| *** rlandy|brb is now known as rlandy | 00:05 | |
| *** dsneddon has joined #oooq | 00:08 | |
| *** Goneri has quit IRC | 00:30 | |
| *** hamzy has joined #oooq | 00:31 | |
| *** apetrich has quit IRC | 01:57 | |
| *** rlandy has quit IRC | 02:55 | |
| *** ykarel|away has joined #oooq | 03:14 | |
| *** udesale has joined #oooq | 03:55 | |
| *** ykarel|away has quit IRC | 04:07 | |
| *** ykarel|away has joined #oooq | 04:23 | |
| *** ykarel_ has joined #oooq | 04:25 | |
| *** ykarel|away has quit IRC | 04:28 | |
| *** raukadah is now known as chandankumar | 04:40 | |
| *** ykarel_ is now known as ykarel | 04:41 | |
| *** ratailor has joined #oooq | 05:04 | |
| *** marios has joined #oooq | 05:42 | |
| *** ksambor has quit IRC | 05:55 | |
| *** ksambor has joined #oooq | 05:55 | |
| *** quiquell|off is now known as quiquell | 06:02 | |
| *** udesale has quit IRC | 06:05 | |
| *** udesale has joined #oooq | 06:05 | |
| *** jfrancoa has joined #oooq | 06:18 | |
| *** jaosorior has joined #oooq | 06:28 | |
| *** yolanda_ has joined #oooq | 06:31 | |
| *** jaosorior has quit IRC | 07:01 | |
| *** udesale has quit IRC | 07:03 | |
| *** udesale has joined #oooq | 07:03 | |
| *** apetrich has joined #oooq | 07:07 | |
| *** dsneddon has quit IRC | 07:07 | |
| *** skramaja has joined #oooq | 07:11 | |
| *** saneax has joined #oooq | 07:14 | |
| *** dsneddon has joined #oooq | 07:33 | |
| chandankumar | quiquell: hello | 07:41 |
|---|---|---|
| chandankumar | quiquell: https://bugs.launchpad.net/tripleo/+bug/1828172 | 07:42 |
| openstack | Launchpad bug 1828172 in tripleo "No match for argument: python2-dnf on tripleo reproducer libvirt Fedora 30" [High,Triaged] | 07:42 |
| chandankumar | quiquell: from where parts/libvirt role is coming? | 07:42 |
| quiquell | chandankumar: https://opendev.org/openstack/tripleo-quickstart/src/branch/master/roles/libvirt | 07:43 |
| quiquell | chandankumar: nop, this one https://opendev.org/openstack/tripleo-quickstart/src/branch/master/roles/parts/libvirt | 07:43 |
| *** ykarel is now known as ykarel|lunch | 07:44 | |
| *** jpena|off is now known as jpena | 07:59 | |
| *** bogdando has joined #oooq | 08:19 | |
| *** ykarel|lunch is now known as ykarel | 08:25 | |
| *** holser_ has joined #oooq | 08:31 | |
| marios | quiquell: does the user we create need passwordless sudo? or can be normal user? | 09:18 |
| quiquell | passwordless | 09:19 |
| marios | quiquell: k thanks sanity checking i already have it like that | 09:19 |
| marios | quiquell: echo "$REPRO_USER ALL=(root) NOPASSWD:ALL" | sudo tee -a /etc/sudoers.d/$REPRO_USER | 09:19 |
| arxcruz | panda: what's the image used to run the standalone job ? | 10:03 |
| panda | arxcruz: image ? | 10:04 |
| arxcruz | yeah, the centos image used | 10:04 |
| panda | arxcruz: in zuul is usually regenerated periodically | 10:05 |
| panda | arxcruz: so it changes frequently | 10:05 |
| arxcruz | panda: what's the image used in the reproducer ? | 10:05 |
| panda | arxcruz: when you are reproducing in rdocloud, it's the same thing as zuul | 10:06 |
| arxcruz | panda: which is? | 10:06 |
| panda | arxcruz: that image that changes periodically | 10:06 |
| panda | arxcruz: you want to download a qcow ? | 10:07 |
| arxcruz | I want to start a vm with that image | 10:07 |
| arxcruz | on rdocloud | 10:08 |
| panda | arxcruz: ok, in sf-ops you can ask someone to point you to the latest built image, it's usually on a server available to the public, but it's quite hidden, and those are cloud images, so you'll have to remove cloud-init, and specify a password | 10:08 |
| panda | arxcruz: OH, wait,m in Rdocloud | 10:08 |
| arxcruz | yup :) | 10:09 |
| *** jaosorior has joined #oooq | 10:11 | |
| arxcruz | panda: found it :) | 10:12 |
| panda | arxcruz: ok, I was trying to look in the nodeset configuration | 10:12 |
| *** holser_ is now known as holser|lunch | 10:15 | |
| panda | quiquell: ... I may need your help | 10:39 |
| *** zbr|pto has quit IRC | 10:42 | |
| quiquell | panda: blue ? | 10:42 |
| panda | quiquell: da bu di da bu da, da bu di , da bu da, da bu di da bu da ? | 10:43 |
