for master | 12:35 |
| jd__ | in the end, would it be anyway possible to have 2 locations for stable and master doc to be both online at the same time? | 12:36 |
| *** ruhe has joined #openstack-infra | 12:40 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Fix launch-node for non-official environments. https://review.openstack.org/44449 | 12:41 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Ensure reports directory is setup correctly. https://review.openstack.org/44391 | 12:41 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make the adduser commands copy-pasteable. https://review.openstack.org/44270 | 12:41 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Install launch/ dependencies on puppetmaster. https://review.openstack.org/44384 | 12:41 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make config updates work with rebased branches. https://review.openstack.org/44383 | 12:41 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document bootstrapping a new infrastructure. https://review.openstack.org/44262 | 12:41 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Create ci-launch directory on the puppetmaster. https://review.openstack.org/44394 | 12:41 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make third-party imports clearer in launch/. https://review.openstack.org/44395 | 12:41 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document setting the correct hostname. https://review.openstack.org/44393 | 12:41 |
| *** thomasbiege has quit IRC | 12:41 | |
| *** alexpilotti has quit IRC | 12:43 | |
| *** boris-42 has joined #openstack-infra | 12:43 | |
| lifeless | jd__: most projects have different release vs master docs published, AFAICT. | 12:43 |
| lifeless | jd__: annegentle_ is possibly the best person to ask about that. | 12:44 |
| * jd__ knocks knocks annegentle_ | 12:44 | |
| *** dims has quit IRC | 12:44 | |
| *** ruhe has quit IRC | 12:44 | |
| *** ruhe has joined #openstack-infra | 12:45 | |
| *** dims has joined #openstack-infra | 12:47 | |
| *** rcleere has quit IRC | 12:47 | |
| *** woodspa has joined #openstack-infra | 12:48 | |
| *** anteaya has joined #openstack-infra | 12:49 | |
| anteaya | hashar: zul builds packages but I think his are ubuntu, I don't know if he also does debian packages, zigo builds debian packages | 12:55 |
| hashar | anteaya: yup zigo does the openstack packages | 12:56 |
| hashar | and pabelanger does the infra ones | 12:56 |
| Mithrandir | hmm, does something like https://gist.github.com/tfheen/5d667003aba41164eab0 make sense? | 12:56 |
| Mithrandir | (yes, it needs adding the delete options too) | 12:56 |
| *** pabelanger has quit IRC | 12:56 | |
| Mithrandir | and it needs to validate that it added the job itself. | 12:56 |
| anteaya | hi dukhlov_ in order to help you anymore, I think I need to see a stacktrace please, including the command you are trying to run | 12:56 |
| hashar | anteaya I though one more people was doing the infra packages as well. I wiil catch up with pabelanger. I could use a package for the python 'gear' module :-] thx! | 12:56 |
| anteaya | hashar: may you get your package | 12:57 |
| anteaya | Mithrandir: I know so little about Jenkins Job Builder, my opinion wouldn't count for much, but I did look at the diff | 12:57 |
| anteaya | hopefully someone with more JJB experience than I will be around soon to weigh in | 12:58 |
| Mithrandir | anteaya: basically, I want to be able to say "purge obsolete jobs" | 12:58 |
| anteaya | a valuable function | 12:58 |
| hashar | Mithrandir: there is a delete-all command | 12:58 |
| hashar | Mithrandir: though that would delete any job | 12:59 |
| Mithrandir | hashar: yes, so it's not what I want. | 12:59 |
| annegentle_ | jd__: aroo | 12:59 |
| hashar | Mithrandir: ah you only want to delete jobs which are not configured in JJB. Clever! | 12:59 |
| Mithrandir | I want to delete obsolete jobs, not all jobs. We still have some hand-maintained jobs. | 12:59 |
| hashar | my recommendation would be to push your change in Gerrit | 12:59 |
| Mithrandir | hashar: no, I want to delete those which are no longer configured by jjb. | 12:59 |
| Mithrandir | which is subtly different. | 12:59 |
| hashar | and if it is not really ready, mark it as a work in progress with the [WIP] button in Gerrit | 12:59 |
| Mithrandir | hashar: sure, and I will eventually. | 12:59 |
| hashar | that will attract reviewers :-] We are a bunch looking at the changes submitted | 13:00 |
| Mithrandir | I was hoping to lure jeblair or mordred to comment, really. :-P | 13:00 |
| annegentle_ | jd__: for docs.o.o/dev/project name we don't publish point releases | 13:00 |
| hashar | i like that feature myself | 13:00 |
| Mithrandir | hashar: I already have a bunch of patches in, I know the flow. | 13:00 |
| Mithrandir | thanks for the pointers though | 13:00 |
| hashar | \O/ | 13:00 |
| annegentle_ | jd__: I think that the jobs push from master | 13:00 |
| lifeless | woo, got a server up and self registered. yay. | 13:00 |
| annegentle_ | jd__: master only | 13:00 |
| annegentle_ | jd__: we do backport from openstack-manuals to prior releases | 13:01 |
| anteaya | Mithrandir: mordred won't be lured to look at anything 'til next week | 13:01 |
| anteaya | he is at burning man | 13:01 |
| annegentle_ | jd__: you can backport docs to past releases from ceiliometer/doc/source/ so that someone could build locally, but they're not published to the web | 13:01 |
| jd__ | annegentle_: reading the infra code, I'm under the impression there's no distinction (that's why I aksed here in the first place) | 13:01 |
| annegentle_ | jd__: yep that's accurate | 13:01 |
| jd__ | annegentle_: could you publish stable releases documentation too then? | 13:02 |
| lifeless | clarkb: / jeblair: is the puppet dashboard meant to be anonymous-access? | 13:02 |
| annegentle_ | jd__: swift was the only project who wanted to, and it became too much trouble/maintenance for the value | 13:02 |
| anteaya | jeblair has been working on overdrive for the past two weeks and will show up in a few hours, I try to answer the phone in the first couple hours of my day to relieve some of the pressure on him | 13:02 |
| jd__ | annegentle_: people deplying Grizzly reads the Havana-dev documentation online right now, that doesn't seem like a good thing | 13:02 |
| annegentle_ | jd__: but it's dev doc, deployers should read deployer doc | 13:02 |
| annegentle_ | jd__: to me that's the distinction | 13:02 |
| jd__ | annegentle_: we don't have a deployer doc :( | 13:03 |
| *** ruhe has quit IRC | 13:03 | |
| lifeless | jd__: working up to it :) | 13:03 |
| *** amotoki has quit IRC | 13:03 | |
| jd__ | annegentle_: our deployer doc is in our dev doc :( | 13:03 |
| annegentle_ | jd__: it would belong in openstack-manuals | 13:03 |
| anteaya | lifeless: \o/ | 13:03 |
| annegentle_ | jd__: I know | 13:03 |
| annegentle_ | jd__: and it's okay, but do take some steps | 13:03 |
| jd__ | annegentle_: I get it :) | 13:03 |
| lifeless | anteaya: https://github.com/testing-cabal/ci-config/tree/testcabal | 13:03 |
| jd__ | annegentle_: well, (how) can we help? | 13:03 |
| *** olaph has quit IRC | 13:03 | |
| lifeless | anteaya: in addition to the somewhat large stack I have floating on top of openstack-infra/config in gerrit ;) | 13:04 |
| anteaya | wahoo, look at that | 13:04 |
| anteaya | ha ha ha | 13:04 |
| jd__ | annegentle_: having the deployer documentation outside our repository seems also quite problematic since our API documentation is automatically generated from the source code for example | 13:04 |
| *** olaph has joined #openstack-infra | 13:05 | |
| annegentle_ | jd__: your API docs are very deployer oriented, I realize, where most other projects their API docs are for end-users | 13:05 |
| annegentle_ | jd__: we can figure this out | 13:05 |
| jd__ | annegentle_: well our API is user oriented too since we got alarming | 13:06 |
| annegentle_ | jd__: we already scrape a bunch of config info from the other projects, how about automating to put into the Config Ref | 13:06 |
| annegentle_ | jd__: here's the lay of the land | 13:06 |
| *** dims has quit IRC | 13:06 | |
| anteaya | lifeless: this is awesome | 13:06 |
| jd__ | annegentle_: yeah generating configuration information is another problem indeed :( | 13:06 |
| anteaya | in other news, the gate is empty | 13:07 |
| annegentle_ | jd__: http://paste.openstack.org/show/45468/ | 13:07 |
| annegentle_ | jd__: hopefully we can get your docs into this sort of pattern | 13:07 |
| annegentle_ | jd__: sorry, typo in the paste | 13:07 |
| *** vogxn has joined #openstack-infra | 13:08 | |
| annegentle_ | jd__: but hopefully it gives you the idea | 13:08 |
| jd__ | annegentle_: yes, I get it | 13:08 |
| jd__ | annegentle_: what's the motivation to have documentation outside the projects? | 13:09 |
| lifeless | incoming | 13:09 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Ensure reports directory is setup correctly. https://review.openstack.org/44391 | 13:09 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Install launch/ dependencies on puppetmaster. https://review.openstack.org/44384 | 13:09 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make config updates work with rebased branches. https://review.openstack.org/44383 | 13:09 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Create ci-launch directory on the puppetmaster. https://review.openstack.org/44394 | 13:09 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make third-party imports clearer in launch/. https://review.openstack.org/44395 | 13:09 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document setting the correct hostname. https://review.openstack.org/44393 | 13:09 |
| annegentle_ | jd__: the tools selected did not enable translations at the time. That's not the whole story but https://wiki.openstack.org/wiki/WebContentRequirements gives more indepth requirements that Sphinx didn't meet. | 13:10 |
| annegentle_ | jd__: doesn't mean we can't revisit, and obviously we are all the time, but when you have to take 12 projects into account you make choices | 13:11 |
| jd__ | annegentle_: sure, I'm just trying to get the context :) | 13:11 |
| annegentle_ | jd__: when people come to docs.openstack.org, they want OpenStack docs. We need to get Ceilometer fitting into the install guide, configuration guide, end user guide, admin user guide, and so on | 13:11 |
| jd__ | I'm already embarassed we are not better at documentation | 13:11 |
| annegentle_ | jd__: I love this morning's post so I know you're on the right track | 13:12 |
| jd__ | annegentle_: my thinking is, it would be better to force people to update documentation with their patches if all the documentation was in our repository | 13:12 |
| annegentle_ | jd__: I do prefer docs checked in with code and it's mortifying as I go through DocImpact bugs now how little I have to go on. | 13:12 |
| *** w_ has joined #openstack-infra | 13:12 | |
| annegentle_ | jd__: yes, do force update docs, use DocImpact, use commit messages for docs that can be integrated elsewhere, all of it | 13:12 |
| jd__ | yeah I don't think DocImpact is a good solution for you | 13:12 |
| *** ruhe has joined #openstack-infra | 13:13 | |
| jd__ | I'd prefer to have the people writing/modifying the feature to update the doc at the same time | 13:13 |
| jd__ | annegentle_: maybe we could experiment something on Ceilometer about this, like having API doc and deployer doc etc in the project rather than outside? | 13:13 |
| annegentle_ | jd__: you're not core so go ahead :) | 13:14 |
| annegentle_ | jd__: I really would encourage an entry on api.openstack.org/api_ref.html at a minimum | 13:14 |
| *** ruhe has quit IRC | 13:14 | |
| *** mriedem has joined #openstack-infra | 13:15 | |
| *** olaph has quit IRC | 13:15 | |
| jd__ | annegentle_: ok | 13:15 |
| jd__ | I'll go through this and try to build some tasks we can tackle to advance on that matters | 13:16 |
| annegentle_ | jd__: and we have almost no DocImpact from ceilometer that I can see. We automated doc bug generation from that so ideally we'd have more of those from the repo and automation | 13:16 |
| *** DennyZhang has joined #openstack-infra | 13:16 | |
| annegentle_ | jd__: sounds good. I can help prioritize as I'm sure it's a bit overwhelming now. | 13:16 |
| jd__ | annegentle_: hm anyway, what would you do with Ceilometer's DocImpact? | 13:16 |
| jd__ | (with the current doc status) | 13:16 |
| annegentle_ | jd__: in any commit that affects docs, put DocImpact and enough info to write docs from or docs themselves. | 13:16 |
| *** ruhe has joined #openstack-infra | 13:16 | |
| *** w_ has quit IRC | 13:17 | |
| annegentle_ | jd__: go through your merged commits and log doc bugs manually | 13:17 |
| jd__ | annegentle_: but who and where this doc will be written? :) | 13:17 |
| jd__ | s/who/by who/ | 13:17 |
| annegentle_ | jd__: your devs, your deployers | 13:17 |
| annegentle_ | jd__: and I can help recruit | 13:17 |
| jd__ | annegentle_: ok that's the problem, nobody's doing it :D | 13:17 |
| annegentle_ | jd__: well, yes. :) | 13:17 |
| jd__ | but I get the picture | 13:17 |
| annegentle_ | jd__: you have several deployers to ask though | 13:17 |
| jd__ | such as? | 13:18 |
| annegentle_ | jd__: Ryan_Lane, right | 13:18 |
| annegentle_ | jd__: sandywalsh, I think | 13:18 |
| jd__ | ack :) | 13:18 |
| annegentle_ | jd__: perhaps "several" is an exaggeration :) | 13:19 |
| annegentle_ | jd__: seems like more than a few though | 13:19 |
| jd__ | yes, for sure | 13:19 |
| *** olaph has joined #openstack-infra | 13:20 | |
| *** sandywalsh_ has joined #openstack-infra | 13:20 | |
| *** mriedem has quit IRC | 13:21 | |
| *** dina_belova has joined #openstack-infra | 13:22 | |
| sandywalsh_ | jd__, ah, just read the email ... makes sense now. We'll need some help from annegentle_ to understand what the "state of the art" is for openstack docs these days. | 13:23 |
| sandywalsh_ | annegentle_, is there a "how to write docs for openstack" doc? | 13:24 |
| sandywalsh_ | :) | 13:24 |
| *** sandywalsh has quit IRC | 13:24 | |
| *** dims has joined #openstack-infra | 13:24 | |
| *** mriedem has joined #openstack-infra | 13:24 | |
| *** w_ has joined #openstack-infra | 13:24 | |
| annegentle_ | sandywalsh_: of course! http://wiki.openstack.org/Documentation/Howto | 13:26 |
| annegentle_ | er | 13:26 |
| annegentle_ | http://wiki.openstack.org/Documentation/HowTo | 13:26 |
| *** dina_belova has quit IRC | 13:27 | |
| *** olaph has quit IRC | 13:27 | |
| annegentle_ | how do I find a list of core members of any project team? | 13:30 |
| annegentle_ | this one didn't work https://launchpad.net/~neutron-core/+members#active | 13:30 |
