| *** jplf_ has joined #openstack-jjb | 00:04 | |
| *** abelur_ has joined #openstack-jjb | 00:05 | |
| *** jplf has quit IRC | 00:06 | |
| *** jplf_ is now known as jplf | 00:06 | |
| *** abelur has quit IRC | 00:06 | |
| *** abelur_ is now known as abelur | 00:06 | |
| *** caphrim007_ has quit IRC | 00:30 | |
| *** alphadose[m] has quit IRC | 00:38 | |
| *** alphadose[m] has joined #openstack-jjb | 00:59 | |
| *** kien-ha has joined #openstack-jjb | 01:37 | |
| *** caphrim007 has joined #openstack-jjb | 02:21 | |
| *** caphrim007 has quit IRC | 02:26 | |
| *** caphrim007 has joined #openstack-jjb | 02:36 | |
| *** caphrim007 has quit IRC | 02:44 | |
| *** caphrim007 has joined #openstack-jjb | 02:45 | |
| *** kien-ha has quit IRC | 05:27 | |
| *** hashar has joined #openstack-jjb | 05:59 | |
| openstackgerrit | Veluri Mithun proposed openstack-infra/jenkins-job-builder master: Use convert_mapping_to_xml helper function https://review.openstack.org/571096 | 06:05 |
|---|---|---|
| *** ssbarnea_ has joined #openstack-jjb | 06:58 | |
| *** hashar has quit IRC | 09:23 | |
| *** hashar has joined #openstack-jjb | 09:23 | |
| *** hashar is now known as hasharAway | 09:50 | |
| electrofelix | ssbarnea zxiiro: regarding https://review.openstack.org/#/c/567943/ I think either that should be reverted or completely remove kerberos support from python-jenkins, as removing the package when the test fails on some peoples system instead of fixing the cause of the failure probably means that kerberos support is broken | 10:02 |
| ssbarnea | electrofelix: this is a tricky issue. currently we do not test kerberos as we mainly cannot. | 10:06 |
| electrofelix | right, but the error reported as the reason to remove the package from the tests suggests that something is likely broken for someone when they have an env for using kerberos but now we'll no longer spot it so we'll continue to break it further | 10:07 |
| electrofelix | In fact that error also means that someone with the kerberos-requests package potentially even without using kerberos will find python-jenkins broken for them, there's no way to tell | 10:08 |
| ssbarnea | true | 10:08 |
| openstackgerrit | Sorin Sbarnea proposed openstack/python-jenkins master: Revert "Do not require requests-kerberos for tests" https://review.openstack.org/571160 | 10:10 |
| ssbarnea | revert is simple, fixing it is not. mainly we should test both, with and without the presence of requests-kerberos packages | 10:11 |
| ssbarnea | but this means new tox target which is not something I enjoy doing, implementing new gate.... | 10:12 |
| ssbarnea | maybe someone has a better idea for testing that it works with both cases. | 10:12 |
| *** ssbarnea_ has quit IRC | 10:42 | |
| *** ssbarnea_ has joined #openstack-jjb | 11:19 | |
| electrofelix | yeah, looks like it's going to be messy to fix with other dependencies having moved forward, I'll see what I can do over the next few days, want to catch up a bit | 12:21 |
| Odd_Bloke | I've just filed https://bugs.launchpad.net/python-cliff/+bug/1774176; looks like we'll need that fixed before any Python 2 gates will pass. | 12:29 |
| openstack | Launchpad bug 1774176 in cliff "Uninstallable using Python 2 (due to cmd2 dropping Py2 support)" [Undecided,New] | 12:29 |
| electrofelix | starting to look like there might have be another major release in 3-6 months where we drop python 2 rather than trying to continue to support | 12:32 |
| openstackgerrit | Vicky Chijwani proposed openstack-infra/jenkins-job-builder master: Add retain_anchors config option. https://review.openstack.org/571197 | 12:51 |
| *** vickyc has joined #openstack-jjb | 12:59 | |
| *** caphrim007 has quit IRC | 13:06 | |
| openstackgerrit | Vicky Chijwani proposed openstack-infra/jenkins-job-builder master: Add retain_anchors config option. https://review.openstack.org/571197 | 13:19 |
| *** kien-ha has joined #openstack-jjb | 14:29 | |
| zxiiro | ok I think I subscribed to the storyboard | 14:38 |
| zxiiro | electrofelix: FYI I think we should give some attention to this issue https://storyboard.openstack.org/#!/story/2001508 as the guy has created a fork of JJB on pypi because of this. | 14:43 |
| zxiiro | we should probably resolve it and ask him to delete his fork to not confuse our users | 14:44 |
| electrofelix | agreed I guess this is what comes from not noticing issues | 14:50 |
| zxiiro | kien-ha: can you help us clean up the storyboard for jjb? I suspect there's a bunch of easy ones that can be closed. | 14:51 |
| kien-ha | sure I'll do what I can | 14:56 |
