*** sshnaidm is now known as sshnaidm|afk | 01:20 | |
*** dsneddon has quit IRC | 01:24 | |
*** jmasud has joined #oooq | 01:50 | |
*** ysandeep|away is now known as ysandeep | 01:54 | |
*** jmasud has quit IRC | 01:56 | |
*** dsneddon has joined #oooq | 01:59 | |
*** dsneddon has quit IRC | 02:30 | |
ykarel|away | weshay|ruck, centos-8 containers should be cleared with new nodepool image i think | 02:43 |
---|---|---|
weshay|ruck | ykarel|away, we have other problems again | 02:43 |
ykarel|away | hmm looks like, i see too many reds | 02:44 |
weshay|ruck | ykarel|away, virtenv is failing to install again | 02:44 |
ykarel|away | :( | 02:44 |
weshay|ruck | ya.. tried a bunch of things | 02:44 |
weshay|ruck | no luck so far | 02:44 |
weshay|ruck | can't repro locally either | 02:44 |
ykarel|away | let me check logs | 02:44 |
ykarel|away | u referring setuptools:44.0.0 ERROR: virtualenv not installed? | 02:45 |
ykarel|away | ImportError: cannot import name ensure_text | 02:45 |
weshay|ruck | ya | 02:46 |
weshay|ruck | https://review.opendev.org/#/c/707491/ https://review.opendev.org/#/c/707492/ https://review.opendev.org/#/c/707504/ https://review.opendev.org/#/c/707505/ | 02:47 |
weshay|ruck | ¯\_(ツ)_/¯ | 02:47 |
weshay|ruck | it's also being discussed in #openstack-infra | 02:47 |
ykarel|away | looks like python-six<1.12 is older there, need to check if it's after new images build | 02:47 |
weshay|ruck | https://review.opendev.org/#/c/707499/1 | 02:47 |
weshay|ruck | ya.. I think we need 1.14 | 02:48 |
ykarel|away | >=1.12 should work | 02:49 |
weshay|ruck | ykarel|away, looks like 3rd party jobs are working on some of these patches | 02:49 |
weshay|ruck | def.. related to the upstream image | 02:49 |
weshay|ruck | ykarel|away, i should be testing w/ https://review.rdoproject.org/r/#/c/24916/ | 02:49 |
weshay|ruck | but.. it's also not failing on check-centos-7 | 02:50 |
weshay|ruck | which *should* be the upstream image | 02:50 |
ykarel|away | weshay|ruck, yes last night we had updated images in rdo, since then it's working there | 02:50 |
ykarel|away | check-centos-7 should be similar to upstream one | 02:52 |
weshay|ruck | rdo shows | 02:53 |
weshay|ruck | 2020-02-13 00:10:44.110751 | primary | pip:20.0.2 virtualenv:20.0.2 setuptools:44.0.0 | 02:53 |
weshay|ruck | not sure how that's possible.. w/ https://review.opendev.org/#/c/707505/1/install-deps.sh | 02:54 |
weshay|ruck | ah.. because it already exists | 02:54 |
weshay|ruck | so where is the virtenv default on upstream jobs | 02:54 |
* weshay|ruck looks | 02:54 | |
weshay|ruck | maybe we can just set OPT_WORKDIR to that | 02:54 |
weshay|ruck | ? | 02:55 |
weshay|ruck | https://opendev.org/opendev/base-jobs/src/branch/master/playbooks/base/pre.yaml#L7-L20 | 02:56 |
ykarel|away | so 20.0.2 is good one | 02:56 |
ykarel|away | so if other upstream jobs non tripleo jobs are working, good to remove that hack in quickstart | 02:56 |
weshay|ruck | ya.. but it failed | 02:57 |
ykarel|away | also i see 20.0.3 virtualenv released https://pypi.org/project/virtualenv/#history | 02:57 |
weshay|ruck | also pinning to 20.0.2 failed | 02:57 |
weshay|ruck | er.. 3 | 02:57 |
* weshay|ruck gets | 02:57 | |