| sandywalsh_ | annegentle_, thanks ... I'll get freshened up on that and try to allocate some time. The RC-phase might be good for that. | 13:30 |
| pblaho | annegentle_: I think you can try https://review.openstack.org/#/admin/groups/ | 13:31 |
| annegentle_ | pblaho: ah that's it, thanks | 13:31 |
| sandywalsh_ | https://launchpad.net/~ceilometer-drivers/+members#active | 13:31 |
| sandywalsh_ | pblaho, that's a handy link ... saved. thanks | 13:32 |
| *** DennyZhang has quit IRC | 13:33 | |
| pblaho | annegentle_: sandywalsh_ : no problem | 13:33 |
| *** tstevenson has joined #openstack-infra | 13:38 | |
| tstevenson | morning | 13:39 |
| *** DennyZhang has joined #openstack-infra | 13:42 | |
| *** w_ has quit IRC | 13:43 | |
| anteaya | morning tstevenson | 13:44 |
| annegentle_ | can I talk to someone (I think it was clarkb or fungi) about what it means to shallow clone a git repo to reuse the content? | 13:44 |
| annegentle_ | such as, does that happen at build time? | 13:46 |
| *** w_ has joined #openstack-infra | 13:47 | |
| *** boris-42 has quit IRC | 13:48 | |
| annegentle_ | and, is it part of a script? | 13:48 |
| *** russellb is now known as rustlebee | 13:49 | |
| anteaya | ah yes, Friday | 13:49 |
| *** anteaya is now known as ColinMochrie | 13:50 | |
| ColinMochrie | bnemec: oh Ryan | 13:50 |
| *** w_ has quit IRC | 13:50 | |
| *** prad_ has joined #openstack-infra | 13:51 | |
| *** burt has joined #openstack-infra | 13:53 | |
| *** jog0 is now known as flashgordon | 13:53 | |
| *** datsun180b has joined #openstack-infra | 13:53 | |
| *** pblaho has quit IRC | 13:54 | |
| *** pblaho has joined #openstack-infra | 13:55 | |
| *** ruhe has quit IRC | 13:56 | |
| *** alexpilotti has joined #openstack-infra | 13:57 | |
| *** fifieldt has quit IRC | 14:02 | |
| ColinMochrie | if anyone is struggling to figure out how to use rvm (Ruby Version Manager) - potentially useful for puppet installations, here is an 8 minute introductory screencast that explains the basics: http://screencasts.org/episodes/how-to-use-rvm the narrator has a very friendly voice | 14:02 |
| *** ruhe has joined #openstack-infra | 14:03 | |
| *** ruhe has quit IRC | 14:11 | |
| *** jamiem has joined #openstack-infra | 14:12 | |
| *** ruhe has joined #openstack-infra | 14:14 | |
| *** thedodd has joined #openstack-infra | 14:16 | |
| *** vogxn has quit IRC | 14:18 | |
| *** thedodd is now known as jim | 14:19 | |
| *** gaelL has quit IRC | 14:19 | |
| *** jim is now known as Guest44551 | 14:19 | |
| *** Guest44551 has quit IRC | 14:20 | |
| *** thedodd has joined #openstack-infra | 14:21 | |
| ColinMochrie | pleia2: as I read the docs on puppet-dashboard, I'll share some thoughts | 14:22 |
| *** dina_belova has joined #openstack-infra | 14:23 | |
| ColinMochrie | first up, major development seems to have stopped about 2 years ago, puppet-dashboard depends on ruby 1.8.7 which is an old version of ruby and isn't really being supported anymore | 14:23 |
| ColinMochrie | ruby 1.9.x is the ruby of choice currently and ruby 2 has been released | 14:23 |
| mgagne | ColinMochrie: puppet-dashboard is now maintained by Aaron Stone | 14:24 |
| ColinMochrie | so the latest round of ruby/rails security issues, I have forgotten exactly but it was database related means that ruby 1.8.7 will be vulnerable | 14:24 |
| ColinMochrie | mgagne: good to know | 14:24 |
| ColinMochrie | is he available for a conversation and where would I find him? | 14:24 |
| ColinMochrie | I'll change to my real nick if he is up for a chat | 14:25 |
| *** ColinMochrie is now known as anteaya | 14:25 | |
| mgagne | ColinMochrie: puppetlabs only put efforts in the enterprise dashboard which I think is called Puppet Console | 14:25 |
| anteaya | right | 14:25 |
| anteaya | which leaves us out, since we don't use enterprise | 14:25 |
| mgagne | anteaya: there is some alternative outside puppetlabs but don't know how they compare to dashboard | 14:26 |
| anteaya | mgagne: I'm willing to look | 14:26 |
| anteaya | have you googleable terms or urls? | 14:26 |
| anteaya | I have to find something that plays nice with what we have | 14:27 |
| mgagne | anteaya: foreman is the one I know about, it comes with a provisioning tool too I think | 14:27 |
| anteaya | okay I will look at foreman, is it ruby? | 14:27 |
| *** dina_belova has quit IRC | 14:28 | |
| *** ruhe has quit IRC | 14:29 | |
| *** SergeyLukjanov has joined #openstack-infra | 14:29 | |
| mgagne | anteaya: looks to use ruby | 14:29 |
| anteaya | mgagne: http://theforeman.org/ ? | 14:29 |
| mgagne | anteaya: yes | 14:29 |
| anteaya | not sure that is what we want | 14:29 |
| anteaya | we are looking for a gui to inferface with what puppet is already doing | 14:30 |
| mgagne | anteaya: as I said, it is a provisioning tool which happens to use puppet. There is a dashboard people are talking. I don't know how easy it would be to strip/disable the provisioning part and only use the dashboard | 14:30 |
| anteaya | ah okay, I think I get it now | 14:30 |
| mgagne | anteaya: or if it plays well with an existing installation | 14:31 |
| anteaya | we are just looking for a dashboard to interface with current provisioning | 14:31 |
| *** gaelL has joined #openstack-infra | 14:31 | |
| anteaya | perhaps figuring out how I talk with Aaron Stone is a better direction | 14:32 |
| anteaya | do you know him? | 14:32 |
| mgagne | anteaya: =) | 14:32 |
| anteaya | :D | 14:32 |
| mgagne | anteaya: not personally, I just happened to read a tweet about him taking over puppet dashboard | 14:32 |
| anteaya | ah okay thanks | 14:33 |
| anteaya | a little bird told you | 14:33 |
| anteaya | :D | 14:33 |
| mgagne | anteaya: should be him: https://github.com/sodabrew | 14:34 |
| anteaya | well foreman will use ruby 1.9 as well as 1.8 so there's that | 14:34 |
| *** comstud is now known as bearhands | 14:34 | |
| mgagne | anteaya: https://github.com/sodabrew/puppet-dashboard | 14:35 |
| anteaya | yeah | 14:35 |
| anteaya | this is a step up, you can use 1.9.3 | 14:35 |
| anteaya | might be worth spinning this up on a test node to see if we can get it working | 14:35 |
| mgagne | anteaya: background info about him willing to take over dev: https://groups.google.com/forum/#!msg/puppet-users/j44EbTJY7HI/EQSJgh-g1fgJ | 14:36 |
| anteaya | oh thank you | 14:36 |
| * anteaya reads | 14:36 | |
| anteaya | "I started porting to Rails 3 last week" written about a year ago - yay! | 14:37 |
| anteaya | this is much better | 14:37 |
| *** ruhe has joined #openstack-infra | 14:38 | |
| anteaya | though the whole use of puppet dashboard looks like a project with a short shelf life | 14:40 |
| *** boris-42 has joined #openstack-infra | 14:40 | |
| *** rnirmal has joined #openstack-infra | 14:42 | |
| anteaya | thanks mgagne | 14:42 |
| mgagne | anteaya: np | 14:43 |
| mgagne | anteaya: I'll keep an eye on what you will choose, I'm interested in your conclusions =) | 14:43 |
| *** sandywalsh_ has quit IRC | 14:44 | |
| *** UtahDave has joined #openstack-infra | 14:45 | |
| *** markmcclain has joined #openstack-infra | 14:47 | |
| ttx | anteaya: reviewing your election setup guidelines. Why are you advising to check the "Present choices on voting page in exactly the given order" box ? | 14:52 |
| anteaya | mgagne: thanks I welcome your opinions during the conversation | 14:58 |
| anteaya | ttx: thanks for reviewing | 14:59 |
| anteaya | to enforce an alphabetical order if that was what was input | 14:59 |
| anteaya | please remove if you disagree | 14:59 |
| dukhlov | anteaya: Hello | 14:59 |
| anteaya | dukhlov: hello | 14:59 |
| anteaya | have you a paste? | 15:00 |
| ttx | anteaya: yeah, I disagree... numerous studies prove that the order influences the voters, so randomizing the order helps in having a fair election | 15:00 |
| ttx | anteaya: willfix | 15:00 |
| anteaya | ttx great thank you | 15:00 |
| *** dina_belova has joined #openstack-infra | 15:00 | |
| *** sandywalsh_ has joined #openstack-infra | 15:00 | |
| anteaya | I didn't know we had been going with randomized order previously | 15:01 |
| dukhlov | anteaya: regarding my yersterday problem (fatal: ICLA contributor agreement requires current contact information.)... I found out the cause | 15:02 |
| anteaya | good | 15:02 |
| anteaya | what was it? | 15:02 |
| dukhlov | anteaya: I registered with launchpad as 'dukhlov' vith mail 'dukhlov@mirantis.com' then I went to review.openstack.org and review profile was created automatically with username dukhlov | 15:04 |
| dukhlov | anteaya:then I had a problem with launchpad and resigned launchpad accaunt. After thar when I came to review.openstack, it seems that new review profile was created | 15:06 |
| anteaya | ah | 15:06 |
| ttx | anteaya: made a few modifications to https://wiki.openstack.org/wiki/Election_Officiating_Guidelines . Please review :) | 15:06 |
| * anteaya reviews | 15:06 | |
| ttx | oh. forgot something | 15:07 |
| dukhlov | anteaya: and it have empty username and when I entrered 'dukhlov' it was rejected and I entered 'dyx' | 15:07 |
| dukhlov | anteaya:so, my current solutiom is the next: I set up gitreview.username=dyx in my git configuration and all works fine | 15:08 |
| anteaya | ah ha, that would do it dukhlov | 15:08 |
| ttx | anteaya: posted new version with a warning about locale. That bit me in the past | 15:09 |
| anteaya | okay, well know we know you have two accounts, I will ask clarkb or fungi once he gets back to look at the db and clean it up so we don't have a ghost entry for you, that has happened before | 15:09 |
| anteaya | dukhlov: glad you have it working and thanks for telling me what happened | 15:09 |
| *** Bada has quit IRC | 15:10 | |
| anteaya | dukhlov: please remind me next week to see if the db clean up for your account(s) have taken place | 15:10 |
| ttx | Oh, and something else | 15:10 |
| * anteaya stands by | 15:10 | |
| anteaya | ha ha ha | 15:10 |
| anteaya | bet the french invitations to vote were well received :D | 15:11 |
| ttx | anteaya: they were. | 15:12 |
| anteaya | ha ha ha | 15:12 |
| ttx | added a bit on picking election officials | 15:12 |
| anteaya | okay /me reviews again | 15:14 |
| anteaya | this is great, thanks ttx | 15:15 |
| anteaya | I missed a few things officiating this election and will do a better job next time | 15:15 |
| anteaya | this election is only 2 days long and I was asked not to do a detailed results report, I will fix those next time | 15:16 |
| ttx | anteaya: the detailed results report lets us run the anonymized ballots through alternative, experimental algorithms and check if they would have given better results | 15:17 |
| anteaya | ttx okay, good to know | 15:18 |
| anteaya | I will do so henceforth | 15:18 |
| *** [wchrisj_ has joined #openstack-infra | 15:19 | |
| ttx | anteaya: (we are considering moving the the "proportional representation" option but it's a bit of a mystery what it would do, so we want to run a few elections results through it to test | 15:19 |
| ttx | s/the the/to the/ | 15:19 |
| anteaya | ah okay | 15:19 |
| ttx | no big deal if that election missed it | 15:19 |
| anteaya | yes, the enforce proportional representation checkbox had me wondering too, glad you are evaluating that setting | 15:19 |
| anteaya | ttx okay thanks | 15:20 |
| *** boris-42 has quit IRC | 15:20 | |
| clarkb | good morning | 15:29 |
| jeblair | clarkb: good morning | 15:30 |
| jeblair | clarkb: are you caught up on code review? i mean, obviously not stuff since yesterday, but there are a few things i reviewed a few days ago that you haven't reviewed | 15:30 |
| ttx | clarkb, jeblair: hola | 15:30 |
| jeblair | clarkb: wondering if you're still working on catching up or missed them. eg: https://review.openstack.org/#/c/40498/ | 15:31 |
| jeblair | ttx: hola! | 15:31 |
| anteaya | morning clarkb and jeblair | 15:31 |
| *** Protux has joined #openstack-infra | 15:31 | |
| * ttx reposts questions lost in backlog | 15:31 | |
| ttx | <ttx> jeblair: trying to push initial commit for openstack/governance (with basic structure and .gitreview) but it does not seem to be in gerrit yet | 15:31 |
| ttx | <ttx> could be a side-effect of not having "upstream" set for it in review.projects.yaml | 15:31 |
| jeblair | ttx: ah, it's probably just manage_projects breaks about half the time it runs :( | 15:32 |
| jeblair | running it manually | 15:32 |
| ttx | jeblair: let me know when it's up, I've got an initial commit for it | 15:32 |
| *** neelashah has joined #openstack-infra | 15:33 | |
| clarkb | jeblair: probably a little bit of both. I know I am behind but have tried to skim the list and review important bits | 15:34 |
| *** vogxn has joined #openstack-infra | 15:34 | |
| clarkb | jeblair: I plan on trying to catch up today | 15:34 |
| jeblair | Failed to add groups file for project: stackforge/clouddocs-maven-plugin | 15:35 |
| clarkb | jeblair: arg that is a known issue | 15:35 |
| clarkb | jeblair: let me push a fix now before I can forget again | 15:35 |
| *** DennyZhang has quit IRC | 15:35 | |
| *** bpokorny has joined #openstack-infra | 15:35 | |
| neelashah | Does anyone know if there are issues currently with gate-nova-python27? | 15:35 |
| anteaya | neelashah: what are you seeing? | 15:36 |
| anteaya | can I get the url to a log? | 15:36 |
| neelashah | anteaya: http://logs.openstack.org/79/30479/14/check/gate-nova-python27/c8aebbe/console.html | 15:36 |
| * anteaya looks | 15:37 | |
| neelashah | its failing 27 at the gate | 15:37 |
| dstufft | mordred: possibly relevant to your interests http://www.python.org/dev/peps/pep-0453/ | 15:37 |
| anteaya | ERROR: InvocationError: '/home/jenkins/workspace/gate-nova-python27/.tox/py27/bin/python setup.py testr --slowest --testr-args=' | 15:38 |
| anteaya | dstufft: he is away until next week and told me he won't be reading backscroll, I suggest you try him then | 15:38 |
| dstufft | okies | 15:39 |
| anteaya | UnknownMethodCallError: Method called is not a member of the object: __deepcopy__ | 15:39 |
| anteaya | neelashah: I'm just pulling out stuff that might be helpful for clarkb or jeblair | 15:40 |
| neelashah | thanks anteaya | 15:40 |
| clarkb | neelashah: anteaya: That looks like a valid failure of the test | 15:42 |
| *** boris-42 has joined #openstack-infra | 15:43 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Add Gerrit ACLs for clouddocs maven plugin project https://review.openstack.org/44471 | 15:43 |
| clarkb | jeblair: ^ that should fix the manage_projects error you saw | 15:43 |
| anteaya | thanks clarkb, glad it isn't an infra issue | 15:43 |
| clarkb | jeblair: manage_projects should configure other projects even with that error though | 15:43 |