| ssbarnea | electrofelix: regarding the SB,... I don't know where to start,... i was planning to write a long blog post about what is wrong with it, mainly is the *UX*. In the end you figured out how to subcribe... and you get few good years go software engineering ;). Should I add impossible to edit descriptions/comments, the project URL is a ... number, very "friendly". Impossible to configure pagination, always will reset,... | 15:33 |
| electrofelix | I guess there is a question, do we abandon it until it's fully working or do we take a philosophical view that we should support a fully open source solution and try to help point out it's defects until it's as good as all the other solutions | 15:35 |
| Odd_Bloke | Launchpad is open source, if that's a concern. | 15:37 |
| ssbarnea | i know for sure that i don't have bandwidth to fix it, i am already overloaded by minor contributions to other projects and fixing SB would need a full time job, if not more. | 15:37 |
| ssbarnea | Odd_Bloke: I know is open source, but maybe not as easy to use. not to mention that I only know one implementation. Last time I attempted to setup my own instance I failed miserably (8+ years ago), and realised that it was not really off the shelf ;) | 15:39 |
| ssbarnea | anyway, there is no pressure to make a decision | 15:39 |
| Odd_Bloke | ssbarnea: Sure, but you couldn't even try to do that with Github without forking over a large amount of cash. | 15:40 |
| ssbarnea | the cmd2 seems like a serious issue. | 15:40 |
| Odd_Bloke | And, yeah, it certainly isn't off-the-shelf. | 15:40 |
| Odd_Bloke | But we wouldn't need to run an instance of it for JJB, we could just use launchpad.net (like much of Openstack). | 15:40 |
| Odd_Bloke | Full disclosure: I work for Canonical, so I already use Launchpad a _lot_ anyway. | 15:42 |
| Odd_Bloke | So I'm going to be biased. :p | 15:42 |
| *** caphrim007 has joined #openstack-jjb | 15:44 | |
| ssbarnea | can launchpad browse code hosted on github? | 15:44 |
| ssbarnea | one plus for LP is the python-jenkins already have a project there. | 15:45 |
| Odd_Bloke | ssbarnea: I think we just wouldn't host any code on Launchpad and direct people to Github. | 15:46 |
| Odd_Bloke | https://launchpad.net/nova does that in the project description, and there's just one random branch if you go to the Code tab. | 15:46 |
| ssbarnea | another question would be if we can have an unified view for both projects jjb and python-jenkins. | 15:46 |
| Odd_Bloke | Bugs can affect more than one project in Launchpad. | 15:47 |
| ssbarnea | still, the "code" tag is misleading users with last chage 2015. | 15:47 |
| ssbarnea | somehow I have the impression that nova received few... more recent contributions. | 15:47 |
| Odd_Bloke | Right, that's not an official branch. | 15:48 |
| ssbarnea | i like GH because they have very good code browser and is very easy to put a link to the code using that, for thus I find a benefit to keep issue tracker and code on the same system. | 15:48 |
| ssbarnea | sadly gerrit lacks on this area too | 15:49 |
| Odd_Bloke | Well, if we aren't using PRs, then some of the linking becomes less useful, right? | 15:49 |
| ssbarnea | on the other hand, going github could bring too much publicity, too easy to raise bugs. | 15:54 |
| ssbarnea | i seen it in the past | 15:55 |
| Odd_Bloke | I would, honestly, be happy with either. | 15:56 |
| zxiiro | Personally I'm fine with either but we need to make sure we can address user concerns in whatever system we use. I'm happy to give SB a try too now that I figured out how to subscribe | 15:57 |
| zxiiro | if we do decide to swtich to GitHub we need to update the instructions on how to raise a bug with us | 15:57 |
| zxiiro | because it currently points people to SB | 15:57 |
| electrofelix | ssbarnea: I wouldn't expect us to fix SB, just to report it's failings to those working on it | 15:58 |
| zxiiro | I'm confident the openstack-infra guys will allow us to move to GH if that's our final decisino but I'd like us all to fully agree on it before we make any major jump. | 15:58 |
| ssbarnea | sure. | 15:58 |
| zxiiro | I do agree with electrofelix though in that if we use GH issues then our users will have the expectation that we will accept PRs which we don't. and there's a whole other mess with that route :) | 15:59 |
| ssbarnea | lets fix urgent issues first, cmd2 | 15:59 |
| ssbarnea | trying to figure out what we can do. | 15:59 |
| zxiiro | ssbarnea: oh yeah, so we fixed that in OpenDyalight by pinning cmd2 to 0.8.5 | 15:59 |
| zxiiro | maybe we can push that into requirements.txt as a temporary workaround? | 15:59 |
| ssbarnea | link? | 16:00 |
| Odd_Bloke | The OpenStack requirements project has a slightly better fix, I think. | 16:00 |