| quiquell | ok going to you rblue | 10:46 |
| *** ykarel is now known as ykarel|afk | 10:51 | |
| chandankumar | quiquell: if we run the reproducer script as a root does it not warns as we need to have a user with password less previliage? | 10:55 |
| *** jaosorior has quit IRC | 10:56 | |
| quiquell | nope | 10:56 |
| *** panda is now known as panda|lunch | 11:04 | |
| *** udesale has quit IRC | 11:14 | |
| *** holser|lunch is now known as holser_ | 11:27 | |
| chandankumar | marios: panda|lunch http://paste.openstack.org/show/751103/ is this known when runnign reproducer with lobvirt on centos 7 | 11:29 |
| marios | chandankumar: dont think i've seen that one | 11:30 |
| chandankumar | marios: it is a fresh centos 7 installation | 11:31 |
| *** jpena is now known as jpena|lunch | 11:31 | |
| marios | chandankumar: try running the command manually to see why it fails | 11:32 |
| marios | chandankumar: the virt-customize but | 11:32 |
| marios | bit | 11:32 |
| chandankumar | marios: still failing, changing the selinux and trying out | 11:35 |
| chandankumar | do you keep selinux to permissive or enforcing? | 11:35 |
| marios | chandankumar: i never change it so is whatever beaker/rdo give i would expect enforcing | 11:35 |
| *** quiquell is now known as quiquell|lunch | 11:36 | |
| marios | chandankumar: but you can try with setenforce 0 to test | 11:36 |
| *** rlandy has joined #oooq | 11:42 | |
| *** amoralej is now known as amoralej|lunch | 11:52 | |
| weshay | morning | 11:54 |
| marios | o/ weshay see email | 11:55 |
| *** ykarel|afk is now known as ykarel | 11:56 | |
| *** jaosorior has joined #oooq | 11:56 | |
| weshay | arxcruz marios rlandy panda|lunch interview | 11:58 |
| arxcruz | weshay:we are there | 11:58 |
| chandankumar | weshay: https://bugs.launchpad.net/tripleo/+bug/1828172 -> from where python2-dnf comes? | 12:12 |
| openstack | Launchpad bug 1828172 in tripleo "No match for argument: python2-dnf on tripleo reproducer libvirt Fedora 30" [High,Triaged] | 12:12 |
| *** quiquell|lunch is now known as quiquell | 12:14 | |
| weshay | chandankumar nice thanks for trying it | 12:14 |
| chandankumar | weshay: I am on f30 already donot want to rollback | 12:14 |
| weshay | chandankumar aye | 12:15 |
| *** ratailor has quit IRC | 12:20 | |
| *** panda|lunch is now known as panda | 12:24 | |
| *** zbr has joined #oooq | 12:25 | |
| *** zbr has quit IRC | 12:26 | |
| quiquell | weshay: have a revie using just local gerrit it's starting to work (but does not support Depends-On yet) | 12:29 |
| *** jpena|lunch is now known as jpena | 12:31 | |
| rfolco|ruck | quiquell, do we have a tag for reproducer bugs ? | 12:52 |
| panda | we definitely should | 12:53 |
| rfolco|ruck | quiquell, panda: have you seen this before? can't access mirrors when deploying... http://10.9.113.13:8000/01/1001/1/check/tripleo-ci-centos-7-standalone-dlrn-hash-tag/03491cd/logs/undercloud/home/zuul/standalone_deploy.log.txt.gz | 12:54 |
| marios | quiquell: http://logs.openstack.org/79/657779/1/check/tripleo-ci-centos-7-undercloud-containers/13c6662/logs/reproducer-quickstart/reproducer-fresh-box-setup.sh | 12:54 |
| marios | :/ | 12:55 |
| marios | quiquell: and we add docs in http://logs.openstack.org/79/657779/1/check/tripleo-ci-centos-7-undercloud-containers/13c6662/logs/reproducer-quickstart/reproducer-fresh-box-setup.sh | 12:55 |
| *** sanjayu_ has joined #oooq | 12:55 | |
| marios | quiquell: err http://logs.openstack.org/79/657779/1/check/tripleo-ci-centos-7-undercloud-containers/13c6662/logs/README-reproducer.html | 12:55 |
| quiquell | rfolco|ruck: zuul-reproducer is the tag | 12:57 |
| *** saneax has quit IRC | 12:58 | |
| quiquell | marios: cool, will take a look, I am hacking something now | 12:58 |
| marios | quiquell: ack gonna update the docs some more (trying to tidy up some more things there) added it to the create reproducer role in https://review.opendev.org/657779 | 12:59 |
| quiquell | marios: btw the autohold is no longer needed | 12:59 |
| quiquell | marios: the role does it automatically | 12:59 |
| marios | quiquell: ack good to know thanks | 12:59 |