| clarkb | now time for code review | 15:43 |
| neelashah | ok clarkb, thanks - something must have changed in underlying code since it last passed jenkins | 15:43 |
| *** mrodden has quit IRC | 15:43 | |
| neelashah | thanks anteaya | 15:44 |
| clarkb | neelashah: very possible | 15:44 |
| clarkb | we merge a lot of code | 15:44 |
| jeblair | clarkb: https://jenkins.openstack.org/job/nova-upstream-translation-update/4020/console | 15:44 |
| *** thomasbiege has joined #openstack-infra | 15:45 | |
| anteaya | thanks for the log neelashah, hope you track down the error | 15:45 |
| clarkb | jeblair: looks like openstackjenkins was removed from the nova maintainer list? /me looks on transifex | 15:45 |
| *** fbo is now known as fbo_away | 15:46 | |
| clarkb | jeblair: yup https://www.transifex.com/projects/p/nova/edit/maintainers/ | 15:46 |
| clarkb | jeblair: do we want to check with fifieldt before adding jenkins again? | 15:46 |
| *** sgviking has quit IRC | 15:47 | |
| jeblair | clarkb: why? i mean, it's great talking to him, but it just seems like a straightforward oversight... | 15:48 |
| *** sgviking has joined #openstack-infra | 15:49 | |
| clarkb | jeblair: maybe it was intentional? though at this point Transifex has changed so much stuff behind the scenes and this is probably related to something transifex did and not fifieldt | 15:49 |
| *** boris-42 has quit IRC | 15:50 | |
| jeblair | clarkb: weird; openstackjenkins is a maintainer of the openstack hub project | 15:51 |
| clarkb | jeblair: there is a Nova_old, I wonder if this is related to the en_US -> en transition | 15:51 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/devstack-gate: Add support for large_ops tempest test https://review.openstack.org/43779 | 15:51 |
| *** markmcclain has quit IRC | 15:51 | |
| clarkb | jeblair: it is possible fifieldt did not add jenkins to the maintainers list so that we could make sure the jenkins scripts output files with the correct source local for the .pot | 15:52 |
| dukhlov | anteaya: ok, thank you | 15:52 |
| *** boris-42 has joined #openstack-infra | 15:52 | |
| flashgordon | clarkb: ping | 15:53 |
| clarkb | flashgordon: pong | 15:53 |
| jeblair | clarkb: ok; we can wait, though i think his weekend has started | 15:53 |
| *** fbo_away is now known as fbo | 15:53 | |
| clarkb | jeblair: the whole en vs en_US thing is a bit of a mess | 15:53 |
| flashgordon | clarkb: so once I get the large-ops job sorted out fully and running what is the next step | 15:53 |
| jeblair | clarkb: true, i don't want to make it worse. | 15:53 |
| *** dkehn has quit IRC | 15:54 | |
| flashgordon | would it be possible to get it to gate, and if not can we get it to run periodically and email or something when it fails? | 15:54 |
| clarkb | jeblair: tl;dr is horizon cannot do en_US, transifex only shares translation memory if the source locale is the same across projects so all projects with en_US (I think this is the default on our ubuntu machines) need to transition to en | 15:54 |
| flashgordon | clarkb: also is it possible to graph the test duration over time. | 15:54 |
| jeblair | i wonder if pootle has that restriction | 15:55 |
| *** dkehn has joined #openstack-infra | 15:55 | |
| jeblair | clarkb: so, puppet is running on gerrit, but when i run it by hand, it is doing stuff | 15:55 |
| clarkb | flashgordon: next step is probably into silent pipeline or non voting (to gather data), then when happy with results there we can gate it | 15:55 |
| *** [wchrisj_ has quit IRC | 15:56 | |
| clarkb | flashgordon: test duration isn't logged in an easily consumable place. | 15:56 |
| *** hashar has quit IRC | 15:57 | |
| jeblair | graphite | 15:57 |
| clarkb | is duration in graphite now? | 15:57 |
| jeblair | clarkb: i think it has been for a while | 15:57 |
| jeblair | lemme check | 15:57 |
| flashgordon | clarkb: what is the silent pipeline? | 15:57 |
| clarkb | flashgordon: it is equivalent to the check pipeline but it does not report to gerrit | 15:58 |
| clarkb | flashgordon: so you can run a lot of tests on real patchsets without noise on reviews | 15:58 |
| clarkb | good for gathering data | 15:58 |
| flashgordon | clarkb: where can you see the results from it? | 15:58 |
| *** ruhe has quit IRC | 15:58 | |
| clarkb | flashgordon: on logs.openstack.org/GITSHA[:2]/GITSHA/silent/$NAME_OF_TEST | 15:59 |
| clarkb | flashgordon: or in elasticsearch or in jenkins | 15:59 |
| *** ruhe has joined #openstack-infra | 15:59 | |
| *** markmcclain has joined #openstack-infra | 16:00 | |
| clarkb | jeblair: https://review.openstack.org/#/c/40498/ reviewed, it needs a rebase | 16:00 |
| flashgordon | clarkb: cool anything in silent pipeline today? | 16:00 |
| flashgordon | I don't see anything in stats/zuul | 16:00 |
| clarkb | flashgordon: no I don't think anything is using it today | 16:01 |
| jeblair | flashgordon: but it's all configured, all you need to do is shift the job to 'silent:' instead of 'experimental:' in layout.yaml | 16:01 |
| *** reed has joined #openstack-infra | 16:01 | |
| jeblair | http://graphite.openstack.org/render/?width=586&height=308&_salt=1377878483.97&lineMode=connected&drawNullAsZero=true&from=-72hours&target=stats.timers.zuul.pipeline.experimental.job.gate-tempest-devstack-vm-large-ops.FAILURE.mean&target=stats.timers.zuul.pipeline.experimental.job.gate-tempest-devstack-vm-large-ops.SUCCESS.mean&target=stats.timers.zuul.pipeline.check.job.gate-cinder-pep8.SUCCESS.mean | 16:02 |
| jeblair | hrm | 16:02 |
| jeblair | i don't see data for large-ops, but you can see that cinder job does have data | 16:02 |
| flashgordon | jeblair: neat, it would be really slick if there was a way to just list all the runs | 16:02 |
| flashgordon | for silent pipeline | 16:03 |
| jeblair | flashgordon: yes, as-is you have to hunt for it. you can get results by searching logstash... | 16:04 |
| flashgordon | when ther was one jenkins master that was easy in jenkins but don't know how to check now that we have 3 | 16:04 |
| jeblair | flashgordon: you can get stats from graphite | 16:04 |
| jeblair | flashgordon: you can look at individual changes through gerrit -- follow a check run and then go up a directory on logs.o.o and you'll find the silent runs | 16:04 |
| jeblair | flashgordon: you can still go to jenkins01 and 02 individually and look at the run history there | 16:05 |
| jeblair | flashgordon: (obviously, that may be harder to scale in the future) | 16:05 |
| ttx | jeblair: I get a "no common ancestry" when I try to git review an openstack/governance change | 16:05 |
| anteaya | pleia2 I haven't read all the backscroll from last night, but the upshot that I get is you want to spin up a new puppet dashboard on a test node and if stable transition over to it | 16:06 |
| ttx | jeblair: should I try to clone the repo again ? | 16:06 |
| jeblair | ttx: yay it ran | 16:06 |
| pleia2 | anteaya: thanks for looking into dashboard, was just catching up | 16:06 |
| ttx | Oh | 16:06 |
| anteaya | pleia2: to that end, mgagne pointed me to: https://github.com/sodabrew/puppet-dashboard which accepts the use of ruby 1.9.3 (a more secure ruby than 1.8.7) and looks like it uses rails 3 | 16:06 |
| ttx | you pushed .gitreview, that explains it | 16:06 |
| flashgordon | jeblair: ahh http://graphite.openstack.org/graphlot/?width=586&height=308&_salt=1377878787.545&target=stats.zuul.pipeline.experimental.job.gate-tempest-devstack-vm-large-ops.FAILURE&target=stats.zuul.pipeline.experimental.job.gate-tempest-devstack-vm-large-ops.SUCCESS | 16:07 |
| ttx | jeblair: willfix | 16:07 |
| anteaya | pleia2: at first blush, I vote for a trial of the sodabrew dashboard on a node and see what happens | 16:07 |
| *** bpokorny has quit IRC | 16:08 | |
| jeblair | ttx: yeah, you may have created your own local repo; you'll want to re-clone, and probably just "patch <" whatever you have in your local one | 16:08 |
| anteaya | might alleviate a host of problems of compatibility we are locked into with our current puppet-dashboard, like db upgrades and such | 16:08 |
| pleia2 | anteaya: so ruby 1.8.7 is in Ubuntu 12.04 main, which means canonical has committed to keeping it up security-wise for 5 years from release | 16:08 |
| openstackgerrit | A change was merged to openstack-infra/config: Build Basic Install Guide for openSUSE https://review.openstack.org/42988 | 16:08 |
| jeblair | clarkb: so why did puppet agent --test do something; was it --disabled ? know how i can tell? | 16:08 |
| anteaya | pleia2: the problem is that the ruby community hasn't | 16:09 |
| pleia2 | anteaya: our servers run precise, no newer version of ruby there | 16:09 |
| *** bpokorny has joined #openstack-infra | 16:09 | |
| anteaya | pleia2: the ruby community has moved on to 1.9.3 | 16:09 |
| pleia2 | anteaya: yeah, even when upstream doesn't take care of it, canonical takes on that burden (in theory) | 16:09 |
| clarkb | jeblair: if it was --disabled puppet agent --test would be disabled as well. Maybe the agent got itself stuck (deadlock, busy loop, etc)? | 16:09 |
| anteaya | pleia2: well you know canonical better than I, so I defer to your judgement in that area | 16:10 |
| jeblair | clarkb: ok, i'll just restart it | 16:10 |
| *** ruhe has quit IRC | 16:10 | |
| anteaya | pleia2: then we are into rails 3/1.8.7 issues vs rails 3/1.9.3 issues | 16:10 |
| pleia2 | anteaya: well, it would be interesting to know exactly what the problems are with it and check to see if the appropriate bugs have been filed and/or fixed | 16:10 |
| jeblair | clarkb: (i did not see signs of it getting hung in the usual ways) | 16:10 |
| *** ArxCruz has quit IRC | 16:10 | |
| anteaya | assuming the sodabrew repo is indeed using rails 3, I haven't confirmed that yet | 16:10 |
| openstackgerrit | A change was merged to openstack-infra/config: Add slow heat job to devstack experimental for debug https://review.openstack.org/43971 | 16:10 |
| clarkb | anteaya: pleia2: we will definitely need to start thinking about newer puppet in part because newer ruby is in use in a bunch of places | 16:10 |
| anteaya | pleia2: ahh | 16:11 |
| *** dina_belova has quit IRC | 16:11 | |
| pleia2 | clarkb: are we using newer ruby on precise, or elsewhere? | 16:11 |
| anteaya | yeah, the better work is spent on moving to newer ruby | 16:11 |
| clarkb | puppetlabs doesn't publish python2.7 for hosts without ruby 1.8 | 16:11 |
| anteaya | newer ruby is 1.9.3 | 16:11 |
| clarkb | pleia2: when we tried to use fedora slaves they all use newer ruby | 16:11 |
| *** dina_belova has joined #openstack-infra | 16:11 | |
| anteaya | and considering ruby 2 | 16:11 |
| pleia2 | clarkb: ah, ok | 16:11 |
| clarkb | so we couldn't install puppet 2.7 from puppetlabs | 16:11 |
| * pleia2 nods | 16:12 | |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/devstack-gate: Add support for large_ops tempest test https://review.openstack.org/43779 | 16:12 |
| *** nicedice_ has joined #openstack-infra | 16:12 | |
| anteaya | pleia2: one of the strengths/weaknesses of the ruby community is its speed, speed of development | 16:12 |
| anteaya | pleia2: but they are not known for long term backwards support | 16:13 |
| clarkb | jeblair: https://review.openstack.org/#/c/43991/ I haven't approved that because of the en/en_US thing. I will look into forcing babel to use en | 16:13 |
| anteaya | which is one of the major frustrations with ruby | 16:13 |
| pleia2 | anteaya: yeah, makes it tricky to use packages shipped with an os | 16:13 |
| anteaya | pleia2: yeah, I never do myself | 16:13 |
| anteaya | pleia2: the ruby community basically has an expectation that you upgrade your ruby constantly if not more often | 16:14 |
| ttx | jeblair, anteaya: feel free to early comment on structure/file formats @ https://review.openstack.org/#/c/44489 - before I push that to the rest of the TC | 16:14 |
| pleia2 | anteaya: seems like a lot of work :) | 16:14 |
| anteaya | pleia2: and those that don't aren't granted much sympathy | 16:14 |
| anteaya | pleia2: it is, but keeps you on the bleeding edge, one of the attractions for working with ruby | 16:14 |
| * anteaya reviews | 16:15 | |
| pleia2 | anteaya: tbh most software is like this, backporting is hard, makes it easier for them not to support old things, if we made a special case for everyone we'd never get any work done | 16:15 |
| ttx | jeblair, anteaya: also we are missing the groups | 16:15 |
| pleia2 | not saying we shouldn't make a special case, just that it has to be considered carefully | 16:15 |
| anteaya | pleia2: very true | 16:16 |
| *** dina_belova has quit IRC | 16:16 | |
| ttx | jeblair, anteaya; I mean, the tc-committee-chair group | 16:16 |
| anteaya | I'm just sharing why in my opinion spinning up something that uses 1.8.7 when we have an option that uses 1.9.3 will pigeonhole us soon | 16:16 |
| *** fbo is now known as fbo_away | 16:16 | |
| pleia2 | anteaya: I'm thinking we should get some notes together on what it would take to run this newer dashboard in development (I can toss something up on a test server today) and then chat about it with everyone at a meeting or something, I'd like to have fungi's input too | 16:17 |
| anteaya | ttx yes, I had wondered how the tc-committee group and the tc-committee group get added | 16:17 |
| anteaya | pleia2: + | 16:17 |
| anteaya | pleia2: I can help | 16:17 |
| ttx | anteaya: tech-committee is there. Only missing tech-committee-chair with me in | 16:17 |
| BobBall | hmmmmm - what's the recursive rebase message about? Zuul said: Your change requires a recursive merge to resolve and asked me to rebase. I've done a rebase to gerrit/master but it "just worked" so I don't have much confidence that resubmitting it will work any better? | 16:17 |
| anteaya | if I have shell access to the test server, I can spin it up, or help you | 16:17 |
| ttx | anteaya: would be great if tech-committee-chair could edit tech-committee members too | 16:18 |
| anteaya | ttx ah okay, sorry two directions at the same time :D | 16:18 |
| anteaya | I'm sure you know the dance well | 16:18 |
| clarkb | jeblair: I am beginning to wonder if this is purely a transifex issue. .pot files don't seem to have a specific source locale | 16:18 |
| jeblair | clarkb: governance repo has no acl in gerrit :( | 16:18 |
| * reed about to rant .... | 16:18 | |
| reed | this dude, I send him a private message and he replies including a mailing list! | 16:19 |
| *** markmcclain has quit IRC | 16:19 | |
| jeblair | i will run manage-projects again | 16:19 |
| ttx | reed: yay | 16:19 |
| reed | a private message ... to a mlist! | 16:19 |