ykarel|away | need to check what .3 fixed | 02:57 |
weshay|ruck | https://review.opendev.org/#/c/707492/1/install-deps.sh | 02:57 |
weshay|ruck | ykarel|away, didn't see anything related in the change log | 02:57 |
ykarel|away | https://github.com/pypa/virtualenv/compare/20.0.2...20.0.3 | 02:59 |
ykarel|away | can u try complete removal of thah hack, install/uninstall | 03:00 |
ykarel|away | else we need to try update of six there | 03:01 |
weshay|ruck | ykarel|away, https://review.opendev.org/#/c/707504/ | 03:01 |
weshay|ruck | anything in that change log strike you as a fix? | 03:01 |
ykarel|away | looking | 03:02 |
weshay|ruck | https://github.com/pypa/virtualenv/commit/80fe52f4a26d59e7b855afbaebe7b7ab1088fb4b | 03:02 |
weshay|ruck | maybe | 03:02 |
weshay|ruck | ykarel|away, if we can fix the images in rdo.. why aren't we fixing them upstream? | 03:04 |
ykarel|away | weshay|ruck, upstream fixes will be going on in parallel i think, but there it takes time, as pushing to multiple clouds | 03:05 |
weshay|ruck | ykarel|away, is there a review #? | 03:06 |
ykarel|away | weshay|ruck, for centos-8 https://review.opendev.org/#/c/707346/ fixed for us | 03:06 |
ykarel|away | images were build after ^^ | 03:07 |
ykarel|away | let me check if i can reproduce locally | 03:08 |
*** apetrich has quit IRC | 03:13 | |
ykarel|away | weshay|ruck, i can reproduce that issue with 20.0.3 locally | 03:15 |
ykarel|away | retrying to confirm | 03:16 |
weshay|ruck | can you paste? | 03:16 |
* weshay|ruck needs to learn | 03:16 | |
weshay|ruck | I think everything is currently failing "TASK [configure-swap : Copy old /opt]" | 03:17 |
weshay|ruck | 2020-02-13 03:16:50.057885 | secondary | included: /var/lib/zuul/builds/d08994440dfb4a4f9cb9f0d057f34160/untrusted/project_3/opendev.org/openstack/openstack-zuul-jobs/roles/configure-swap/tasks/ephemeral.yaml | 03:17 |
ykarel|away | python -c "from __future__ import print_function; import virtualenv; print('$1:%s ' % virtualenv.__version__, end='')" | 03:17 |
ykarel|away | ^^ is used in quickstart | 03:17 |
ykarel|away | i used same locally | 03:17 |
ykarel|away | with virtualenv 20.0.3 and six 1.11.0 | 03:18 |
ykarel|away | and get same traceback | 03:18 |
weshay|ruck | in here? https://opendev.org/openstack/tripleo-quickstart/src/branch/master/install-deps.sh | 03:19 |
weshay|ruck | https://opendev.org/openstack/tripleo-quickstart/src/branch/master/install-deps.sh#L86 | 03:20 |
weshay|ruck | see it | 03:20 |
ykarel|away | yes | 03:20 |
ykarel|away | weshay|ruck, ok so on rdo side centos-8 six is handled with https://softwarefactory-project.io/r/#/c/17510/1/nodepool/elements/fix-virtualenv/install.d/99-fix-virtualenv | 03:20 |
weshay|ruck | python -c "from __future__ import print_function; import virtualenv; print('$1:%s ' % virtualenv.__version__, end='')" | 03:21 |
weshay|ruck | :20.0.3 (wes) | 03:21 |
weshay|ruck | meh | 03:21 |
ykarel|away | weshay check six in your env | 03:22 |
ykarel|away | you would have >=1.12 | 03:22 |
ykarel|away | only reproduceable with lesser version | 03:22 |
weshay|ruck | six-1.14.0.dist-info/ | 03:22 |
ykarel|away | yup that's why u didn't hit it | 03:22 |
weshay|ruck | k | 03:22 |
weshay|ruck | uninstalled, install 1.12.0 | 03:23 |