| weshay | panda | 13:08 |
| *** aakarsh has quit IRC | 13:13 | |
| *** amoralej|lunch is now known as amoralej | 13:32 | |
| *** Goneri has joined #oooq | 13:35 | |
| arxcruz | rlandy: I've been working with tristan this morning, and I got zuul-runner working | 13:36 |
| rlandy | arxcruz: nice | 13:36 |
| rlandy | arxcruz: is it easier to use? | 13:36 |
| arxcruz | rlandy: I am executing right now tripleo-ci-centos-7-standalone | 13:36 |
| rlandy | arxcruz: I took on the rhel 8 work | 13:37 |
| arxcruz | rlandy: once the zuul patch get merged it's just a matter of call one command | 13:37 |
| rlandy | arxcruz: what is your impression? | 13:37 |
| arxcruz | rlandy: of course, you must have the vm already prepared | 13:37 |
| rlandy | devil in the details | 13:37 |
| arxcruz | rlandy: yeah, so what I did was run the old reproducer to prepare the machine | 13:37 |
| arxcruz | with all nodepool ci etc | 13:37 |
| arxcruz | and the directories that tqe assume it exists in the vm | 13:38 |
| rlandy | so what did we really replace? | 13:38 |
| arxcruz | then just issue the command zuul-runner --verbose --tenant runner --api http://localhost:9000/ --project openstack/tripleo-ci --job tripleo-ci-centos-7-standalone execute --nodes ssh:centos-7:38.145.34.208:zuul:/home/zuul | 13:38 |
| arxcruz | now i'm waiting it finishes | 13:38 |
| arxcruz | and | 13:38 |
| arxcruz | 2019-05-08 13:36:42 | Deployment successful! | 13:38 |
| arxcruz | rlandy: no need to run zuul service | 13:39 |
| arxcruz | rlandy: I update the doc | 13:39 |
| *** vinaykns has joined #oooq | 13:39 | |
| quiquell | arxcruz: sounds promising at the end for libvirt we are preparing already the nodes in a playbook we can reuse that | 13:39 |
| arxcruz | quiquell: rlandy yes, and we are not using the same vm as upstream which already have all this set perhaps that's why | 13:43 |
| arxcruz | but for me was pretty straightfoward download a reproducer script from a job setup the vm properly | 13:43 |
| arxcruz | and run the zuul-runner command | 13:43 |
| rlandy | I think evaluating all the use cases will lead to whether is this more usable or not | 13:44 |
| arxcruz | well, it worked :D | 13:45 |
| arxcruz | i'll execute again to fix the cirros stuff | 13:46 |
| arxcruz | but yeah, worked with standalone job | 13:46 |
| rfolco|ruck | arxcruz, nice, good news | 13:50 |
| rfolco|ruck | arxcruz, if you can point me to a doc/instructions I can also test this path | 13:51 |
| *** ykarel is now known as ykarel|away | 13:52 | |
| *** aakarsh has joined #oooq | 13:52 | |
| *** quiquell is now known as quiquell|brb | 13:53 | |
| *** ykarel|away has quit IRC | 13:59 | |
| *** sanjayu__ has joined #oooq | 14:01 | |
| *** sanjayu_ has quit IRC | 14:04 | |
| *** quiquell|brb is now known as quiquell | 14:15 | |
| chandankumar | Friends, if you are trying on f30, check this bug https://bugs.launchpad.net/tripleo/+bug/1828172 I made progress | 14:24 |
| openstack | Launchpad bug 1828172 in tripleo "No match for argument: python2-dnf on tripleo reproducer libvirt Fedora 30" [High,Triaged] | 14:24 |
| chandankumar | we need to do a mass update the package list to py3 with few workarounds | 14:24 |
| chandankumar | rlandy: Hello | 14:25 |
| rlandy | chandankumar: ack | 14:25 |
| rlandy | chandankumar: I am keeping a list of what needs updating wrt reproducer on rhel-8 | 14:25 |
| rlandy | it's a lot | 14:25 |
| rlandy | https://etherpad.openstack.org/p/rhel-8-reproducer | 14:25 |
| chandankumar | rlandy: today I have presented a talk on zuul in pune office got hit 2 questions | 14:25 |
| rlandy | chandankumar: ok ... go ahead | 14:26 |
| chandankumar | rlandy: how nodepool provisions nodes on baremetal and if we create a test node does it has nested virt supported if needed during testing | 14:26 |
| chandankumar | ? | 14:26 |
| rlandy | chandankumar: so there are two options ... | 14:26 |
| rlandy | you can add baremetal nodes to nodepool with the static driver | 14:27 |
| rlandy | we went down that route at one point and ditched it | 14:27 |
| chandankumar | one of the code ready (openshift 4 installer) developer was asking they wanted an infra where they can test their stuff | 14:27 |