| anteaya | BobBall: this is zuul's way of saying that your patch might have merged with a lot of work, but it kicked it out and asked you to rebase to increase the chances of other patches merging | 16:19 |
| * ttx disappears, it's way past beer o'clock time | 16:19 | |
| reed | ttx, back in France? | 16:19 |
| clarkb | BobBall: Gerrit uses jgit which is different than Cgit which makes merges slightly odd. Zuul is now trying to approximate what Gerrit does with jgit. tl;dr local merges may not be a good indication of what Gerrit will do | 16:19 |
| ttx | reed: yep | 16:19 |
| reed | ttx, have a good beer | 16:19 |
| BobBall | heh... nice :D ok | 16:20 |
| clarkb | if anyone knows how to set the source locale for a .pot such that transifex will learn about it correctly please let me know, but I can't find anything. | 16:20 |
| BobBall | well I'll do what it says... I can follow instructions even if I don't understand why ;) | 16:20 |
| clarkb | I really think this may be a transifex thing. Maybe the tx command can set it? | 16:20 |
| anteaya | jeblair: have I erred on my original governance repo creation patch? | 16:20 |
| anteaya | or is this the manage_repos bug you mentioned earlier? | 16:21 |
| *** pblaho has quit IRC | 16:21 | |
| anteaya | BobBall: it helps to chant an incantation and spin a rubber chicken in the air | 16:22 |
| clarkb | http://support.transifex.com/customer/portal/articles/996211-pushing-new-translations doesn't indicate a way to set the source locale | 16:22 |
| * BobBall gets the rubber chicken out again | 16:22 | |
| BobBall | Fourth time I've needed to use that today! | 16:22 |
| anteaya | ha ha ha | 16:22 |
| anteaya | helps to keep the chicken handy | 16:22 |
| *** dina_belova has joined #openstack-infra | 16:23 | |
| *** markmcclain has joined #openstack-infra | 16:24 | |
| pleia2 | anteaya: it's going to take me some time to get around to -dashboard today, in the meantime can you update https://bugs.launchpad.net/openstack-ci/+bug/1218631 with version info, location, etc of new dashboard we should be trying? | 16:26 |
| uvirtbot | Launchpad bug 1218631 in openstack-ci "Build new puppet-dashboard server" [Undecided,New] | 16:26 |
| openstackgerrit | A change was merged to openstack-infra/config: Fix new sqlalchemy-migrate jobs https://review.openstack.org/44021 | 16:27 |
| anteaya | pleia2: can do, I think I will try to read a bit more first but I will update with what I have soon | 16:27 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/devstack-gate: Add support for large_ops tempest test https://review.openstack.org/43779 | 16:27 |
| * anteaya heads for lunch | 16:27 | |
| pleia2 | anteaya: perfect, thank you! | 16:27 |
| anteaya | :D | 16:27 |
| jeblair | clarkb: if manage-projects fails to do anything with the acl file, what do we do? | 16:28 |
| *** BobBall is now known as BobBallAway | 16:28 | |
| clarkb | jeblair: was there a traceback when run manually? | 16:28 |
| jeblair | (after mordred gets back, i don't intend to approve any new projects until this is fixed) | 16:28 |
| jeblair | clarkb: nope, just silently exited | 16:28 |
| jeblair | except for that one error from earlier | 16:28 |
| clarkb | jeblair: usually fails to do things with ACL files are a result of a syntax error in the file | 16:29 |
| jeblair | but no traceback | 16:29 |
| clarkb | I thought that generated a traceback when the git push failed, but I could be mistaken | 16:29 |
| openstackgerrit | A change was merged to openstack-infra/config: Generate heat docs on check and gate https://review.openstack.org/43234 | 16:29 |
| *** hemnafk is now known as hemna | 16:30 | |
| jeblair | of course manage-projects --help | 16:30 |
| jeblair | just runs manage-projects again | 16:30 |
| jeblair | github.GithubException.GithubException: 403 {u'message': u'API rate limit exceeded for openstack-project-creator. See http://developer.github.com/v3/#rate-limiting for details.'} | 16:30 |
| jeblair | that's exciting | 16:30 |
| clarkb | nice | 16:31 |
| clarkb | there are print messages for the ACL push failure (so not tracebacks but you should get a message) | 16:31 |
| *** gyee has joined #openstack-infra | 16:32 | |
| jeblair | clarkb: i can not think of a way to debug this | 16:33 |
| *** thomasbiege has quit IRC | 16:33 | |
| jeblair | this seems like a black box. no logging, no debug mode, no verbose mode, no noop mode, nothing. | 16:33 |
| clarkb | it has "logging" | 16:34 |
| jeblair | it logs to /dev/null | 16:34 |
| jeblair | so it has webscale logging. great. | 16:34 |
| clarkb | jeblair: creating a pruned projects.yaml so that you are only interacting with the problem project and running in pdb may be the immediate answer | 16:35 |
| clarkb | that will help keep the github rate limiter happy and simplify debugging | 16:36 |
| *** nati_ueno has joined #openstack-infra | 16:37 | |
| *** nati_ueno has quit IRC | 16:38 | |
| *** nati_ueno has joined #openstack-infra | 16:38 | |
| *** ArxCruz has joined #openstack-infra | 16:44 | |
| *** dina_belova has quit IRC | 16:47 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/jeepyb: Add options to manage-projects https://review.openstack.org/44496 | 16:48 |
| *** jfriedly has joined #openstack-infra | 16:48 | |
| *** dina_belova has joined #openstack-infra | 16:48 | |
| jeblair | clarkb: apparently what was needed was for it to run several times with debugging output. <sigh> | 16:48 |
| jeblair | clarkb: because as soon as i did that ^ the acls showed up in gerrit. | 16:48 |
| openstackgerrit | A change was merged to openstack-infra/config: Translation update jobs now supports any org https://review.openstack.org/44212 | 16:49 |
| openstackgerrit | A change was merged to openstack-infra/config: Update Japanese Ops Guide build job. https://review.openstack.org/43734 | 16:49 |
| jeblair | also, wow the neutron people really want to sudo. | 16:49 |
| clarkb | jeblair: :/ but ++ to your change. reviewing it now | 16:49 |
| clarkb | jeblair: yes ovcsctl all over the plce | 16:49 |
| ArxCruz | ALL: I'm trying to configure zuul in my ifrastructure to listen review.openstack.org streams. I configure it to use my account on review.openstack.org and my key. However, I'm wondering, how can I autehticate since zuul is running as a daemon… | 16:51 |
| jeblair | ArxCruz: in production you'd use a service account with a key with no passphrase | 16:52 |
| *** dina_belova has quit IRC | 16:52 | |
| jeblair | ArxCruz: for testing, you can use ssh-agent which will avoid you needing to enter the passphrase | 16:52 |
| *** dina_belova has joined #openstack-infra | 16:53 | |
| *** ruhe has joined #openstack-infra | 16:54 | |
| clarkb | jeblair: jeepy manage-project change reviweed | 16:55 |
| jeblair | clarkb: any idea what that error is? | 16:55 |
| clarkb | jeblair: that was causing it to fail when you ran it? | 16:56 |
| jeblair | clarkb: no, that is causing it to not install gerritlib. | 16:56 |
| clarkb | oh is that why the tests failed? I didn't even look there yet. Looking now | 16:57 |
| jeblair | clarkb: and so you know, i don't plan on fixing manage_projects. i have too much on my plate and creating stackforge projects is not a priority for me. | 16:57 |
| jeblair | mordred can fix it when he gets back | 16:57 |
| clarkb | jeblair: thats fine | 16:57 |
| clarkb | jeblair: it almost looks like it is complaining that the real name of the gerritlib is gerritlib | 16:58 |
| clarkb | dstufft: http://logs.openstack.org/96/44496/1/check/gate-jeepyb-pep8/9c379e0/console.html | 16:58 |
| dstufft | clarkb: http://status.python.org/incidents/4h01t5v8r343 | 16:59 |
| *** pcrews has quit IRC | 16:59 | |
| *** wenlock_ has joined #openstack-infra | 16:59 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/jeepyb: Add options to manage-projects https://review.openstack.org/44496 | 16:59 |
| clarkb | dstufft: that would cause "Real name of requirement gerritlib is gerritlib" ? also, ignore me if you are busy :) | 17:00 |
| annegentle_ | what's a way to trigger the Japanese Ops Guide build? | 17:00 |
| ArxCruz | jeblair: How can I use ssh-agent ? editing the /etc/init.d/zuul script ? | 17:00 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/jeepyb: Add options to manage-projects https://review.openstack.org/44496 | 17:01 |
| clarkb | annegentle_: it is in the post queue so you will need to merge a change to that project | 17:01 |
| annegentle_ | clarkb: even a change to the english guide builds the jp one? | 17:01 |
| clarkb | annegentle_: I would give JJB and Zuul anohter 20 minutes to apply their changes | 17:01 |
| clarkb | annegentle_: yes, it is the same project | 17:01 |
| jeblair | ArxCruz: you can google how to use ssh-agent; nothing special is needed for zuul, but if you're running it from an init script, you should get a service account. | 17:01 |
| clarkb | ArxCruz: no, ssh-agent will run independent of zuul. | 17:01 |
| annegentle_ | clarkb: ok, yeah, that's logical :) der, thanks | 17:01 |
| ArxCruz | krtaylor: ^ | 17:02 |
| dstufft | clarkb: it might yea, if /simple/gerritlib/ was unavailable and it fell back to looking at /simple/ and discovered "gerritlib" | 17:02 |
| *** sarob has joined #openstack-infra | 17:05 | |
| flashgordon | do you think the failures from the regular tests are related to my changes: https://review.openstack.org/#/c/43779/ | 17:07 |
| *** zeus has joined #openstack-infra | 17:07 | |
| clarkb | mtreinish: ^ http://logs.openstack.org/79/43779/11/check/gate-tempest-devstack-vm-full/f66dfdb/console.html | 17:08 |
| *** mgagne has quit IRC | 17:08 | |
| *** mgagne has joined #openstack-infra | 17:09 | |
| *** mgagne has quit IRC | 17:09 | |
| *** mgagne has joined #openstack-infra | 17:09 | |
| *** nicedice_ has quit IRC | 17:09 | |
| flashgordon | it seems to always happen | 17:10 |
| *** Ryan_Lane has joined #openstack-infra | 17:10 | |
| clarkb | flashgordon: you shouldn't have changed anything in normal tempest though. The localrc changes only happen for your test | 17:11 |
| clarkb | flashgordon: I would see if mtreinish recognizes the failures | 17:11 |
| flashgordon | clarkb: thats what I thought but maybe I goofed | 17:12 |
| flashgordon | I did make one change actaully | 17:12 |
| clarkb | oh you changed fixed network size | 17:12 |
| clarkb | but didn't change the fixed range | 17:12 |
| clarkb | flashgordon: that may be the problem | 17:12 |
| flashgordon | fixed_range isn't used | 17:12 |
| clarkb | it isn't? | 17:12 |
| flashgordon | if i remember correctly | 17:12 |
| clarkb | I thought it was necessary to prevent VM ip collisions with HPCloud IOPs | 17:13 |
| flashgordon | actually I can drop the fixed_range change | 17:13 |
| clarkb | s/IOPs/IPs/ | 17:13 |
| flashgordon | err network_size | 17:13 |
| clarkb | flashgordon: if you bump the network size you need to bump the fixed range too I think | 17:13 |
| *** derekh has quit IRC | 17:13 | |
| clarkb | because a /24 has less than 1024 usable addresses in it | 17:13 |
| flashgordon | clarkb: yeah fixed_range isn't in nova.conf | 17:14 |
| *** adalbas has joined #openstack-infra | 17:14 | |
| flashgordon | but I don't need to change it anymroe | 17:14 |
| flashgordon | since will be spinning up <256 VMs now | 17:14 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/devstack-gate: Add support for large_ops tempest test https://review.openstack.org/43779 | 17:15 |
| flashgordon | clarkb: looks like the magic number of VMs I can spin up is 200 | 17:15 |
| flashgordon | and 250 fails | 17:15 |
| *** david-lyle is now known as david-lyle_afk | 17:16 | |
| *** david-lyle_afk is now known as david-lyle_away | 17:16 | |
| mgagne | flashgordon: pardon my ignorance, is large_ops test using neutron? | 17:16 |
| flashgordon | mgagne: it is not | 17:16 |
| mgagne | flashgordon: ok thanks | 17:17 |
| flashgordon | mgagne: no reason it cannot though | 17:17 |
| *** beagles is now known as seagulls | 17:17 | |
| flashgordon | mgagne: this is the localrc used http://logs.openstack.org/79/43779/8/experimental/gate-tempest-devstack-vm-large-ops/1aed0c3/logs/localrc.txt.gz | 17:17 |
| clarkb | flashgordon: it would be interesting to see if neutron would pass :) | 17:17 |
| mgagne | flashgordon: I ask the question because we are facing serious scalability issues with quantum (grizzly) and are still trying to figure out the reason | 17:18 |
| flashgordon | clarkb mgagne: I picked n-net because that is the default setup | 17:18 |
| flashgordon | mgagne: let me try a neutron version | 17:18 |
| clarkb | flashgordon: my curiousity comes from some of the problems neutron has with normal tempest (different sort of test though) | 17:20 |
| mtreinish | clarkb, flashgordon: I don't recognize any of those failures | 17:21 |
| *** shardy is now known as shardy_afk | 17:21 | |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/devstack-gate: Add support for large_ops tempest test https://review.openstack.org/43779 | 17:23 |
| clarkb | pleia2: https://review.openstack.org/#/c/44209/ though the parents of that commit also have -1's so no rush | 17:23 |
| flashgordon | mtreinish: Ithink it is related to the network size change | 17:23 |
| flashgordon | which I dropped | 17:23 |
| flashgordon | clarkb: ^ neutron version of test | 17:23 |
| pleia2 | clarkb: my comment from the 3rd patch set is still valid, hiera is misspelled | 17:24 |
| flashgordon | mgagne: ^ | 17:24 |
| mgagne | flashgordon: waiting for results =) | 17:24 |
| flashgordon | mgagne: so just a warning the logs aren't complete for this test due to https://bugs.launchpad.net/devstack/+bug/1218575 | 17:25 |
| uvirtbot | Launchpad bug 1218575 in devstack "When using NUMBER_FAKE_NOVA_COMPUTE only one copy of n-cpu is logged" [Undecided,New] | 17:25 |
| flashgordon | so if it fails n-cpu logs are missing | 17:25 |
| flashgordon | for 9/10 n-cpus | 17:25 |
| flashgordon | mgagne: test takes ~10-15 min | 17:26 |
| clarkb | pleia2: new patchset must have been part of a rebase and not an actual edit | 17:26 |
| clarkb | pleia2: thank you for looking | 17:26 |
| wenlock_ | puppet install question on very bare bones ubuntu system, has anyone seen this error before? | 17:27 |
| wenlock_ | puppet : Depends: puppet-common (= 2.7.23-1puppetlabs1) but it is not going to be installed | 17:27 |
| wenlock_ | E: Unable to correct problems, you have held broken packages. | 17:27 |
| mgagne | flashgordon: I talked to my coworkers and we think we figure out part of our problem. (to be validated) We create a bunch of instances in the same tenant. Those instances share a common security groups which contains a rule: all traffic from members of the security groups. We believe it has the potential to create some kind of exponential workload on quantum, each instance triggering the update on the previously created instances. | 17:27 |