| Odd_Bloke | Pin it for Python 2, but not for Python 3. | 16:00 |
| zxiiro | Odd_Bloke: oh? what did they do? | 16:00 |
| electrofelix | lets try getting SB working for 1 month and then report on the deficiencies? can then decide | 16:00 |
| ssbarnea | i was considering cmd<=0.85;python.... | 16:00 |
| Odd_Bloke | Yeah, something like that. | 16:00 |
| ssbarnea | like pinning only for unsupported versions of python. | 16:00 |
| zxiiro | OpenDaylight just did a simple cmd==0.8.5 | 16:00 |
| *** hasharAway is now known as hashar | 16:01 | |
| zxiiro | cmd2==0.8.5 | 16:01 |
| electrofelix | pin for py2 with env marker and that makes it clear when removing python2 support in the future that we can drop the pin | 16:02 |
| Odd_Bloke | I'd prefer cmd2<0.9. | 16:02 |
| Odd_Bloke | cmd2 0.8.x is up to 0.8.7. | 16:02 |
| zxiiro | +1 I like the more specific pinning | 16:02 |
| Odd_Bloke | Which one are you +1'ing? | 16:02 |
| Odd_Bloke | I'd say mine is less specific. :p | 16:03 |
| zxiiro | cmd2<0.9 against py2 | 16:03 |
| openstackgerrit | Sorin Sbarnea proposed openstack/python-jenkins master: Freeze cmd2 library on python2 https://review.openstack.org/571243 | 16:03 |
| electrofelix | is there an issue with the 0.8.3 version? | 16:04 |
| ssbarnea | electrofelix: this is copy/paste from openstack, i bet there was a problem | 16:05 |
| Odd_Bloke | I believe that's following... yeah, that. | 16:05 |
| electrofelix | ah | 16:05 |
| ssbarnea | i am wondering if it would not make sense to point to openstack requirements directly, so we would delegate these issues to them ;) | 16:06 |
| electrofelix | does the check-requirements job take care of such issues? | 16:06 |
| electrofelix | marked it CR+2 but left the workflow as is for the moment | 16:07 |
| ssbarnea | yeah, me waiting for ci. | 16:08 |
| *** electrofelix has quit IRC | 16:08 | |
| openstackgerrit | Merged openstack/python-jenkins master: Freeze cmd2 library on python2 https://review.openstack.org/571243 | 16:37 |
| *** caphrim007 has quit IRC | 16:40 | |
| openstackgerrit | Sorin Sbarnea proposed openstack/python-jenkins master: Revert "Do not require requests-kerberos for tests" https://review.openstack.org/571160 | 16:51 |
| *** hashar is now known as hasharAway | 17:00 | |
| *** ssbarnea_ has quit IRC | 17:27 | |
| openstackgerrit | Veluri Mithun proposed openstack-infra/jenkins-job-builder master: Use convert_mapping_to_xml helper function https://review.openstack.org/571096 | 17:30 |
| openstackgerrit | Veluri Mithun proposed openstack-infra/jenkins-job-builder master: Remove trailing white space and convert tabs to spaces https://review.openstack.org/571262 | 17:30 |
| *** ssbarnea_ has joined #openstack-jjb | 17:45 | |
| *** caphrim007 has joined #openstack-jjb | 17:45 | |
| *** caphrim007 has quit IRC | 17:47 | |
| *** caphrim007 has joined #openstack-jjb | 17:47 | |
| openstackgerrit | Veluri Mithun proposed openstack-infra/jenkins-job-builder master: Use convert_mapping_to_xml helper function https://review.openstack.org/571278 | 18:16 |
| *** Mithun has joined #openstack-jjb | 18:21 | |
| openstackgerrit | Veluri Mithun proposed openstack-infra/jenkins-job-builder master: Use convert_mapping_to_xml helper function https://review.openstack.org/571278 | 18:24 |
| *** ssbarnea_ has quit IRC | 18:31 | |
| *** Mithun has quit IRC | 18:45 | |
| *** ssbarnea_ has joined #openstack-jjb | 19:33 | |
| Odd_Bloke | Does someone want to throw the StoryBoard project link (https://storyboard.openstack.org/#!/project/723) in to the topic? | 19:40 |
| Odd_Bloke | Not sure how to do that. | 19:40 |
| Odd_Bloke | Not being able to triage importance in StoryBoard is annoying. | 19:46 |
| Odd_Bloke | Also, I assume these were imported from somewhere; YAML is semi-valid MarkDown and none of the issues I've looked at thus far wrapped the YAML in ```s to make it code. ;.; | 19:47 |
| *** kien-ha has quit IRC | 19:59 | |
| openstackgerrit | Daniel Watkins proposed openstack-infra/jenkins-job-builder master: Make the deprecated plugin config messaging clearer https://review.openstack.org/571300 | 20:27 |
| *** hasharAway has quit IRC | 20:51 | |
| openstackgerrit | Daniel Watkins proposed openstack-infra/jenkins-job-builder master: Handle string skip-vote parameters in gerrit trigger https://review.openstack.org/571311 | 21:02 |
| *** ssbarnea_ has quit IRC | 21:14 | |
| *** caphrim007_ has joined #openstack-jjb | 21:55 | |
| *** caphrim007 has quit IRC | 21:59 | |
| *** kien-ha has joined #openstack-jjb | 22:08 | |
| *** caphrim007_ has quit IRC | 23:47 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!