| rlandy | it's not clean | 14:27 |
| rlandy | chandankumar: here is what we do ... | 14:27 |
| chandankumar | currently they are using centos ci but it is not feasible | 14:27 |
| rlandy | we use a nodepool node to act just a zuul operator | 14:28 |
| rlandy | and we do the real work on the baremetal box that we ssh to | 14:28 |
| rlandy | so zuul has no direct contact with the baremetal node | 14:28 |
| rlandy | it's a much cleaner break | 14:28 |
| rlandy | but there are complication as there are a lot of assumptions that the node you are testing on is a noedpool | 14:29 |
| rlandy | nodepool node | 14:29 |
| rlandy | chandankumar: the developers can contact me and I'll take then through it | 14:29 |
| rlandy | it works well for us | 14:29 |
| *** ykarel|away has joined #oooq | 14:29 | |
| chandankumar | rlandy: I will introduce with him tomorrow | 14:30 |
| chandankumar | rlandy: thanks :-) | 14:31 |
| rlandy | chandankumar: sure - we would be happy to help them get free of ci.centos - it was not workable for us either | 14:31 |
| weshay | arxcruz I'm ready in a few | 14:31 |
| weshay | arxcruz oh crap I have to run the cix call | 14:32 |
| arxcruz | weshay:we can do later, i have german class in a few, when I get back :) | 14:32 |
| chandankumar | rlandy: another question why openshift guys are not using zuul, and I have no answers | 14:37 |
| *** tesseract has joined #oooq | 14:38 | |
| rlandy | chandankumar: they have their own test system | 14:39 |
| rlandy | trown mentioned it and seemed to think it was pretty cool | 14:40 |
| chandankumar | they asked beacuse of cross dependency and speculative merging nature | 14:40 |
| rlandy | I don;t know the real answer here - it could be technical or organizational | 14:40 |
| chandankumar | prown a wrapper over jenkins | 14:40 |
| chandankumar | *prow | 14:41 |
| rlandy | well anything that gets you away from jenkins is a plus | 14:41 |
| quiquell | zuulv2 was a wrapper around jenkins, isn't it ? | 14:41 |
| rlandy | I did discuss this with trown briefly at one point - where we were looking at using baremetal to test openshift on openstack | 14:42 |
| rlandy | my every limited understanding is that there is a lot of work on aws | 14:43 |
| rlandy | and some testing done there | 14:43 |
| rlandy | trown mentioned vexxhost as well | 14:43 |
| rlandy | really that is more a question to the openshift guys that we can answer | 14:44 |
| weshay | rfolco|ruck fix-released is closed fyi. re: https://bugs.launchpad.net/tripleo/+bug/1826452 | 14:44 |
| openstack | Launchpad bug 1826452 in tripleo "pip install tripleo-repos: DistributionNotFound" [Critical,Fix released] - Assigned to Sorin Sbarnea (ssbarnea) | 14:44 |
| weshay | arxcruz ping bj | 14:44 |
| *** Vorrtex has joined #oooq | 14:45 | |
| arxcruz | weshay:ok | 14:45 |
| *** skramaja has quit IRC | 14:47 | |
| *** jaosorior has quit IRC | 14:47 | |
| *** rfolco|ruck is now known as rfolco|out | 14:48 | |
| rlandy | oh my gosh - this python2/python3 thing will kill me | 14:49 |
| quiquell | rlandy: btw we should have a f30 nodepool images at RDO, yesterday they merged a change to have them | 14:50 |
| rlandy | setting the interpreter to python3 gives ansible a fit when it hits the centos7 nodes | 14:50 |
| rlandy | quiquell: thanks- working on rhel-8 | 14:50 |
| rlandy | to be exact a centos7 deploy on a rel-8 host | 14:50 |
| rlandy | so python of two versions not living happily together | 14:50 |
| quiquell | humm looks like for rhel8 there is only python2 or python3 but not python | 14:52 |
| quiquell | so shebags need to point to either of them | 14:52 |
| rlandy | rhel 8 is pythonn 3 only | 14:52 |
| quiquell | you have python2 too | 14:52 |
| rlandy | which is ok - got a bunch of changes to get by that | 14:52 |
| rlandy | but centos 7 is python2 | 14:53 |
| rlandy | so need to sync with that | 14:53 |
| quiquell | btw, wip with reproducer without accessing gerrits https://review.rdoproject.org/r/#/c/20597/ | 14:53 |
| quiquell | weshay: ^ but it does not support Depends-On | 14:53 |