weshay|ruck | worked | 03:23 |
ykarel|away | install 1.11 | 03:24 |
ykarel|away | there the issue | 03:24 |
ykarel|away | 1.12.0 is fine | 03:24 |
weshay|ruck | k | 03:24 |
weshay|ruck | ERROR: virtualenv 20.0.3 has requirement six<2,>=1.12.0, but you'll have six 1.11.0 which is incompatible. | 03:24 |
*** ykarel|away is now known as ykarel | 03:25 | |
weshay|ruck | yup.. that did it :) | 03:25 |
ykarel | u got traceback? | 03:25 |
weshay|ruck | yes | 03:25 |
ykarel | okk | 03:25 |
weshay|ruck | so do we just need to upgrade six? | 03:25 |
weshay|ruck | in install-deps? | 03:25 |
ykarel | yes that will workaround the issue | 03:26 |
ykarel | but not sure about the actual fix | 03:26 |
* ykarel checks image build logs | 03:27 | |
ykarel | https://nb01.openstack.org/centos-7-0000121697.log | 03:29 |
ykarel | 20.0.3 is installed there | 03:29 |
*** rlandy|bbl is now known as rlandy | 03:29 | |
ykarel | so it started happening after new images | 03:30 |
weshay|ruck | we could add python-six to the bindep | 03:32 |
ykarel | weshay|ruck, that should not help | 03:32 |
weshay|ruck | no --site-packages | 03:32 |
ykarel | as distro's don't have updated package | 03:32 |
weshay|ruck | I gues | 03:32 |
weshay|ruck | ah | 03:32 |
weshay|ruck | ykarel, still busted https://review.opendev.org/#/c/707505/3/install-deps.sh | 03:33 |
ykarel | okk rdo images don't have issues | 03:33 |
weshay|ruck | http://zuul.openstack.org/stream/f66befa7ef834c13a0e78ee745fcd9c0?logfile=console.log | 03:33 |
ykarel | https://softwarefactory-project.io/nodepool-log/upstream-centos-7-0000095413.log | 03:33 |
weshay|ruck | 2020-02-13 03:32:40.797876 | primary | pip:20.0.2 Traceback (most recent call last): | 03:34 |
weshay|ruck | I think we're patching in the wrong place | 03:35 |
ykarel | checking logs | 03:35 |
ykarel | hmm that part is not reached | 03:35 |
ykarel | error happens earlier | 03:35 |
ykarel | at check_python thing | 03:35 |
weshay|ruck | think ur right | 03:36 |
weshay|ruck | it's happening on line 86 | 03:36 |
weshay|ruck | https://opendev.org/openstack/tripleo-quickstart/blame/branch/master/install-deps.sh#L86 | 03:36 |
weshay|ruck | in the check | 03:36 |
ykarel | yes so need to do these hacks at that line | 03:36 |
ykarel | just upgrade six should work | 03:36 |
weshay|ruck | ya | 03:36 |
ykarel | new patch looks fine | 03:41 |
ykarel | https://review.opendev.org/#/c/707525/1/install-deps.sh | 03:41 |
weshay|ruck | that's good to hear :) | 03:41 |
weshay|ruck | install-deps sucks | 03:41 |
weshay|ruck | it's a little better now that it's all in one script | 03:41 |
weshay|ruck | but ugh | 03:41 |
weshay|ruck | sshnaidm|afk, had a good idea to build a container w/ the deps and run from that | 03:42 |
ykarel | on the fly container build, or build container in promotion and use that in all jobs? | 03:44 |
ykarel | ok got it | 03:45 |
weshay|ruck | 2020-02-13 03:45:27.359743 | primary | Looking in indexes: http://mirror.regionone.limestone.openstack.org/pypi/simple, http://mirror.regionone.limestone.openstack.org/wheel/centos-7.7-x86_64 | 03:46 |
weshay|ruck | 2020-02-13 03:45:27.422251 | primary | Collecting six>=1.14.0 | 03:46 |
