Wednesday, 2015-01-21

*** hashar has joined #openstack-sprint00:41
*** hashar has quit IRC01:55
*** hashar has joined #openstack-sprint05:14
*** hashar has quit IRC06:03
*** rfolco has joined #openstack-sprint10:54
*** asselin_ has joined #openstack-sprint15:08
*** hashar has joined #openstack-sprint15:12
*** omrim has joined #openstack-sprint15:14
*** mmedvede has joined #openstack-sprint15:16
*** ja has joined #openstack-sprint15:32
*** hashar has quit IRC15:39
*** hashar has joined #openstack-sprint15:40
krtaylor#startmeeting third-party-ci-documentation16:02
openstackMeeting started Wed Jan 21 16:02:16 2015 UTC and is due to finish in 60 minutes.  The chair is krtaylor. Information about MeetBot at http://wiki.debian.org/MeetBot.16:02
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.16:02
openstackThe meeting name has been set to 'third_party_ci_documentation'16:02
krtaylorhello everyone16:03
krtaylorthis is the start of the third-party-ci-documentation virtual sprint16:03
*** luqas has joined #openstack-sprint16:03
krtaylorthe goal is to rewrite/refresh the third-party ci documentation16:04
krtaylorhere is the link to the etherpad for the sprint:16:04
krtaylor#link https://etherpad.openstack.org/p/third-party-ci-documentation16:04
krtayloranyone here for the sprint?16:05
mmedvedeI am lurking. Looking through docs16:07
krtaylorok, so the first thing is, if you have not already, read third-party.rst16:09
krtaylor#link http://ci.openstack.org/third_party.html16:09
krtaylorand16:10
krtaylor#link http://ci.openstack.org/running-your-own.html16:10
krtayloras you are reading through it, make note of the areas that need refreshing16:10
krtaylorthere are many16:10
krtayloradd notes to that section in the etherpad, at a minimum16:11
krtayloreven better, put your nick on the etherpad next to a section and re-write it16:11
krtaylorI put myself next to the Overview and Requirements sections, but that does not mean that I won't take help there16:12
krtaylorbut it would be better to have all the sections covered first16:13
krtaylorbut, first, some housekeeping16:13
krtaylorjust to remind everyone, the sprint will run for 48 hours16:14
krtaylorso we will be in this channel round-the-clock for the next 2 days16:14
krtaylorI will try to be here as much as I can over that timeframe16:14
krtaylorok, I guess I didnt need to start a meeting, the channel is logged16:18
krtaylor#endmeeting16:18
openstackMeeting ended Wed Jan 21 16:18:44 2015 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)16:18
openstackMinutes:        http://eavesdrop.openstack.org/meetings/third_party_ci_documentation/2015/third_party_ci_documentation.2015-01-21-16.02.html16:18
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/third_party_ci_documentation/2015/third_party_ci_documentation.2015-01-21-16.02.txt16:18
openstackLog:            http://eavesdrop.openstack.org/meetings/third_party_ci_documentation/2015/third_party_ci_documentation.2015-01-21-16.02.log.html16:18
krtaylorok, omrim you said you could help on the other side of the world, keep things moving ?16:20
krtayloromrim, thanks, I know it is late for you16:20
krtaylorwe don't have anyone signed up for running-your-own.html16:21
krtaylorthat is probably most out of date16:21
krtaylorthe requirements section is rather lacking, it would be good to have some expectation on what will be needed there16:22
*** luqas has quit IRC16:26
krtaylorremember to set your patch topics to: third-party-ci-documentation16:26
krtaylordoes anyone have any questions? on the etherpad? on our goal?16:28
krtaylorhttps://review.openstack.org/#/c/146968/   is pretty well reviewed, just needs another core to merge16:34
krtaylormmedvede, is there an area that you want to focus on?16:35
*** jhesketh has quit IRC16:37
mmedvedekrtaylor: not  anything in particular. I can just get the next unassigned section (Event Stream)16:38
*** jhesketh has joined #openstack-sprint16:39
krtaylormmedvede, great16:40