| quiquell | just running plain jobs | 14:53 |
| quiquell | droping now, read you tomorrow | 14:54 |
| *** quiquell is now known as quiquell|off | 14:54 | |
| marios | rlandy: i remember you mentioned this a while back, added some info on the private network creation for first run @ https://review.opendev.org/#/c/657779/2/roles/create-zuul-based-reproducer/templates/README-reproducer-zuul-based-quickstart.html.j2@99 i'll point to the rendered file once the jobs run there | 14:56 |
| *** ykarel|away is now known as ykarel | 14:57 | |
| rlandy | marios: yeah - not sure what happened there | 15:09 |
| rlandy | I think we had it in as small note | 15:09 |
| *** tesseract has quit IRC | 15:09 | |
| rlandy | or maybe we removed it? | 15:09 |
| chandankumar | rlandy: I reached to the same state in zuul reproducer as you hit for rhel 8 | 15:10 |
| *** ccamacho has joined #oooq | 15:10 | |
| marios | rlandy: ack not sure. its a bit of a weird one as unlike all the other things, that one is really first time only | 15:11 |
| rlandy | chandankumar: you also ran on rhel-8? | 15:11 |
| chandankumar | rlandy: I am on fedora 30 with python3 | 15:12 |
| chandankumar | Here is the notes https://bugs.launchpad.net/tripleo/+bug/1828172 | 15:12 |
| openstack | Launchpad bug 1828172 in tripleo "No match for argument: python2-dnf on tripleo reproducer libvirt Fedora 30" [High,Triaged] | 15:12 |
| rlandy | chandankumar: so I get as far as [libvirt/setup/overcloud : Set hostname correctly for subnode-0] | 15:12 |
| rlandy | here this is a centos-7 node | 15:12 |
| rlandy | and I have set -e python_interpreter=/usr/bin/python3 | 15:13 |
| chandankumar | yes, I need to fix some part | 15:13 |
| rlandy | in the ansible call | 15:13 |
| chandankumar | I have already setted python3 to fix libvirt installation | 15:13 |
| rlandy | which means I have to override that for the nodes themselves | 15:13 |
| rlandy | correct | 15:13 |
| rlandy | that first setting gets by the libvirt install | 15:14 |
| * rlandy looks into set_fact | 15:14 | |
| rlandy | TASK [libvirt/setup/overcloud : Add subnode-0 to inventory] ******************************************************************************************************************************* | 15:15 |
| rlandy | task path: /tmp/reproduce-tmp.hBqSE/roles/libvirt/setup/overcloud/tasks/register_vm.yml:20 | 15:15 |
| rlandy | creating host via 'add_host': hostname=subnode-0 | 15:15 |
| rlandy | changed: [localhost] => {"add_host": {"groups": ["subnodes"], "host_name": "subnode-0", "host_vars": {"ansible_fqdn": "192.168.122.57", "ansible_host": "192.168.122.57", "ansible_python_interpreter": "/usr/bin/python2", "ansible_user": "zuul", "subnode_private_ip": "192.168.122.57", "subnode_public_ip": "192.168.122.57"}}, "changed": true} | 15:15 |
| rlandy | ^^ interesting thing | 15:15 |
| rlandy | chandankumar: for the node itself, I have tried to set multiple values for "ansible_python_interpreter" | 15:15 |
| rlandy | in this node, it should be "/usr/bin/python2" | 15:15 |
| chandankumar | ok let me try that | 15:16 |
| rlandy | tried -e consul_install_remotely=true | 15:17 |
| rlandy | chandankumar: ^^ didn't help me either | 15:17 |
| rlandy | chandankumar: going to try set a hostvar | 15:17 |
| chandankumar | not tried this one | 15:18 |
| rlandy | maybe these mixed marriages are no good | 15:18 |
| rlandy | python 3 guest on python 3 host | 15:18 |
| rlandy | and don;t support this any version guest on only version host | 15:19 |
| *** altlogbot_0 has quit IRC | 15:21 | |
| *** altlogbot_2 has joined #oooq | 15:21 | |
| *** irclogbot_3 has quit IRC | 15:22 | |
| *** irclogbot_3 has joined #oooq | 15:23 | |
| weshay | quiquell|off dang | 15:25 |
| weshay | missed him... quiquell|off ping me tomorrow re: reproducer | 15:25 |
| rlandy | chandankumar: trying removing the "ansible_python_interpreter" from -e and setting it per host | 15:25 |
| quiquell|off | weshay: I am semi off | 15:26 |
| *** Goneri has quit IRC | 15:26 | |
| rlandy | quiquell|off can't leave us :) | 15:27 |
| *** altlogbot_2 has quit IRC | 15:27 | |