| mgagne | wenlock_: you need to pin puppet-common too | 17:28 |
| clarkb | wenlock_: the isntall_puppet.sh script should do this for you | 17:28 |
| flashgordon | mgagne: hmm interesting, we can test this here by updating the large_ops test to do that | 17:29 |
| flashgordon | mgagne: https://github.com/openstack/tempest/blob/master/tempest/scenario/test_large_ops.py | 17:29 |
| clarkb | mgagne: flashgordon I think that issue exists in not quantum/neutron as well | 17:29 |
| flashgordon | mgagne: want to update tempest to do that case | 17:30 |
| mgagne | flashgordon: I have yet to have tempest installed in our infra =) | 17:30 |
| flashgordon | in the file above ^, because its a superset of what we do now | 17:30 |
| mgagne | flashgordon: our backlog starts to drop requests =) | 17:30 |
| flashgordon | flashgordon: that sounds right, RPC timeouts ? | 17:31 |
| mgagne | flashgordon: talking to yourself? | 17:31 |
| flashgordon | mgagne: I guess so | 17:31 |
| flashgordon | TGIF | 17:31 |
| *** vogxn has quit IRC | 17:31 | |
| mgagne | flashgordon: backlog: TODO list, tempest is low in the list =( | 17:31 |
| flashgordon | mgagne: ohh that backlog | 17:31 |
| *** morganfainberg|a is now known as morganfainberg | 17:32 | |
| clarkb | darn backlogs | 17:32 |
| mgagne | clarkb: tell me about it =) | 17:32 |
| mgagne | clarkb: but our customers need that heated cup-holder first! Please stop working on airbags now. | 17:33 |
| flashgordon | mgagne: this is the rule you use: http://paste.openstack.org/show/45487/ | 17:35 |
| flashgordon | if so this is an easy change, just add a rule to default security group before launching VMs | 17:36 |
| *** SlickNik has quit IRC | 17:37 | |
| *** dina_belova has quit IRC | 17:37 | |
| *** SlickNik has joined #openstack-infra | 17:37 | |
| *** dina_belova has joined #openstack-infra | 17:38 | |
| flashgordon | clarkb mgagne: neutron just failed where n-net worked http://logs.openstack.org/79/43779/13/experimental/gate-tempest-devstack-vm-large-ops/89f96f8/ | 17:39 |
| clarkb | flashgordon: make fake libvirt isn't good enough for it? | 17:39 |
| flashgordon | clarkb: huh? | 17:40 |
| flashgordon | markmcclain: ^ | 17:40 |
| mgagne | flashgordon: OverQuota: Quota exceeded for resources: ['port'] =) | 17:40 |
| flashgordon | mgagne: doh | 17:41 |
| flashgordon | good catch | 17:41 |
| *** senk has quit IRC | 17:41 | |
| *** dina_belova has quit IRC | 17:42 | |
| *** sandywalsh_ has quit IRC | 17:43 | |
| *** markmcclain has quit IRC | 17:43 | |
| *** markmcclain has joined #openstack-infra | 17:46 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/devstack-gate: Remove node management from devstack-gate https://review.openstack.org/44503 | 17:46 |
| *** seagulls is now known as beagles | 17:49 | |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/devstack-gate: Add support for large_ops tempest test https://review.openstack.org/43779 | 17:50 |
| *** beagles is now known as seagulls | 17:50 | |
| *** [wchrisj_ has joined #openstack-infra | 17:50 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add Gerrit ACLs for clouddocs maven plugin project https://review.openstack.org/44471 | 17:50 |
| *** ruhe has quit IRC | 17:51 | |
| *** neelashah has left #openstack-infra | 17:55 | |
| *** sandywalsh_ has joined #openstack-infra | 17:56 | |
| *** dina_belova has joined #openstack-infra | 17:58 | |
| openstackgerrit | Elizabeth Krumbach Joseph proposed a change to openstack-infra/config: Update references to GerritJenkinsGit wiki https://review.openstack.org/44507 | 17:59 |
| *** xBsd has joined #openstack-infra | 18:01 | |
| *** pcrews has joined #openstack-infra | 18:03 | |
| *** wenlock_ has quit IRC | 18:06 | |
| *** dukhlov has quit IRC | 18:08 | |
| clarkb | jeblair: https://review.openstack.org/#/c/42643/ LGTM. I did not approve it as it requires a corresponding zuul layout change | 18:18 |
| clarkb | jeblair: should we try pushing that through today? | 18:18 |
| Alex_Gaynor | clarkb: Can I ask you to take a look at https://review.openstack.org/#/c/44369/ and https://review.openstack.org/#/c/44246/ when you have a chance? | 18:20 |
| clarkb | Alex_Gaynor: sure | 18:20 |
| Alex_Gaynor | thanks | 18:20 |
| clarkb | Alex_Gaynor: all of those tests run under pypy and pass? | 18:21 |
| Alex_Gaynor | clarkb: swiftclient ones definitely do, I didn't double check jd__ for hacking and pbr though | 18:21 |
| *** Protux has quit IRC | 18:21 | |
| *** Protux has joined #openstack-infra | 18:22 | |
| clarkb | Alex_Gaynor: jd__: it would be great to double check the others because once the hacking and pbr change merges pypy will gate | 18:22 |
| Alex_Gaynor | clarkb: I'll doubel check for him (assuming he's not around to verify now) | 18:22 |
| clarkb | ya I think he is near/in UTC so this would be really late on Friday | 18:23 |
| clarkb | Alex_Gaynor: great let me know. I am slightly worried about pbr | 18:23 |
| openstackgerrit | A change was merged to openstack-infra/config: Added PyPy to gate and check for python-swiftclient https://review.openstack.org/44369 | 18:24 |
| *** sarob has quit IRC | 18:24 | |
| Alex_Gaynor | clarkb: hmm, is there something more to running the pbr tests than `tox`? | 18:24 |
| *** sarob has joined #openstack-infra | 18:24 | |
| clarkb | Alex_Gaynor: I don't think pbr has pypy in its default envs list. `tox -epypy` should work though | 18:25 |
| Alex_Gaynor | clarkb: yeah, I just meant, running `tox` doesn't work for me in the pbr repo: http://bpaste.net/show/127822/ | 18:25 |
| clarkb | that should work. Jenkins runs the tests that way | 18:26 |
| clarkb | though you do have nested virtualenvs looks like | 18:27 |
| Alex_Gaynor | yeah, I'm running tox inside another venv | 18:27 |
| Alex_Gaynor | which usually works ok | 18:27 |
| clarkb | Alex_Gaynor: I think pbr in the outer venv might be having trouble installing pbr | 18:27 |
| clarkb | Alex_Gaynor: try with a clean outer venv? | 18:27 |
| *** markmcclain has quit IRC | 18:30 | |
| Alex_Gaynor | clarkb: so hacking definitely works | 18:31 |
| jeblair | clarkb: well, i don't really want to touch zuul again for a while; we could let those sit; make a feature branch; or pin prod to a sha | 18:31 |
| clarkb | jeblair: wfm | 18:32 |
| clarkb | I just figured if it was going to go in it would have to be today | 18:33 |
| jeblair | clarkb: yep | 18:33 |
| openstackgerrit | Alex Gaynor proposed a change to openstack-dev/hacking: Use the name for this attribute that exists on Python3. https://review.openstack.org/44512 | 18:33 |
| openstackgerrit | Alex Gaynor proposed a change to openstack-dev/hacking: Added PyPy env to tox https://review.openstack.org/44513 | 18:34 |
| clarkb | jeblair: https://review.openstack.org/#/c/42645/7 the comments say I left an inline comment on patchset 7 but it doesn't show up for me | 18:34 |
| clarkb | jeblair: can you see my inline comment? | 18:34 |
| clarkb | I am 98% sure I know what the comment was, but it is weird that I can't refer back to it | 18:34 |
| jeblair | clarkb: i don't see the flag next to the file | 18:34 |
| jeblair | clarkb: which file? | 18:34 |
| clarkb | jeblair: neither do I. It would've been on the smtp.py file | 18:35 |
| clarkb | I think | 18:35 |
| clarkb | around how error handling is done | 18:35 |
| jeblair | clarkb: nope, don't see it. ;/ | 18:35 |
| clarkb | weird | 18:35 |
| Alex_Gaynor | clarkb: I can't seem to get pbr tests running at all :/ | 18:35 |
| clarkb | Alex_Gaynor: even with a clean virtualenv? oh I wonder, how new is your setuptools in that virtualenv? | 18:36 |
| clarkb | setuptools >=1.0 seems to have broken us | 18:36 |
| jeblair | reed: can you poke mrmartin and have him look at clark's comment on https://review.openstack.org/#/c/42608/ ? | 18:36 |
| Alex_Gaynor | clarkb: it's 0.9.8, I guess my system venv is old | 18:36 |
| reed | jeblair, done | 18:36 |
| clarkb | Alex_Gaynor: what version of python failed in that paste? | 18:38 |
| Alex_Gaynor | clarkb: 2.6, also happens if I manually run testr under 2.7 though | 18:39 |
| clarkb | could it be an OS X thing? | 18:39 |
| Alex_Gaynor | possibly! | 18:39 |
| *** gaelL has quit IRC | 18:40 | |
| clarkb | it is trying to do subprocess stuff which is at least slightly OS specific | 18:40 |
| Alex_Gaynor | yeah, I'm investigating | 18:40 |
| Alex_Gaynor | clarkb: so yeah. something is hella messed up: http://bpaste.net/show/AL79IlsuJCHtJiUdItfF/ | 18:41 |
| clarkb | it is splitting on every byte with a null byte? | 18:43 |
| clarkb | that is awesome | 18:43 |
| Alex_Gaynor | I think it's looking at the underly byte representation of the unicode, instead of ... anything remotely sane | 18:43 |
| Alex_Gaynor | so it's utf-16, so of course there are some \x00 bytes | 18:43 |
| Alex_Gaynor | amazing | 18:44 |
| clarkb | oh yes | 18:44 |
| clarkb | shlex doesn't work with unicode on py26 | 18:44 |
| Alex_Gaynor | this is 2.7 | 18:44 |
| clarkb | but you said it does the same on py27? | 18:44 |
| clarkb | hmm | 18:44 |
| Alex_Gaynor | yup | 18:44 |
| clarkb | jeblair: what do you think about WIP'ing jhesketh's changes? I am semi afraid mordred might do drive by approvals next week :) | 18:45 |
| *** gaelL has joined #openstack-infra | 18:46 | |
| jeblair | clarkb: heh, probably a good idea | 18:47 |
| *** david-lyle_away is now known as david-lyle | 18:48 | |
| flashgordon | clarkb: so I think I have the large_ops test working | 18:51 |
| flashgordon | can you review https://review.openstack.org/#/c/44343 so I can update the d-g patch accordingly | 18:51 |
| clarkb | sure | 18:51 |
| clarkb | let me finish reviewing jhesketh_'s second zuul change then I will look at the large ops chane | 18:51 |
| flashgordon | clarkb: thanks | 18:51 |
| clarkb | flashgordon: done. that was easy | 18:53 |
| clarkb | Alex_Gaynor: your hacking change seems to have not liked __code__ | 18:54 |
| *** thomasbiege1 has joined #openstack-infra | 18:54 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add word tempest to large_ops variable https://review.openstack.org/44343 | 18:55 |
| flashgordon | clarkb: thanks | 18:56 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/devstack-gate: Add support for large_ops tempest test https://review.openstack.org/43779 | 18:56 |
| clarkb | jeblair: did you want to look at the mysql backups on review-dev and see if they look good to you? Is merging the change for review.o.o and wiki something you are comfortable doing next week or should I focus on that now? | 18:56 |
| *** UtahDave has quit IRC | 18:57 | |
| jeblair | clarkb: either way; i'll look at review-dev now | 18:59 |
| *** mriedem has quit IRC | 18:59 | |
| *** mriedem1 has joined #openstack-infra | 18:59 | |
| jeblair | clarkb: i think it needs work :( | 19:00 |
| jeblair | -rw-r--r-- 1 root root 0 Aug 30 06:30 review-dev.sql.gz | 19:00 |
| jeblair | -rw-r--r-- 1 root root 1234398 Aug 30 00:00 review-dev.sql.gz.1 | 19:00 |
| *** datsun180b has quit IRC | 19:00 | |
| clarkb | oh that | 19:01 |
| Alex_Gaynor | clarkb: will investigate after lunch | 19:01 |
| clarkb | the dump and the rotate run at different times and logrotate insists that the file actually exist | 19:01 |
| *** wenlock has joined #openstack-infra | 19:02 | |
| clarkb | jeblair: so at 0000 review-dev.sql.gz is created with many bytes by mysqldump. then at 0630 review-dev.sql.gz is moved to review-dev.sql.gz.1 and a zero byte review-dev.sql.gz is craeted | 19:02 |
| *** wenlock has quit IRC | 19:02 | |
| *** wenlock has joined #openstack-infra | 19:03 | |
| jeblair | oh, ok | 19:03 |
| Mithrandir | hmm, I submitted a draft branch for review (for jjb), doesn't that show up on irc? | 19:06 |
| annegentle_ | clarkb: hm. I triggered the build for operations-guide, but looks like it hasn't run since Aug. 5th. https://jenkins.openstack.org/view/Openstack-manuals/job/openstack-operations-guide/ | 19:07 |
| jeblair | clarkb: backups lgtm | 19:07 |
| annegentle_ | clarkb: and the JP version has never attempted to be built | 19:07 |
| *** UtahDave has joined #openstack-infra | 19:07 | |
| jeblair | Mithrandir: i don't believe it does; no one can view the change unless you've explicitly added them. we pretty much hate drafts around here, and use work-in-progress instead | 19:08 |
| jeblair | Mithrandir: (they never work like anyone expects them to) | 19:08 |
| Mithrandir | oh, ok. | 19:08 |
| Mithrandir | how do I submit a WIP branch? | 19:08 |
| jeblair | Mithrandir: you have to mark it wip after submitting it | 19:08 |
| Mithrandir | ok, submitting one now. I'm on 3G so this takes a while. | 19:09 |
| *** dina_belova has quit IRC | 19:09 | |
| *** dina_belova has joined #openstack-infra | 19:10 | |
| clarkb | annegentle_: we have three jenkins masters now. jenkins.o.o doesn't run many tests | 19:10 |
| clarkb | annegentle_: you will want to look on jenkins01 and jenkins02 | 19:10 |
| clarkb | jeblair: good to approve https://review.openstack.org/#/c/44129/1 then? | 19:11 |
| *** dina_belova has quit IRC | 19:12 | |
| *** dina_belova has joined #openstack-infra | 19:13 | |
| *** SergeyLukjanov has quit IRC | 19:13 | |
| annegentle_ | clarkb: oh! | 19:14 |
| *** SergeyLukjanov has joined #openstack-infra | 19:15 | |
| *** ruhe has joined #openstack-infra | 19:15 | |
| annegentle_ | clarkb: ok, still have the problem of the slicing script going to the openstack-manuals repo error while opening "./doc/src/docbkx/openstack-ops/locale/ja.po" for reading: No such file or directory | 19:15 |
| annegentle_ | clarkb: I'll cirlce back with Daisy on the openstack-docs mailing list | 19:15 |
| Mithrandir | gah: ! [remote rejected] HEAD -> refs/publish/master/purge-old-jobs (no new changes) | 19:17 |
| *** boris-42 has quit IRC | 19:17 | |
| * Mithrandir stabs gerrit with a rusty spoon | 19:17 | |
| *** ruhe has quit IRC | 19:18 | |
| Alex_Gaynor | clarkb: hmm, so they pass locally for me | 19:20 |
| Mithrandir | hmm, I don't think it shows up here even when I published it | 19:22 |
| Mithrandir | so, https://review.openstack.org/#/c/44525/ ; reviews welcome, but I don't really think it's in a mergeable state | 19:23 |
| openstackgerrit | Alex Gaynor proposed a change to openstack-dev/hacking: Use the name for this attribute that exists on Python3 https://review.openstack.org/44512 | 19:23 |
| *** dina_belova has quit IRC | 19:27 | |
| *** dina_belova has joined #openstack-infra | 19:27 | |
| Alex_Gaynor | clarkb: https://review.openstack.org/#/c/44512 looks good now, I claim it was a phantom in the machine :/ (I hate not knowing why it was broken before) | 19:28 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/devstack-gate: Add support for large_ops tempest test https://review.openstack.org/43779 | 19:28 |