*** hashar has quit IRC16:40
krtaylorI did get permission from jaypipes to use anything from his blogs as input as needed16:40
krtaylorbut we should give him co-author credit for sure16:41
krtaylorthe links for those blogs are at the bottom of the etherpad16:41
krtaylorrfolco 's blog is there as well, with another perspective16:41
krtaylorrfolco, I assume we can use wording from that blog as needed, with co-author credit?16:42
krtaylorhm, wish I could set the channel topic...17:03
*** mestery has joined #openstack-sprint17:03
krtaylorI am reviewing the infra manual process for advice on  the proper way to link to other docs in various places17:09
krtayloranyone know off hand?17:09
rfolcokrtaylor, yes you can grab info from the article17:11
rfolcokrtaylor,     Posting Result To Gerrit (rfolco)17:11
rfolcokrtaylor, I need some education on the doc process but I assigned one workt item to myself17:11
krtaylorrfolco, great!17:12
krtaylorI am looking for any style references we may need17:12
krtaylorbut I feel like we should just get  the words in a patch, and folks that know can add comments17:13
krtaylorin other words, not worry about formatting as much as content17:14
pleia2good morning17:15
krtaylorbut for links `Words to describe link <http link>`_. is the format17:15
krtaylorpleia2, hi! welcome!17:15
pleia2krtaylor: have a moment for a quick PM?17:16
krtaylorpleia2, sure17:16
krtaylorpleia2, thanks for the tweet!17:26
pleia2sure thing :)17:26
pleia2should do another now actually17:26
krtaylorwe need everyone to spread the word, this was so last minute17:27
krtaylorI'd really like to make sure these doc changes link into existing docs as much as possible17:28
*** mestery has quit IRC17:28
krtaylorwe need to move forward with this thinking about how we can make it easier to maintain17:28
*** mestery has joined #openstack-sprint17:28
pleia2I'll also mention that there is some work happening to create a puppet module for all of this, but I think these docs will be valuable in creating said puppet module, and even once we have that it won't always be perfect for everyone17:29
pleia2so I'm sure these docs will actively evolve once that comes into play17:29
pleia2but that's how things go :)17:30
krtaylorpleia2, yes, we discussed that in the third-party meeting17:30
krtaylorasselin_, agreed that after in-tree, we'd update docs17:31
krtaylorhere is a link to that spec, btw: https://review.openstack.org/#/c/139745/17:31
pleia2thanks17:31
krtaylorit is shaping up very nicely17:32
krtaylorso, I was looking at the 2 documents we have for third-party testing17:33
krtaylorthe first third_party.rst is basically what a system needs to do to interact with infra services17:33
krtaylorthe second, running-your-own, is how to grab, modify and run17:34
krtaylorwe have FAQ in the first, but the question there really belongs in the second17:35
krtaylorso, 2 FAQs?17:35
krtayloror, and maybe this is too big, why do we have 2 documents anyway? makes it more manageable? or is it just because thats how it was done initially?17:36
krtayloranyone know?17:36
pleia2I think I'd forget which faq to reference if there were 217:40
pleia2the first document was created by someone who was doing a presentation on running your own and didn't know until after he committed to it that it wasn't well documented :)17:41
pleia2so in the course of talk prep, he documented17:41
krtaylorcool, good to know, I know we had to feel our way through the dark  :)17:42
krtayloryeah, 2 FAQs don't make sense17:42
krtaylorwhat if we pulled the docs together?17:42
pleia2so running-your-own is kind of CI-focused document, since our project is "an open source infrastructure" and we want folks to be able to use it when they bring up their system in their organization that has nothing to do with openstack17:45
pleia2my vision here would be that running-your-own would finally be maintained, and third-party would reference it as needed, adding in bits of information that are 3rd party specific17:46