| weshay | quiquell|off want to chat live now or tomorrow | 15:27 |
| *** altlogbot_1 has joined #oooq | 15:27 | |
| quiquell|off | Let's blue in a few | 15:28 |
| *** irclogbot_3 has quit IRC | 15:28 | |
| *** irclogbot_3 has joined #oooq | 15:29 | |
| marios | weshay: quiquell|off fyi fresh-box.sh and docs in https://review.opendev.org/#/c/657779/ waiting for one of the jobs to finish so i can sanity check (and docs will be nicer to read, hopefully making it clearer) | 15:30 |
| weshay | quiquell|off k.. just ping me | 15:30 |
| rlandy | chandankumar: k - that worked | 15:30 |
| *** Goneri has joined #oooq | 15:30 | |
| lmiccini | o/ anyone having few minutes to check https://review.opendev.org/#/c/656937/ ? | 15:31 |
| marios | anyone else seen this thing? I had it on centos rdo cloud vm http://paste.openstack.org/raw/750874/ it comes from this task https://github.com/rdo-infra/ansible-role-tripleo-ci-reproducer/blob/79034622331b0d4386d2b68b4e95048a265737e9/tasks/launch-job.yaml#L20 am looking to understand why it broke and hopefuly fix but wondering if anyone else saw it | 15:32 |
| quiquell|off | weshay: now | 15:33 |
| weshay | k | 15:33 |
| *** tcw has joined #oooq | 15:34 | |
| weshay | quiquell|off I'm in | 15:34 |
| marios | quiquell|off: weshay are you throwing an exception at my proposal and not responding at all (no response worse than any ;) ) | 15:34 |
| marios | lmiccini: ack | 15:34 |
| lmiccini | marios: cheers | 15:34 |
| weshay | marios re: anyone else seen this? | 15:34 |
| marios | weshay: no the fresh-box.sh | 15:34 |
| rlandy | marios: ^^ no - but I don't have a split personality situation - one user name and key on all | 15:35 |
| marios | weshay: thing i was on about yesterday for where you want to run reproducer on some new rdo cloud vm or beaker box | 15:35 |
| weshay | oh sorry lots of things | 15:35 |
| * weshay looking now | 15:35 | |
| marios | weshay: fyi fresh-box.sh and docs in https://review.opendev.org/#/c/657779/ | 15:35 |
| weshay | ah there we go | 15:35 |
| * weshay looks | 15:35 | |
| marios | weshay: the readme will make more sense once the jobs report though so you can find it | 15:35 |
| marios | weshay: i added the freshboot into the create reproducer role so it lives in the logs. the readme just points to it so they know they can/should run it | 15:36 |
| marios | weshay: also added commands for 'you need a network called private that is externally routable' as some folks will not have that (I didn't) | 15:38 |
| marios | weshay: so in general try to make it a bit more easy for that first time case, and especially when they are gonna use a cattle not a pet (i used those terms in the readme) | 15:38 |
| rlandy | starting podman - we're getting there!! | 15:39 |
| marios | what does one call a singular cattle? | 15:39 |
| * marios asking the important questions | 15:39 | |
| rlandy | a cow? | 15:39 |
| marios | ha | 15:40 |
| marios | yah guess? | 15:40 |
| rlandy | a special cow? | 15:40 |
| marios | single instance of cow() | 15:40 |
| rlandy | a pet cow | 15:41 |
| marios | weshay: when are we calling the test day. is it over tomorrow? imo we should keep it going till end of sprint. though nothing else would likely happen like scrap sprint. not sure we can afford to do that. esp. standalone-upgrade still needs sorting. perhaps rhel8/osp15 not as urgent as still likely blocked | 15:42 |
| marios | weshay: hmm... maybe prune sprint board to essentials | 15:42 |
| marios | and keep going? | 15:42 |
| rlandy | marios: I have other tasks | 15:42 |
| rlandy | but others can keep going | 15:43 |
| rlandy | baremetal never sleeps | 15:43 |
| marios | rlandy: ack | 15:43 |
| marios | rlandy: just thinking out loud but don't know what weshay has in mind | 15:43 |
| * marios almost out | 15:43 | |
| * weshay looking now | 15:45 | |
| weshay | marios I like the patch and idea | 15:46 |
| weshay | marios the test day is over for you now | 15:47 |
| weshay | eob today | 15:47 |
| marios | weshay: ack ok so i know to refocus on sprint stuff tomorrow AM | 15:47 |
| weshay | I'll setup a review meeting for what we have found, fixed, etc.. | 15:47 |