| *** dina_belova has quit IRC | 19:28 | |
| *** dina_bel_ has joined #openstack-infra | 19:28 | |
| *** SergeyLukjanov has quit IRC | 19:30 | |
| *** nicedice_ has joined #openstack-infra | 19:33 | |
| *** sarob has quit IRC | 19:34 | |
| *** thomasm has quit IRC | 19:35 | |
| *** SergeyLukjanov has joined #openstack-infra | 19:38 | |
| clarkb | annegentle_: I am on the list now so will try to follow along :) | 19:39 |
| clarkb | Mithrandir: gerrit is very specific about not pushing redundant patchsets... that means sometimes you push a trivial change. It is also picky about not merging anything but the latest patchset | 19:40 |
| *** bpokorny has quit IRC | 19:43 | |
| *** dkehn_ has joined #openstack-infra | 19:43 | |
| *** openstack has joined #openstack-infra | 19:43 | |
| *** bpokorny has joined #openstack-infra | 19:44 | |
| *** dkehn has quit IRC | 19:45 | |
| lifeless | clarkb: oh hai :) | 19:45 |
| lifeless | clarkb: is there meant to be anything on puppet-dashboard:80/ ? | 19:46 |
| lifeless | clarkb: and is the :3000 service meant to be secured in some fashion ? | 19:46 |
| lifeless | clarkb: http://puppet-dashboard.testing-cabal.org:3000/ lists lots of pending tasks; I presume there is some undocumented setup ? | 19:47 |
| clarkb | lifeless: I bet the log process workers are not running. | 19:47 |
| *** gyee has quit IRC | 19:48 | |
| *** dkehn_ is now known as dkehn | 19:48 | |
| clarkb | lifeless: the way dashboard works is it has a web frontend that displays what you see and stores reports from clients. Then the worker processes grab the stored logs do stuff with them and updte the mysql DB | 19:48 |
| clarkb | lifeless: look in /etc/init.d/ for puppet things that aren't puppet agent and start that | 19:49 |
| clarkb | if you have it running on the same host as the master I thnk you need to secure it as it can operate in write mode. When run on its own host I believe it is semi secure as it is read only . Do note that puppet logs go there though which may be problematic if you need to keep thing secret | 19:50 |
| lifeless | clarkb: so does openstack need to keep it's one secure? | 19:50 |
| clarkb | we don't keep ours secure | 19:50 |
| lifeless | clarkb: other than it being broken-ass right now ? | 19:50 |
| lifeless | :) | 19:50 |
| clarkb | lifeless can you address jeblair and pleia2's comments near the bottom of your stack? once that is done I think we can merge another batch of changes | 19:52 |
| *** svarnau has joined #openstack-infra | 19:52 | |
| Mithrandir | clarkb: arguably, it was different, since it was no longer a draft. | 19:54 |
| jeblair | Mithrandir: if you just wanted to wip your draft, you could publish it, then mark it wip | 19:55 |
| Mithrandir | jeblair: I ended up doing that, or rather, marking the draft as published and then wip-ing it. | 19:55 |
| clarkb | jeblair: logrotate take a couple days to do its thing with the mysqldumps. I am leaning towards erring on the side of caution and approving the backup change next week. Avoids anything that may happen over the long weekend | 19:59 |
| clarkb | if however you feel that is too paranoid let me know and I will approve (the change has been running on 3 hosts now) | 20:00 |
| *** markmcclain has joined #openstack-infra | 20:00 | |
| jeblair | clarkb: i like your plan | 20:04 |
| *** thomasbiege1 has quit IRC | 20:07 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Ensure reports directory is setup correctly. https://review.openstack.org/44391 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document hiera setup a bit more comprehensively. https://review.openstack.org/44357 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make the adduser commands copy-pasteable. https://review.openstack.org/44270 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Install launch/ dependencies on puppetmaster. https://review.openstack.org/44384 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Add a pointer to launch/README for root operators. https://review.openstack.org/44163 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make setting up a Puppetmaster easier. https://review.openstack.org/44209 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make config updates work with rebased branches. https://review.openstack.org/44383 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Fix launch-node for non-official environments. https://review.openstack.org/44449 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document bootstrapping a new infrastructure. https://review.openstack.org/44262 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Start documenting running your own CI infra. https://review.openstack.org/44177 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Create ci-launch directory on the puppetmaster. https://review.openstack.org/44394 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make third-party imports clearer in launch/. https://review.openstack.org/44395 | 20:12 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document setting the correct hostname. https://review.openstack.org/44393 | 20:12 |
| lifeless | jeblair: clarkb: whats the next service to bring up after the dashboard? Gerrit or zuul? | 20:12 |
| lifeless | clarkb: I'll file a bug about the dashboard | 20:12 |
| clarkb | lifeless: gerrit | 20:12 |
| clarkb | gerrit -> jenkins -> zuul I think | 20:13 |
| *** sarob has joined #openstack-infra | 20:13 | |
| *** mriedem has joined #openstack-infra | 20:14 | |
| *** mriedem1 has quit IRC | 20:16 | |
| *** xBsd has quit IRC | 20:17 | |
| *** pcm__ has joined #openstack-infra | 20:20 | |
| *** gyee has joined #openstack-infra | 20:22 | |
| *** adalbas has quit IRC | 20:22 | |
| *** thomasbiege1 has joined #openstack-infra | 20:23 | |
| *** pcm_ has quit IRC | 20:24 | |
| *** adalbas has joined #openstack-infra | 20:26 | |
| lifeless | clarkb: so whats meant to be on port 80 on the dashboard? | 20:28 |
| clarkb | lifeless: right now nothing | 20:28 |
| clarkb | lifeless: there is history behind that and it has to do with mordred wresting dashboard and dashboard winning the port argument | 20:28 |
| openstackgerrit | A change was merged to openstack-infra/config: Fix stackforge description for Neutron module https://review.openstack.org/44361 | 20:28 |
| lifeless | so why is the port open ? | 20:28 |
| clarkb | lifeless: he couldn't get dashboard to reliably serve out of port 80 for some reason | 20:28 |
| lifeless | https://bugs.launchpad.net/openstack-ci/+bug/1219076 filed | 20:29 |
| uvirtbot | Launchpad bug 1219076 in openstack-ci "puppet-dashboard does not run background services" [Undecided,New] | 20:29 |
| lifeless | I haven't seen 'start an init.d service' in puppet yet to cargo cult the fix | 20:29 |
| *** dina_bel_ has quit IRC | 20:30 | |
| *** dina_belova has joined #openstack-infra | 20:30 | |
| *** thomasbiege1 has quit IRC | 20:31 | |
| openstackgerrit | A change was merged to openstack-dev/hacking: Add pypy on tox envlist https://review.openstack.org/44248 | 20:31 |
| lifeless | jeblair: clarkb: I can rebase that full stack if you want to get it in w/o merge commits | 20:32 |
| lifeless | jeblair: I believe I've answered the questions you and pleia2 posed | 20:32 |
| *** sarob_ has joined #openstack-infra | 20:33 | |
| *** sarob has quit IRC | 20:33 | |
| *** Shrews has quit IRC | 20:34 | |
| *** SergeyLukjanov has quit IRC | 20:34 | |
| jeblair | lifeless: no need, and thanks. | 20:34 |
| clarkb | lifeless: merge commits don't bother me. Was the heira typo fixed? | 20:34 |
| clarkb | should be hiera | 20:34 |
| *** dina_belova has quit IRC | 20:35 | |
| jeblair | lifeless: re https://review.openstack.org/#/c/44262/9/doc/source/running-your-own.rst | 20:35 |
| jeblair | lifeless: jenkins01 is not a slave, it's a jenkins master (#2 out of 3) | 20:36 |
| jeblair | lifeless: jenkins.o.o, jenkins01.o.o, jenkins02.o.o are all jenkins masters | 20:36 |
| jeblair | lifeless: the slaves take the form precise#.slave.openstack.org | 20:36 |
| *** dims has quit IRC | 20:37 | |
| *** rnirmal has quit IRC | 20:38 | |
| jeblair | lifeless: in our case, jenkins01 and jenkins02 do the bulk of the work; jenkins.o.o has more privileged (eg, can publish things) slaves attached to it (to avoid complications from potential jenkins privilege escalations from the test runners) | 20:38 |
| openstackgerrit | A change was merged to openstack-infra/config: Add a pointer to launch/README for root operators. https://review.openstack.org/44163 | 20:38 |
| jeblair | lifeless: from a scaling pov, one jenkins master can scale to at least hundreds of slaves. | 20:38 |
| jeblair | lifeless: i think the thing pleia2 and i are picking up on is the word 'minimally'; perhaps there should be a minimal set, and then recommendations for how to scale out the next steps | 20:40 |
| jeblair | lifeless: and i don't think dashboard should be in minimal; i mean, heck, we're not even using it right now. | 20:40 |
| jeblair | lifeless: and same for the additional jenkins masters -- probably a good idea if you're going to scale, but not minimally needed. | 20:40 |
| *** whoops has joined #openstack-infra | 20:42 | |
| lifeless | clarkb: fixed now. | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Ensure reports directory is setup correctly. https://review.openstack.org/44391 | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document hiera setup a bit more comprehensively. https://review.openstack.org/44357 | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make the adduser commands copy-pasteable. https://review.openstack.org/44270 | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Install launch/ dependencies on puppetmaster. https://review.openstack.org/44384 | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make setting up a Puppetmaster easier. https://review.openstack.org/44209 | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make config updates work with rebased branches. https://review.openstack.org/44383 | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Fix launch-node for non-official environments. https://review.openstack.org/44449 | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document bootstrapping a new infrastructure. https://review.openstack.org/44262 | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Create ci-launch directory on the puppetmaster. https://review.openstack.org/44394 | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make third-party imports clearer in launch/. https://review.openstack.org/44395 | 20:43 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document setting the correct hostname. https://review.openstack.org/44393 | 20:43 |
| lifeless | jeblair: so it's more complex to make the dashboard optional | 20:43 |
| lifeless | jeblair: right now the base config looks for a dashboard to report to. | 20:44 |
| lifeless | jeblair: I don't see the point in supporting arbitrary variation | 20:44 |
| jeblair | lifeless: ok, fair enough. i think we should make the dashboard optional, but if it is not in practice now, that's fine. | 20:44 |
| jeblair | lifeless: that leaves the multiple jenkinses | 20:45 |
| lifeless | jeblair: so the security separation is more important than scaling | 20:45 |
| lifeless | jeblair: isn't it? | 20:46 |
| lifeless | jeblair: ack on the word minimal; I'm expecting multiple futher iterations to this doc as we progress through the exercise | 20:46 |
| jeblair | lifeless: i think the separation is important; though it's afaik a very theoretical vector, and also depends on what you want to use the system for. if you only want to automate testing, but not, say, releases, then it's of less concern. | 20:49 |
| jeblair | that sentence was constructed poorly; it should have been two. but i think you get the idea. | 20:50 |
| *** thomasm has joined #openstack-infra | 20:56 | |
| lifeless | jeblair: so, I think keeping the two jenkins makes sense then: we can document why there are two; and that 01 is the first of a scalable set available if needed. | 20:56 |
| lifeless | jeblair: but collapsing to just one removes that protection - and one of the main things *I* want is release automation that isn't bound to my laptop or jml's laptop. | 20:56 |
| *** pcm__ has quit IRC | 20:57 | |
| lifeless | jeblair: I probably haven't spelt this out... I want a) to follow openstack-ci as closely as possible: every delta in design or deployment is ongoing maintenance. I want near-zero of that. | 20:57 |
| *** pcm_ has joined #openstack-infra | 20:57 | |
| lifeless | jeblair: b) we want a broad spectrum of python test gates, and a small number of more complex things that will drive cloud instances etc. | 20:58 |
| lifeless | jeblair: c) release management | 20:58 |
| lifeless | jeblair: d) but we probably don't need wiki etherpad lists etc, at least for now. | 20:58 |
| jeblair | lifeless: wfm (with additional docs about why) | 21:00 |
| lifeless | jeblair: having this all AAS but not as noddy as e.g. travis, would be ideal. | 21:00 |
| lifeless | jeblair: would you like me to write 'why' docs now, or in a later iteration ? | 21:00 |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Missing space at the end of the conditional expression https://review.openstack.org/44223 | 21:01 |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Missing default value https://review.openstack.org/44226 | 21:01 |
| jeblair | lifeless: if it's more convenient, we can go with later | 21:01 |
| lifeless | jeblair: I'd kind of like to flush the queue; the rebase is getting ridonkulous :) | 21:01 |
| *** dims has joined #openstack-infra | 21:02 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add puppet modules we maintain to gerrit bot https://review.openstack.org/44370 | 21:02 |
| jeblair | lifeless: k, +2d | 21:02 |
| *** rfolco has quit IRC | 21:02 | |
| *** xBsd has joined #openstack-infra | 21:02 | |
| lifeless | jeblair: what doc file would 'why' docs about the openstack-ci layout go in? top level README.md ? | 21:02 |
| clarkb | jeblair: https://review.openstack.org/#/c/44348/1/modules/openstack_project/templates/review.projects.yaml.erb line 192. Any concern with duplicatign a project without a clear why? (it appears that the existing repo does a couple things and they will prune it but I don't see confirmation of that) | 21:03 |
| jeblair | clarkb: i did not notice that, thanks. i'll change to a -1. | 21:04 |
| jeblair | clarkb: are you going to leave a comment, or shall i? | 21:04 |
| lifeless | jeblair: one thing that struck me as I work through this is that there is a mix of 'what to run', 'how to run it' and 'what projects are hosted' all in the same repo. | 21:05 |