weshay|ruck | 2020-02-13 03:45:27.426208 | primary | Downloading http://mirror.regionone.limestone.openstack.org/pypifiles/packages/65/eb/1f97cb97bfc2390a276969c6fae16075da282f5058082d4cb10c6c5c1dba/six-1.14.0-py2.py3-none-any.whl (10 kB) | 03:46 |
weshay|ruck | ykarel++ | 03:46 |
weshay|ruck | ykarel++ | 03:46 |
weshay|ruck | ykarel, hitting retry limits. | 03:47 |
weshay|ruck | this is not going to merge quickly | 03:47 |
weshay|ruck | thanks rlandy | 03:50 |
rlandy | weshay|ruck: sure - least I could do | 03:50 |
rlandy | see you are still battling this | 03:50 |
ykarel | mmm some jobs are passing | 03:54 |
ykarel | and some i see RETRY_LIMIT | 03:55 |
ykarel | do you have the logs for retry_limit thing | 03:55 |
weshay|ruck | sshnaidm|afk, ^ zbr was right :) | 03:56 |
weshay|ruck | re: images | 03:56 |
weshay|ruck | ykarel, you don't get logs on retry | 03:56 |
ykarel | weshay|ruck, i though u had console open | 03:56 |
weshay|ruck | ykarel, probably a bad image | 03:56 |
weshay|ruck | ykarel, posted to openstack-infra channel | 03:56 |
ykarel | ohkk possible | 03:57 |
ykarel | once infra is clear, abandon/restore can be done to restart the jobs | 03:59 |
*** ykarel is now known as ykarel|afk | 03:59 | |
weshay|ruck | aye | 03:59 |
weshay|ruck | thanks ykarel|afk | 03:59 |
*** rlandy has quit IRC | 04:03 | |
*** udesale has joined #oooq | 04:07 | |
*** ykarel|afk is now known as ykarel | 04:23 | |
*** raukadah is now known as chandankumar | 04:59 | |
*** skramaja has joined #oooq | 04:59 | |
ykarel | ohhk so issue hit both upstream and rdo softwarefactory | 05:13 |
ykarel | i see some third party jobs failing too | 05:13 |
*** jmasud has joined #oooq | 05:37 | |
*** holser has joined #oooq | 05:41 | |
*** holser has quit IRC | 06:00 | |
*** marios has joined #oooq | 06:23 | |
*** marios is now known as marios|ruck | 06:23 | |
*** dsneddon has joined #oooq | 06:27 | |
marios|ruck | chandankumar: o/ you free to jump on phone in ~ 1 hour? let me do some reviews etc first | 06:31 |
*** dsneddon has quit IRC | 06:32 | |
*** ccamacho has quit IRC | 06:39 | |
chandankumar | ykarel, aye | 06:49 |
ykarel | chandankumar, hi | 06:50 |
*** jfrancoa has joined #oooq | 06:50 | |
chandankumar | ykarel, Hello | 06:50 |
chandankumar | ykarel, apart from this https://review.opendev.org/#/c/707525, does any more patches needed to clear the gates? | 06:52 |
ykarel | chandankumar, yes | 06:53 |
chandankumar | ykarel, my rhel8 patch and? | 06:53 |
ykarel | chandankumar, https://review.opendev.org/#/c/707330/ and your rhel8 fix patch | 06:53 |
chandankumar | ykarel, ok thanks! | 06:54 |
ykarel | with ^^ i think we need we need to adjust some of rdo jobs, but that's doable | 06:54 |
ykarel | to add ansible-role-collect logs in job roles: | 06:55 |
ykarel | sorry ansible-role-container-registry | 06:55 |
marios|ruck | 08:31 < marios|ruck> chandankumar: o/ you free to jump on phone in ~ 1 hour? let me do some reviews etc first | 07:00 |
chandankumar | marios|ruck, yes, how about in another 30 mins? | 07:02 |
chandankumar | marios|ruck, https://review.rdoproject.org/r/24924 | 07:03 |
marios|ruck | chandankumar: sounds good and ack will check in a bit | 07:03 |