| weshay | marios correct... | 15:47 |
| marios | weshay: ack thanks | 15:48 |
| *** sanjayu__ has quit IRC | 15:49 | |
| *** sanjayu__ has joined #oooq | 15:50 | |
| *** dtrainor has quit IRC | 15:51 | |
| chandankumar | time for deep nap | 16:08 |
| chandankumar | see ya tomorrow | 16:08 |
| *** chandankumar is now known as raukadah | 16:08 | |
| *** jfrancoa has quit IRC | 16:12 | |
| *** marios has quit IRC | 16:16 | |
| *** ykarel is now known as ykarel|away | 16:18 | |
| *** bogdando has quit IRC | 16:22 | |
| *** tcw has quit IRC | 16:35 | |
| *** vinaykns has quit IRC | 16:38 | |
| *** vinaykns has joined #oooq | 16:43 | |
| weshay | rlandy https://review.opendev.org/#/c/656014/ ? | 16:57 |
| *** holser_ has quit IRC | 17:04 | |
| *** jpena is now known as jpena|off | 17:10 | |
| *** jbadiapa has quit IRC | 17:19 | |
| *** tcw has joined #oooq | 17:24 | |
| *** jbadiapa has joined #oooq | 17:33 | |
| rlandy | weshay: I'm fine with https://review.opendev.org/#/c/656014 | 18:06 |
| weshay | rlandy ok.. I guess I need to test it.. will do so now | 18:07 |
| weshay | rlandy can we trigger the jobs that kick from ansible-role-tripleo-reproducer w/ these changes? | 18:07 |
| rlandy | {% if depends_on|bool == true -%} | 18:07 |
| weshay | or we just need molecule I guess | 18:08 |
| rlandy | just rechecking the spacing | 18:08 |
| rlandy | fine - because end of line | 18:09 |
| weshay | rlandy hit this on centos-7 https://review.rdoproject.org/r/20602 | 18:10 |
| rlandy | weshay: actually I was too quick ... | 18:11 |
| * weshay about to try it | 18:11 | |
| weshay | so was I ;) | 18:12 |
| rlandy | weshay: so the use_depends_on is never used | 18:13 |
| rlandy | and the default needs to stand at run time | 18:13 |
| weshay | ah dang | 18:13 |
| rlandy | not template create time | 18:13 |
| rlandy | see comments | 18:13 |
| weshay | should just be depends_on | 18:13 |
| rlandy | sorry to be a pain | 18:14 |
| rlandy | and sorry for delay - was watching the women's forum at summit - interesting | 18:14 |
| weshay | np or other way around | 18:14 |
| rlandy | comments make sense? | 18:14 |
| * rlandy checks other ereview | 18:14 | |
| weshay | launch_job_branch: master | 18:17 |
| weshay | {% if depends_on|bool == true -%} | 18:17 |
| weshay | ^ here | 18:17 |
| weshay | ya.. rlandy as long as you are good on the patch in theory | 18:17 |
| weshay | I'll fix the issues | 18:17 |
| rlandy | weshay: in theory - that si exactly what I was suggesting | 18:18 |
| rlandy | just default that to true | 18:18 |
| weshay | aye | 18:18 |
| rlandy | then it will resolve at the reproducer run time | 18:18 |
| weshay | patch is up | 18:18 |
| weshay | w/ the change | 18:18 |
| weshay | just have to patch my local version | 18:18 |
| * rlandy looks again | 18:18 | |
| * rlandy needs a podman debugging cheatsheet | 18:19 | |
| weshay | oh.. I do need to default that in the jinja2 | 18:19 |
| weshay | not just the role | 18:19 |
| weshay | because it's run time | 18:19 |
| rlandy | weshay: exactly | 18:21 |
| * rlandy looks if ok to default in j2 | 18:21 | |
| weshay | so that would be |default(true) or |bool|default(true) | 18:22 |
| * weshay forgets the syntax | 18:22 | |
| weshay | default("true")|bool | 18:23 |
| rlandy | usually the first but I don;t know if we ever default within j2 | 18:23 |
| rlandy | weshay: example: https://github.com/openstack/tripleo-quickstart-extras/blob/1aa9747a4eab3ee857e7dc7ae7e50862bbf64123/roles/validate-tempest/templates/configure-tempest.sh.j2#L123 | 18:25 |
| rlandy | can skip == | 18:27 |
| weshay | aye.. updated | 18:27 |
| weshay | doh | 18:27 |
| weshay | fixed | 18:28 |
| *** ykarel|away has quit IRC | 18:28 | |
| *** sanjayu__ has quit IRC | 18:34 | |
| *** Goneri has quit IRC | 18:38 | |
| * rlandy needs recommended podman cheatsheet for the uninitiated? | 18:46 | |
| weshay | rlandy it's not going to work as written :( | 18:46 |
| rlandy | which part? | 18:46 |
| rlandy | your change? | 18:46 |
| weshay | {% if 'periodic' not in zuul.job -%} | 18:46 |
| weshay | {% raw -%} | 18:46 |