| notmyname | FYI python-swiftclient 1.6.0 tag pushed up just now | 21:05 |
| lifeless | jeblair: 'what and how' make sense to clone, but for many infra-downstreams 'hosted projects' will be totally different. | 21:05 |
| clarkb | jeblair: I can do it | 21:05 |
| lifeless | jeblair: it would be nice to hive that off to a different tree I think | 21:05 |
| clarkb | lifeless: https://review.openstack.org/#/c/44209/6/install_puppet.sh what is with the comment there? | 21:06 |
| clarkb | lifeless: YES! | 21:06 |
| jeblair | lifeless: can you elaborate more about 'hosted projects'? | 21:06 |
| openstackgerrit | A change was merged to openstack-infra/config: Start documenting running your own CI infra. https://review.openstack.org/44177 | 21:06 |
| lifeless | clarkb: that comment notes that there is copied code. | 21:06 |
| lifeless | clarkb: there is both a manual and a puppet enforced setting. | 21:07 |
| lifeless | clarkb: if one changes, so should the other. | 21:07 |
| lifeless | clarkb: though this is apparently not obvious :) | 21:07 |
| clarkb | I think we need to refactor to use roles and profiles https://github.com/hunner/roles_and_profiles. Then we can use r10k to split the hows into distinct modules | 21:07 |
| clarkb | then only how will live in openstack-infra/config | 21:07 |
| clarkb | lifeless: 'NB:' is what confused me | 21:08 |
| lifeless | jeblair: to add a new project to gerrit for instance, one patches the same git tree that describes how to add a new jenkins master. | 21:08 |
| lifeless | clarkb: note bene | 21:08 |
| lifeless | jeblair: these two things are not the same, nor even tightly coupled. | 21:08 |
| clarkb | doing the refactoring that I describe is a significant amount of work but definitely doable | 21:08 |
| lifeless | clarkb: so I'm more ok with how and what in the same tree, vs list-of-hosted-projects | 21:09 |
| clarkb | lifeless: the nice thing about what I describe is it will be easier to swap in puppet modules from puppetforge | 21:10 |
| lifeless | clarkb: 90% of openstack_project/ seems to be applicable to me so far, I'm unlikely to refactor it as I go, but I'll start a low-priority thread to consolidate all the stuff i"m copying | 21:10 |
| *** sarob_ has quit IRC | 21:10 | |
| clarkb | lifeless: for example there is a good logstash module there which didn't exist when I started working with logstash so I wrote my own | 21:10 |
| lifeless | clarkb: what is r10k ? | 21:10 |
| clarkb | lifeless: it is a tool for managing puppet modules either from puppet forge or from source | 21:11 |
| clarkb | lifeless: it uses puppetfiles which come from puppet-librarian but I am told that r10k actually works :) | 21:12 |
| clarkb | mgagne: have you used r10k yet? finch says it will solve all my problems | 21:12 |
| mgagne | clarkb: I just read about you talking about r10k. no, haven't used it. | 21:12 |
| jeblair | lifeless: this shows signs of being a much longer conversation than i am able to have atm, but openstack-infra/config is "The configuration of the infrastructure of the OpenStack project"; i think that includes the hosts we run and the configuration of those hosts, including what projects are hosted on them. | 21:12 |
| clarkb | lifeless: using it allows you to avoid dealing with git submodules or git subtree and instead treat dependencies like packages | 21:12 |
| *** fifieldt has joined #openstack-infra | 21:12 | |
| *** jamiem has quit IRC | 21:13 | |
| jeblair | lifeless: i believe if we continue in the direction we want to, including what clarkb is talking about, then many operational details of, eg, how to run a gerrit server, should be pulled out of that repo. | 21:13 |
| clarkb | ++ | 21:13 |
| lifeless | jeblair: So I don't much care which direction the split occurs in :) | 21:13 |
| mgagne | clarkb: IMO, it looks like an organization philosophy | 21:14 |
| lifeless | jeblair: however, I think the split I'm suggesting is useful irrespective of downstreams, though of course both repos should be openstack-infra repos. | 21:14 |
| lifeless | jeblair: but - the main reason I mentioned it is that you wanted feedback on the design and layout and approach. | 21:15 |
| mgagne | clarkb: rename openstack_project to role, split reusable piece of manifests to profile and you now have a similar setup | 21:15 |
| clarkb | mgagne: yup | 21:15 |
| clarkb | hunner brought roles and profiles up after complaining that our gerrit module is too monolithic | 21:15 |
| lifeless | jeblair: and this plus the fact that I'm copy-pasting unmodified many pp files, some with a minor search-replace. | 21:15 |
| lifeless | jeblair: is the feedback I have so far ;) | 21:15 |
| clarkb | but after reading his exampes I realized openstack_project was half an attempt at doing roles and profiles | 21:16 |
| clarkb | mgagne: but there is a bit more to pull some of the monolithicness out and into profiles | 21:16 |
| mgagne | clarkb: could be that some stuff are openstack specific and/or does not allow more granular configuration of gerrit | 21:16 |
| *** UtahDave has quit IRC | 21:17 | |
| clarkb | mgagne: there is some of that too | 21:17 |
| clarkb | but I guess when you write pretty puppet all day for clients you want more of the same :) | 21:17 |
| mgagne | clarkb: we could say it's much like a 600 lines method, doing to much | 21:17 |
| clarkb | yup | 21:17 |
| clarkb | I absolutely agree with his suggsetions. With a bit of refactoring our puppet becomes quite flexible | 21:18 |
| *** UtahDave has joined #openstack-infra | 21:18 | |
| mgagne | clarkb: this should be split in an other manifest within the module itself: https://github.com/openstack-infra/config/blob/master/modules/gerrit/manifests/init.pp#L512-543 | 21:18 |
| mgagne | clarkb: easy example | 21:18 |
| clarkb | jeblair: lifeless' change at hte bottom of the stack needs one more review | 21:20 |
| mgagne | clarkb: have to go | 21:21 |
| clarkb | jeblair: I am moving up the stack and approving things that look good. once the one at the bottom is approved we should see a bunch of things go into the gate | 21:21 |
| clarkb | mgagne: have a good weekend | 21:21 |
| mgagne | clarkb: thanks, you too | 21:21 |
| clarkb | mgagne: I don't think there is a major rush on the refactoring but it would be good to think of this stuff as we make changes | 21:21 |
| mgagne | clarkb: cool, don't mind giving my input ;) | 21:21 |
| *** weshay has quit IRC | 21:22 | |
| *** svarnau has quit IRC | 21:22 | |
| *** openstackgerrit has quit IRC | 21:25 | |
| *** openstackgerrit has joined #openstack-infra | 21:25 | |
| clarkb | lifeless: https://review.openstack.org/#/c/44262/11/doc/source/running-your-own.rst what is meant by Migrate: on line 113? | 21:26 |
| clarkb | lifeless: migrate from openstack-infra's configs? | 21:26 |
| *** woodspa has quit IRC | 21:26 | |
| lifeless | clarkb: yes, copy and paste | 21:29 |
| clarkb | lifeless: and for https://review.openstack.org/#/c/44383/8/modules/openstack_project/manifests/puppetmaster.pp are you running that on your master and it works fine? | 21:30 |
| clarkb | @{u} is going to send me to cron manpages I think | 21:30 |
| clarkb | or is that a git thing? | 21:31 |
| *** melwitt has joined #openstack-infra | 21:31 | |
| openstackgerrit | A change was merged to openstack-infra/config: Make setting up a Puppetmaster easier. https://review.openstack.org/44209 | 21:31 |
| openstackgerrit | A change was merged to openstack-infra/config: Document hiera setup a bit more comprehensively. https://review.openstack.org/44357 | 21:31 |
| openstackgerrit | A change was merged to openstack-infra/config: Fix launch-node for non-official environments. https://review.openstack.org/44449 | 21:32 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/nodepool: Add a nodepool command https://review.openstack.org/44553 | 21:32 |
| openstackgerrit | Russell Bryant proposed a change to openstack-infra/config: Report cinder patches to #openstack-cinder https://review.openstack.org/44554 | 21:33 |
| openstackgerrit | A change was merged to openstack-infra/config: Document bootstrapping a new infrastructure. https://review.openstack.org/44262 | 21:33 |
| lifeless | clarkb: I am | 21:33 |
| lifeless | clarkb: it does | 21:33 |
| lifeless | clarkb: it means 'upstream branch' | 21:34 |
| *** mriedem has quit IRC | 21:35 | |
| clarkb | equivalent to origin/master in this case? | 21:35 |
| *** whoops has quit IRC | 21:36 | |
| clarkb | I found it in gitrevisions(7) | 21:37 |
| *** [wchrisj_ has quit IRC | 21:38 | |
| lifeless | clarkb: for you yes | 21:38 |
| lifeless | clarkb: for https://github.com/testing-cabal/ci-config, equivalent to origin/testcabal | 21:39 |
| lifeless | clarkb: it's the branch that 'git pull' would pull from, but I couldn't find a pull recipe that == reset --hard. | 21:39 |
| *** thomasm has quit IRC | 21:40 | |
| clarkb | lifeless: should the git fetch be a git remote update instead to get all remotes? | 21:40 |
| *** dina_belova has joined #openstack-infra | 21:41 | |
| clarkb | I guess in this case it probably doesn't matter. everything is talking to origin | 21:41 |
| lifeless | clarkb: it will fetch the current pull tracking branch remote, IIRC. | 21:42 |
| lifeless | clarkb: but yeah, it should work for you as-is, and works for me. | 21:42 |
| lifeless | clarkb: other folk can submit patches. | 21:42 |
| *** sarob has joined #openstack-infra | 21:43 | |
| notmyname | suppose the swift team wanted to create a 3rd deliverable as a separate code repo (in addition to the server and the client). what is the feasibility, risks, etc? | 21:44 |
| *** dina_belova has quit IRC | 21:45 | |
| *** thedodd has quit IRC | 21:46 | |
| clarkb | notmyname: that seems to becoming more common with other projects, it is definitely feasible | 21:46 |
| clarkb | lifeless: I got picky in https://review.openstack.org/#/c/44384/ sorry | 21:46 |
| *** sarob_ has joined #openstack-infra | 21:46 | |
| notmyname | clarkb: k. nothing formal yet. just a discussion this afternoon that I wanted to explore | 21:47 |
| *** pcm_ has quit IRC | 21:47 | |
| clarkb | lifeless: and will have a comment for its child that can be addressed together | 21:47 |
| clarkb | notmyname: heat is doing this for addtional stuff and so is trove | 21:47 |
| notmyname | clarkb: ah. we're thinking about it for the swift-bench tool | 21:47 |
| clarkb | notmyname: I think one of the risks is packagers see it as another dependency but often that is the right choice | 21:47 |
| notmyname | clarkb: the good news in this case is that it wouldn't be a dependency on either the server or the client (it would depend on the client though) | 21:48 |
| *** prad_ has quit IRC | 21:48 | |
| lifeless | clarkb: is it possible to list multiple packages, or do they have to be separate stanzas? | 21:48 |
| *** sarob_ has quit IRC | 21:49 | |
| *** sarob has quit IRC | 21:49 | |
| clarkb | lifeless: you can list multiple as long as they share all of the same parameters. So python-novaclient and python-cinderclient can be combined. I will paste and example of that the looks like | 21:49 |
| jeblair | clarkb: did cookiecutter land? | 21:49 |
| *** sarob has joined #openstack-infra | 21:49 | |
| clarkb | jeblair: no https://review.openstack.org/#/c/42530/ | 21:50 |
| jeblair | notmyname: swift-bench currently exists within swift repo? | 21:50 |
| notmyname | jeblair: yup | 21:50 |
| clarkb | lifeless: http://paste.openstack.org/show/45500/ | 21:50 |
| *** bpokorny has quit IRC | 21:50 | |
| jeblair | notmyname: we have 'git filter-branch'-ed projects out of existing repos before; we can help with that if/when the time comes, if you like. | 21:51 |
| lifeless | clarkb: do you want VWS between the stanzas, though the comment applies to both ? | 21:51 |
| mgagne | clarkb: I'm still here. => yes | 21:51 |
| *** morganfainberg is now known as morganfainberg|a | 21:51 | |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/devstack-gate: Add support for large_ops tempest test https://review.openstack.org/43779 | 21:51 |
| notmyname | jeblair: nah, if we were to split it, I'd want to keep the history intact | 21:51 |
| flashgordon | clarkb: I think large_ops test is ready | 21:51 |
| mgagne | clarkb: check jenkins::slave | 21:52 |
| jeblair | notmyname: that's what git filter-branch does. :) | 21:52 |
| mgagne | clarkb: doing it already | 21:52 |
| clarkb | mgagne: ? | 21:52 |
| *** morganfainberg|a is now known as morganfainberg | 21:52 | |
| lifeless | notmyname: it gives you the history of just the files you keep | 21:52 |
| mgagne | clarkb: forget it, you were replying to lifeless =) | 21:52 |
| notmyname | jeblair: as in, I don't want to prune it from the history. I'd want to have an initial commit of "initial migration from the swift repo" as the first commit of the new thing | 21:52 |
| notmyname | jeblair: lifeless: but I'm sure I could be convinced otherwise | 21:52 |
| lifeless | notmyname: tis mainly asthetics | 21:53 |
| clarkb | lifeless: VWS == whitespace? | 21:53 |
| lifeless | bah, aesthetics. | 21:53 |
| lifeless | clarkb: vertical whitespace | 21:53 |
| lifeless | clarkb: aka a blank line | 21:53 |
| clarkb | lifeless: no you can group them together since they are all related and belong to the comment | 21:53 |
| notmyname | jeblair: lifeless: ya, I understand. perhaps it would be cool. it's not what I normally think of when I hear `git filter-branch` :-) | 21:53 |
| jeblair | notmyname: your choice -- i like the filter-branch approach because you have the history of the relevant files without the extra size of the rest of the repo; git blame, etc still work. | 21:54 |
| notmyname | jeblair: ya, that makes sense | 21:54 |
| clarkb | notmyname: and you can remove it from the swift repo in the way you describe | 21:54 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Ensure reports directory is setup correctly. https://review.openstack.org/44391 | 21:54 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make the adduser commands copy-pasteable. https://review.openstack.org/44270 | 21:54 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Install launch/ dependencies on puppetmaster. https://review.openstack.org/44384 | 21:54 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make config updates work with rebased branches. https://review.openstack.org/44383 | 21:54 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Create ci-launch directory on the puppetmaster. https://review.openstack.org/44394 | 21:54 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Make third-party imports clearer in launch/. https://review.openstack.org/44395 | 21:54 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document setting the correct hostname. https://review.openstack.org/44393 | 21:54 |