krtaylorhm, good point, it would make it more broadly consumable to keep is separate17:47
krtaylorand we can always merge later if someone feels so inclined17:48
pleia2yeah17:49
krtaylorneed to grab some lunch, brb17:52
*** mestery has quit IRC18:37
krtaylormmedvede, any work needed to event stream section?18:54
mmedvedekrtaylor: not sure. Looks good superficially, was going to test the commands that are there19:00
asselin_hi, sorry very very busy morning, and my ci system went down....19:13
asselin_(still down)19:13
pleia2asselin_: ouch :\19:15
asselin_so I'd like to start migrating japipes blogs over. any comments and where that should go?19:17
pleia2I think the main work there is merging his posts with an update to http://ci.openstack.org/running-your-own.html19:21
pleia2since there is a lot of overlap, and we do want to keep running-your-own.html as a more general CI configuration document (unrelated to openstack, to anyone who wants to run the CI)19:22
krtaylorasselin_, no worries, I have a couple of meetings atm19:27
krtaylorbut I agree with pleia219:27
krtaylorasselin_, I think running-your-own needs a lot of attention, and no one is looking at it (yet)19:28
* asselin_ looking at running your own19:28
pleia2added the third-party-ci-documentation patch topic note to the etherpad, and link we can use for reviewing patches, once they start coming in https://review.openstack.org/#/q/topic:third-party-ci-documentation,n,z19:36
asselin_I think http://www.joinfu.com/2014/01/understanding-the-openstack-ci-system/ should be a pre-read to Running your own19:40
pleia2that makes sense, it does duplicate some of the other docs we have too so maybe not a drop in replacement, but I'm happy to help identify spots that we don't need to duplicate19:48
asselin_pleia2, honestly, the information seems more part of Third Party Testing. A bit more background for those who don't know what it is or why have it19:55
asselin_perhaps a "background" section?19:56
asselin_or not. I'm down to the section of "Continuous Integration Testing" Everything before should be in other docs20:01
pleia2asselin_: so there are things like talking about zuul's layout.conf where I think we can write a quick note about it, and then link to the CI docs for learning more, right now the blog post is out of date because /config repo no longer exists and the layout.yaml is in a completely different place, not having to maintain this on the third party side too would be nice20:01
pleia2not that I think it will change soon, but layout examples also change and cause confusion20:03
asselin_noted20:03
asselin_I like the idea of basic snips + link for more info20:04
pleia2great20:04
*** rfolco has quit IRC20:06
asselin_pleia2, krtaylor https://review.openstack.org/149032 Take a look at the structure. I'll update the content after grabbing some lunch.20:20
pleia2asselin_: want us to comment now, or hold off until you have some changes in?20:20
jait sounds like he's after comments on structure (now) while he's at lunch.  presumably he'll respond/iterate when done eating.20:27
pleia2yeah, that's what I'll go with :)20:27
* krtaylor reading20:30
krtaylorbtw, all of jaypipes blogs are linked at the bottom of the etherpad, as well as any other infromation sources that we use20:30
krtaylorand information sources20:30
jakrtaylor, meta-comments starting to spew out of my brain.  here?  etherpad?20:34
krtaylorja, etherpad please, todo section or comments section, or however you want20:36
krtaylorwow comments from jaypipes 7 minutes from pushing the patchset! very nice asselin_20:42
krtaylor:)20:42
pleia2I'm really happy to see he's involved in the process20:44
krtaylorvery20:45
*** mestery has joined #openstack-sprint20:47
pleia2did anyone invite him over here? :)20:48
krtaylor2x via email, and discussions about using his content, but not today  :)20:58
pleia2just sent him a msg in case he forgot :)21:01
*** jaypipes has joined #openstack-sprint21:01
pleia2jaypipes: welcome to the party!21:02