| weshay | {% if use_depends_on|default(true)|bool -%} | 18:46 |
| weshay | depends_on: | 18:46 |
| weshay | - {{ zuul.change_url }} | 18:46 |
| weshay | {% endif %} | 18:46 |
| weshay | {% endraw %} | 18:46 |
| weshay | {% endif %} | 18:46 |
| weshay | so w/ the raw above the if statement | 18:46 |
| weshay | the if statement ends up in the rendered template | 18:47 |
| rlandy | weshay: take away the default value | 18:47 |
| rlandy | not needed | 18:47 |
| rlandy | ie: ... | 18:47 |
| weshay | the value doesn't matter | 18:47 |
| rlandy | https://review.opendev.org/#/c/656014/10/roles/create-zuul-based-reproducer/defaults/main.yml | 18:47 |
| weshay | in this case it's a syntax issue | 18:48 |
| rlandy | if it's raw it should not render | 18:48 |
| *** tcw has quit IRC | 18:50 | |
| weshay | rlandy /me gets a cheatsheet | 18:51 |
| *** panda is now known as panda|off | 19:01 | |
| *** dsneddon has quit IRC | 19:01 | |
| *** trown|brb has joined #oooq | 19:05 | |
| *** trown|brb is now known as regain | 19:05 | |
| *** regain is now known as trown | 19:05 | |
| rlandy | trown: hi there | 19:10 |
| rlandy | trown: chandan asked me why openshift devs aren't using zuul - and I didn't know the answer | 19:11 |
| trown | they have prow which is part of upstream kubernetes | 19:11 |
| trown | rlandy: any clue if the issue with rdo-release.rpm is known? I came on IRC to ask on #rdo, but crickets there :P | 19:16 |
| weshay | trown hrm.. for stein? | 19:18 |
| weshay | I don't if anyone is using that, we are not | 19:18 |
| trown | ya just started this afternoon, and it breaks all of our openshift CI :( | 19:18 |
| rlandy | yeah - we don;t rely on it | 19:18 |
| weshay | oh fun | 19:18 |
| rlandy | trown: why do you need that? just lay down the repos | 19:19 |
| trown | is there a more stable way I could get openstackclient? | 19:19 |
| weshay | trown why should openshift ci be any diff than anyone else's | 19:19 |
| *** Goneri has joined #oooq | 19:19 | |
| weshay | ;olz | 19:19 |
| weshay | lolz | 19:19 |
| trown | rlandy: ya I am trying that now, but I have to do it in a dockerfile | 19:19 |
| rlandy | stable and openstackclient in the same sentence ... hmmm | 19:19 |
| trown | it is actually a bit hard to get a file in | 19:19 |
| weshay | rlandy heh.. the client should be stable | 19:20 |
| trown | ya I am thinking of just uploading a binary for it and pointing at that :P | 19:20 |
| weshay | rlandy another use case for molecule ^ test the rdo-release rpm | 19:20 |
| rlandy | well ... | 19:20 |
| weshay | I haven't pulled that in years | 19:20 |
| rlandy | for stein, things should be more stable | 19:21 |
| rlandy | so if you pulled release version | 19:21 |
| weshay | trown from here? https://repos.fedorapeople.org/repos/openstack/ | 19:21 |
| weshay | NA | 19:22 |
| rlandy | haven't tried rdo-release in years | 19:23 |
| trown | k I think I figured out how to make a static repo file... trying https://github.com/trown/installer/commit/60f6fe9d311b9518fb02ac234f4eda78f61a438f | 19:26 |
| trown | is there a different static repo I should use there? | 19:26 |
| *** dsneddon has joined #oooq | 19:26 | |
| trown | it is that mirrorlist URL that is busted, I switched it to just hardcode the baseurl | 19:27 |
| *** tcw has joined #oooq | 19:39 | |
| *** rf0lc0 has joined #oooq | 20:12 | |
| *** hamzy_ has joined #oooq | 20:12 | |
| *** ccamacho has quit IRC | 20:13 | |
| *** tcw has quit IRC | 20:14 | |
| *** yolanda_ has quit IRC | 20:15 | |
| *** ksambor has quit IRC | 20:15 | |
| *** rfolco|out has quit IRC | 20:15 | |
| *** jjoyce has quit IRC | 20:15 | |
| *** hamzy has quit IRC | 20:15 | |
| *** jjoyce has joined #oooq | 20:17 | |
| *** yolanda_ has joined #oooq | 20:28 | |
| *** Goneri has quit IRC | 20:33 | |
| *** trown has quit IRC | 20:40 | |
| *** Goneri has joined #oooq | 20:47 | |
| *** aakarsh has quit IRC | 20:51 | |
| *** Vorrtex has quit IRC | 21:06 | |
| *** Goneri has quit IRC | 21:59 | |
| *** rlandy is now known as rlandy|bbl | 22:17 | |
| *** aakarsh has joined #oooq | 22:46 | |
| *** vinaykns has quit IRC | 22:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!