| clarkb | notmyname: wtih a single commit that deletes files and explains why | 21:54 |
| lifeless | yay, no longer a full page. | 21:54 |
| *** sarob has quit IRC | 21:54 | |
| notmyname | clarkb: jeblair: who do we talk to if/when we decide to do it? | 21:54 |
| lifeless | clarkb: so how do you tell puppet to run the dashboard services? | 21:55 |
| clarkb | lifeless: there should be an init script for them so you can create a service resource for the service and ensure => running | 21:56 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/nodepool: Add a nodepool command https://review.openstack.org/44553 | 21:56 |
| clarkb | lifeless: if you grep for 'service {' in the repo you shoudl find examples | 21:56 |
| clarkb | lifeless: did that recent stack reorder dependencies? | 21:56 |
| jeblair | notmyname: let clarkb, fungi, mordred or i know, and we can help with setting it up | 21:56 |
| lifeless | clarkb: no | 21:57 |
| jeblair | notmyname: also i'm sure ttx will be interested | 21:57 |
| lifeless | clarkb: I just rebase -i'd to edit the patch, then review -y to upload. | 21:57 |
| notmyname | jeblair: ttx is interested in /everything/ | 21:57 |
| clarkb | did you rebase -i against master/ | 21:57 |
| jeblair | notmyname: i thought that was a safe bet. :) | 21:57 |
| clarkb | I am trying to figure out why it repusehd a couple things I had approved but were through the gate yet. Not a big deal | 21:57 |
| lifeless | clarkb: other changes would have landed in the interim, so the review scripts fetch + rebase combo would change the hash on all the not-in-master patches. | 21:57 |
| lifeless | clarkb: 'review' aka 'all your rebase done for you' | 21:58 |
| clarkb | git review only does that if there are merge conflicts, which may be the case | 21:59 |
| lifeless | oh, has it gotten smarter? | 21:59 |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Remove node management from devstack-gate https://review.openstack.org/44503 | 21:59 |
| clarkb | yeah anything >= 1.19 or so | 21:59 |
| *** tstevenson has quit IRC | 21:59 | |
| lifeless | Press ENTER or type command to continue | 21:59 |
| lifeless | git-review version 1.21 | 21:59 |
| openstackgerrit | A change was merged to openstack-infra/config: Make the adduser commands copy-pasteable. https://review.openstack.org/44270 | 22:00 |
| lifeless | does it log it's actions? | 22:00 |
| openstackgerrit | A change was merged to openstack-infra/config: Make config updates work with rebased branches. https://review.openstack.org/44383 | 22:00 |
| clarkb | lifeless: no, but you can pass it the -v flag and it will be very verbose | 22:00 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Update nodepool init script to nodepoold https://review.openstack.org/44557 | 22:00 |
| lifeless | clarkb: logging would be good | 22:01 |
| lifeless | clarkb: then one can answer 'why did X happen' | 22:01 |
| *** xBsd has quit IRC | 22:03 | |
| clarkb | jeblair: I think if you give the existing stack one more pass we may be able to merge them all | 22:03 |
| *** burt has quit IRC | 22:04 | |
| *** sarob has joined #openstack-infra | 22:07 | |
| *** sarob has quit IRC | 22:08 | |
| *** jamiem has joined #openstack-infra | 22:08 | |
| *** sarob has joined #openstack-infra | 22:08 | |
| *** sarob_ has joined #openstack-infra | 22:09 | |
| jeblair | lifeless: did you note my comment about bug refs in commit msgs? | 22:10 |
| openstackgerrit | A change was merged to openstack-infra/config: Install launch/ dependencies on puppetmaster. https://review.openstack.org/44384 | 22:11 |
| openstackgerrit | A change was merged to openstack-infra/config: Ensure reports directory is setup correctly. https://review.openstack.org/44391 | 22:12 |
| openstackgerrit | A change was merged to openstack-infra/config: Document setting the correct hostname. https://review.openstack.org/44393 | 22:12 |
| *** sarob has quit IRC | 22:13 | |
| *** morganfainberg is now known as morganfainberg|a | 22:13 | |
| openstackgerrit | A change was merged to openstack-infra/config: Create ci-launch directory on the puppetmaster. https://review.openstack.org/44394 | 22:13 |
| openstackgerrit | A change was merged to openstack-infra/config: Make third-party imports clearer in launch/. https://review.openstack.org/44395 | 22:14 |
| *** woodspa has joined #openstack-infra | 22:16 | |
| *** woodspa has left #openstack-infra | 22:18 | |
| *** zeus has quit IRC | 22:19 | |
| *** woodspa has joined #openstack-infra | 22:19 | |
| *** mgagne has quit IRC | 22:23 | |
| lifeless | jeblair: I didn't sorry; which patch is it on? | 22:25 |
| jeblair | lifeless: not important; i wanted to point out that there's a canonical way of referring to bugs now: https://wiki.openstack.org/wiki/GitCommitMessages#Including_external_references | 22:27 |
| jeblair | lifeless: (if you want automation to act on it) | 22:27 |
| jeblair | lifeless: lots of backwards compat forms are still supported though. | 22:27 |
| jeblair | lifeless: but it might be good to start adjusting habits | 22:28 |
| *** pcrews has quit IRC | 22:28 | |
| *** jamiem has quit IRC | 22:29 | |
| lifeless | jeblair: nuts, I thought I was following the canonical form :) | 22:30 |
| lifeless | jeblair: thanks | 22:30 |
| *** jhesketh_ has quit IRC | 22:30 | |
| *** sarob_ has quit IRC | 22:34 | |
| *** morganfainberg|a is now known as morganfainberg | 22:43 | |
| jeblair | we have 1129 jobs defined in jenkins | 22:45 |
| jeblair | it was like 450 the last time i looked. | 22:45 |
| clarkb | jeblair: Alex_Gaynor and jd__ have been busy with the py3k and pypy jobs :) | 22:45 |
| * Alex_Gaynor added like 5... jobs... | 22:46 | |
| jeblair | Alex_Gaynor: to maybe 100 projects :) | 22:47 |
| jeblair | also, there... are... four... jobs. | 22:47 |
| Alex_Gaynor | Ah, right, so I'm responsible for *maybe* 100 jobs :) Someone else is responsible for the other 600 | 22:48 |
| jeblair | i think it's pretty cool. :) | 22:51 |
| Alex_Gaynor | :) | 22:57 |
| *** woodspa has quit IRC | 23:04 | |
| clarkb | jeblair: are you going to this OpenStack Core talk thing? | 23:05 |
| clarkb | or did that already happen? | 23:05 |
| clarkb | nevermind it happened | 23:06 |
| notmyname | clarkb: and it was exciting! | 23:06 |
| clarkb | does anyone else think "Core" is the wrong term because it is too overloaded? | 23:07 |
| jeblair | notmyname: were all our problems solved? | 23:07 |
| clarkb | openstack-core-positions == people that review code to me | 23:07 |
| notmyname | jeblair: well, first off, there was pizza | 23:07 |
| jeblair | notmyname: then it was a success! | 23:07 |
| notmyname | jeblair: so according to maslow, we're done | 23:07 |
| *** reed has quit IRC | 23:09 | |
| notmyname | clarkb: jeblair: it was less silly than the AWS API stuff. overall, I support the general idea, but I think there are still too many undefined or overloaded words in the current conversation | 23:09 |
| notmyname | clarkb: but the problem is that the openstack bylaws say "core" | 23:09 |
| clarkb | notmyname: yeah I think the ship has sailed, I was just confused for a minute | 23:10 |
| jeblair | the bylaws can be changed. it's only _nearly_ impossible to do so. | 23:10 |
| clarkb | Item 6.B bothers me | 23:12 |
| clarkb | why is open sourcing the alternate implementation matter if it isn't compliant? you remove the mark and move one | 23:13 |
| clarkb | s/one/on/ anyways, I clearly missed the discssion | 23:13 |
| *** Protux has quit IRC | 23:17 | |
| jeblair | clarkb: what are you reading? | 23:18 |
| clarkb | jeblair: http://robhirschfeld.com/2013/08/27/taking-openstack-core-discussions-to-community/ | 23:20 |
| *** pcrews has joined #openstack-infra | 23:21 | |
| clarkb | jeblair: I followed that to http://robhirschfeld.com/2013/08/13/openstack-core-positions/ | 23:22 |
| pleia2 | the pizza wasn't bad | 23:22 |
| pleia2 | reed was there too | 23:22 |
| clarkb | it is also slightly weird that they would do the Core talk as a non design session considering it is directly related to the TC | 23:23 |
| notmyname | clarkb: they are planning on having the final decision made (or thingy approved) on the board meeting right before the summit. on sunday in hong kong | 23:24 |
| notmyname | ie before design sessions | 23:24 |
| clarkb | oh its before | 23:24 |
| notmyname | clarkb: my concern with the wording is that it seems geared towards the provisioning API framework projects (like cinder and neutron) and not towards stuff like swift, horizon, marconi, or ceilometer | 23:25 |
| koolhead17 | clarkb: we need to sit drink beer on 7th when am in Seattle & talk about Core :D | 23:27 |
| notmyname | also the language around basing "must pass" tests (ie those that define compliance) based on common deployments--IOW the results of a survey | 23:27 |
| koolhead17 | pleia2: hi there. | 23:27 |
| pleia2 | koolhead17: hey | 23:27 |
| koolhead17 | pleia2: howdy? | 23:28 |
| koolhead17 | notmyname: hellos | 23:28 |
| anteaya | how long are you in seattle, koolhead17? | 23:28 |
| koolhead17 | anteaya: only for one day :( | 23:28 |
| koolhead17 | 7th | 23:28 |
| anteaya | :( | 23:28 |
| anteaya | going to be in seattle on the 14th | 23:28 |
| anteaya | I'll miss seeing you | 23:28 |
| koolhead17 | sigh.. | 23:28 |
| clarkb | notmyname: the stuff about vendor specific implementations bothers | 23:28 |
| clarkb | *bothers me | 23:28 |
| anteaya | koolhead17: you will be at the summit, yeah? | 23:29 |
| koolhead17 | anteaya: hopefully yes. :D | 23:29 |
| clarkb | notmyname: the bits about must open source a thing if you don't pass tests seems unenforceable, and not necessary | 23:29 |
| anteaya | koolhead17: keep me posted, I hope to see you there | 23:29 |
| notmyname | clarkb: indeed | 23:29 |
| koolhead17 | anteaya: for sure | 23:29 |
| anteaya | :D | 23:29 |
| clarkb | notmyname: I would also question being an openstack core thing if you have a completely private implementation that simply passes the must pass tests. That isn't OpenStack that is a proprietary clone | 23:29 |
| koolhead17 | anteaya: i will be at doc bootcamp. hoping to meet annegentle_ fifieldt and many others :D | 23:30 |
| clarkb | koolhead17: yes beer | 23:30 |
| koolhead17 | clarkb: for sure sir. | 23:30 |
| pleia2 | clarkb: yeah, that came up, it can't just be an API test | 23:30 |
| koolhead17 | DM me your number so can SMS you once am there as i might not have IRC access via cellphone | 23:31 |
| anteaya | koolhead17: that sounds great, the infra boot camp was awesome, so I hope the doc bootcamp is equally awesome | 23:31 |
| clarkb | pleia2: but any amount of testing doesn't change that it is a clone that we have zero control of | 23:31 |
| koolhead17 | anteaya: yups am sure of it | 23:31 |
| pleia2 | could have a really lousy proprietary clone, and that gives people a lousy experience, not good for openstack | 23:31 |
| notmyname | clarkb: indeed. but I think that's covered by section 4 (although that has its own fuzzy definition issues) | 23:31 |
| pleia2 | clarkb: presumably the control comes from not allowing them to call it openstack (trademark) if it's not | 23:32 |
| clarkb | pleia2: but vendors may substitute. that is what 6 is all about :) | 23:32 |
| clarkb | pleia2: and passing the must pass tests is the only hurdle to getting the mark | 23:33 |
| pleia2 | clarkb: but it's still bound by 4 | 23:33 |
| pleia2 | pretty much the entire discussion the other night was on 4 and 6 btw :) | 23:33 |
| clarkb | even if you are bound by 4 you can be closed and the foundation/project have no control over the actual projects | 23:34 |
| pleia2 | yeah | 23:34 |
| clarkb | (assuming the frameworks in 4 don't require being free as in libre) | 23:34 |
| pleia2 | notmyname made good points about 4 being vague enough that it's hard to tell exactly what it is/requires | 23:35 |
| notmyname | actually, I had issues with 6 too (I think). the whole lack of definition around "plugin" or "alternate implementation" is too vague right now | 23:36 |
| *** zul has joined #openstack-infra | 23:36 | |
| pleia2 | yeah | 23:36 |
| notmyname | eg other vendors are starting to announce "Swift API compatibility" now. which is cool. but also emphasizes the need for definition. (eg ceph, emc vipr, cleversafe, etc) | 23:37 |
| *** lbragstad has joined #openstack-infra | 23:40 | |
| *** Shrews has joined #openstack-infra | 23:41 | |
| *** yjiang5 is now known as bmqq123 | 23:42 | |
| *** gyee has quit IRC | 23:43 | |
| koolhead17 | zul: hola dir | 23:45 |
| koolhead17 | clarkb: Nova is not Nova if you are not using RabbitMQ as per CORE definiton | 23:46 |
| koolhead17 | :D | 23:46 |
| * koolhead17 is kidding | 23:46 | |
| koolhead17 | pleia2: will you be occupied over weekend? We can go grab beer somewhere | 23:46 |
| pleia2 | koolhead17: yeah, I have a friend coming into town tomorrow, sorry :\ | 23:47 |
| koolhead17 | pleia2: np. :) | 23:47 |
| pleia2 | was good to see you last night though! | 23:47 |
| clarkb | koolhead17: thats fine with me :) (replace rabbitmq with qpid, 0mq, etc and it is still fine with me) | 23:47 |
| clarkb | koolhead17: because there is still a thing that is controlled by the body giving you the mark, that is open and accessible to all | 23:47 |
| clarkb | if there is a decision to standardize on some components that does not break ^ | 23:48 |
| koolhead17 | i heard someone bragging over twitter CORE certificate will be given/provided | 23:48 |
| koolhead17 | :D | 23:48 |
| koolhead17 | i wonder what that certification means | 23:48 |
| koolhead17 | haha | 23:49 |
| koolhead17 | and current CORE can become un-CORE | 23:49 |
| koolhead17 | as well | 23:50 |
| bmqq123 | not sure if gerrit can provide a method to differnt two patchset, like patchset 24/25 difference. | 23:55 |
| clarkb | bmqq123: it can, in the cahnge there is a dropdown that says "base" change that to patchset 24 | 23:55 |
| clarkb | bmqq123: you can also use git review -m to do the diffs locally. The benefit of using git review -m is it will rebase both sides to a common base so you only see differences between the patchsets and not their bases | 23:56 |
| bmqq123 | clarkb: that base will include everything changed (i.e. master changes between 24/25) and it's very difficlut to distinguish the pure-patch changes. | 23:56 |
| clarkb | bmqq123: that is what git review -m aims to solve | 23:56 |
| bmqq123 | clarkb: aha, that's sound really perfect, will have a try. | 23:56 |
| openstackgerrit | A change was merged to openstack/requirements: Add jsonpatch https://review.openstack.org/44253 | 23:59 |
| *** markmcclain has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!