krtaylorhi jaypipes, thanks for joining!21:02
jaypipesheya :)21:02
krtaylorthanks for the crazy fast comments on asselin_  patchset21:03
jaypipesno worries. will try to do reviews on anything you ping me about.21:04
jaypipesin between meetings and more meetings ;)21:04
krtaylorjaypipes, perfect, thanks21:04
asselin_I'm back21:04
krtaylorja, I am interested in your comment on a structural picture21:05
krtaylorja, like - here are the components and how they hook together?21:05
asselin_wow, got more than I expected :)21:05
jayes, like rfolco's21:05
asselin_I just go have lunch more often :)21:06
krtaylorja, I have one I have done for presentations that is more general21:06
ja...his blog entry has a good starting point, although I would be careful to add in the firewall since that really affects planning21:06
krtaylorrfolco's is pretty specific to our environment, if it is the one I am thinking of21:06
krtayloryeah, that is what I was thinking for the run-your-own requirements section21:07
krtaylorthere isnt much discussion on how much hardware youll need, firewall considerations, options for hosting logs, etc21:07
jakrtaylor, I could posit several actually.  1=the actual openstack one, as infra "ships" it.  2=a generic third party.  3+= specific third party examples (probably as hyperlinks)21:07
krtayloryes, I think picture would really help some basic understanding of the layout of things and how they work together, it has really helped internally anyway21:09
ja... 1 is the base a 3P is "copying".  2 is the logical model (some of the specifics removed, so it's down to roles & responsibilities).  need at least one 3 somewhere so people have 2 diff examples (1 and 3) of applying 2 to a specific case.21:09
jakrtaylor, the pic on 2 would also be careful to point out places where the implementation would vary (i.e. where that's the expected state).  there might also be specific cases noted where "most 3Ps do X, but you can do otherwise" eg run zuul on x86 vs something else.21:18
ja... and it's another "good pub writing" practice, that whereever you give readers an option you give/point to the criteria they'd use to decide.21:18
krtaylorja, good point21:19
krtaylorI'll look at the doc sdague did for gerrit and see how it is linked in the doc21:20
krtaylorok fairly straightforward, I'll take that task21:25
jakrtaylor, I think I'm done etherpadding for today.  I'll probably fuss with getting git-review and friends config'd tomw AM; already installed git and git-review, so if anything else I should be doing to prep feel free to shoot email with particulars at me.21:43
*** ja has quit IRC21:48
*** jaypipes has quit IRC22:37
krtaylor will be back on later, have to do some schools stuff for my son22:40
asselin_https://review.openstack.org/#/c/149032/ updated22:58
*** patrickeast has joined #openstack-sprint23:08
pleia2asselin_: reviewing23:25
*** mestery has quit IRC23:27
jheskethHey guys23:33
jheskethhow can I help :-)23:33
*** mestery has joined #openstack-sprint23:34
pleia2o/ jhesketh23:34
pleia2jhesketh: not many patches in flight, but what we have is here: https://review.openstack.org/#/q/topic:third-party-ci-documentation,n,z23:35
jheskethcool, I'll review23:35
asselin_what do you think. Should this be used? http://www.joinfu.com/2014/01/understanding-the-openstack-ci-system/#Jenkins_Job_Creation_and_Configuration Or is there a better place that explains JJB?23:35
asselin_I found this: http://ci.openstack.org/jenkins-job-builder/definition.html23:35
pleia2yeah, there's also http://ci.openstack.org/jjb.html23:36
pleia2both have examples already23:36
pleia2leveraging one or both would be my preference23:36
asselin_I like http://ci.openstack.org/jjb.html#yaml-format23:38
pleia2yeah23:38
*** gothicmindfood has quit IRC23:42
pleia2woo, one patch down! thanks jhesketh23:47
jhesketh:-)23:47
asselin_great! I'll rebase on that now. :)23:48
asselin_well before my next patch set23:48

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!