*** saneax has joined #oooq | 07:20 | |
*** holser has joined #oooq | 07:23 | |
*** ykarel is now known as ykarel|lunch | 07:31 | |
marios|ruck | panda: sshnaidm: zbr: arxcruz: chandankumar: please review when you have time thanks https://review.rdoproject.org/r/#/c/24807/ | 07:35 |
*** jpena|off is now known as jpena | 07:51 | |
*** ccamacho has joined #oooq | 08:13 | |
arxcruz | marios|ruck: from ansible/python perspective, looks ok, only what sshnaidm|afk pointed, but if he's not blocking, i won't | 08:18 |
marios|ruck | arxcruz: ack thanks i will check sshnaidm|afk comments more carefully in a bit might update thank you for looking | 08:19 |
*** tosky has joined #oooq | 08:19 | |
*** tesseract has joined #oooq | 08:20 | |
*** kopecmartin has quit IRC | 08:21 | |
dpawlik | chandankumar, marios|ruck: maybe I miss the answer, but its really important: how old tripleO image (dirs with overcloud-full.tar) should I keep? May I set retention_days to 60 days ? | 08:22 |
dpawlik | currently you have images older than 150days | 08:23 |
dpawlik | server volume is almost full and I need to know, if I should add new volume or not | 08:23 |
*** ykarel|lunch is now known as ykarel | 08:28 | |
*** dsneddon has joined #oooq | 08:28 | |
marios|ruck | dpawlik: which images are you talking about sorry i am out of context. chandankumar where are the notes you showed me on the call about this (rcm move & images server i don't see them on the etherpad) | 08:29 |
marios|ruck | dpawlik: chandankumar: are these the ones from fs2-upload job? | 08:29 |
*** amoralej|off is now known as amoralej | 08:29 | |
marios|ruck | chandankumar: if its the ones from fs2 upload I can't think of any reason to keep them more than 60 days maybe even less. | 08:30 |
dpawlik | marios|ruck, dunno if its from fs2-upload job (I'm new), but | 08:30 |
dpawlik | it is those images that are sent to rcm-share host (38.145.34.141) | 08:30 |
*** dsneddon has quit IRC | 08:33 | |
marios|ruck | dpawlik: so not sure. afaik we use rcm-guest for rhel8 stuffs/images. even then though i can't think of a job that would want to use an image older than 60 days... | 08:34 |
marios|ruck | dpawlik: i would set retention as you suggest and send note to pcci | 08:34 |
dpawlik | marios|ruck, cool. THanks | 08:34 |
*** kopecmartin has joined #oooq | 08:45 | |
ykarel | marios|ruck, do you know the reasoning behind recent changes at https://review.opendev.org/#/c/706572/? | 08:52 |
ykarel | mainly the inclusion of installation u-c part in this change | 08:52 |
marios|ruck | ykarel: looks like weshay|ruck rebased ontop of your patch if that is what you mean? | 08:53 |
marios|ruck | ykarel: and added the 7/8 vs yum/dnf | 08:53 |
ykarel | marios|ruck, but my patch didn't have 7/8 thing | 08:53 |
ykarel | yes, ok and andy reasoning for that change? | 08:53 |
marios|ruck | ykarel: 10:53 < ykarel> marios|ruck, but my patch didn't have 7/8 thing | 08:54 |
marios|ruck | ykarel: yes but this is different patch for the centos8 check job | 08:54 |
marios|ruck | ykarel: the reasoning, i guess it broke for centos or rhel8 | 08:54 |
marios|ruck | ykarel: we had discussed it recently let me find the review sec | 08:54 |
ykarel | marios|ruck, ack don't think that change breaked it | 08:55 |
ykarel | it was virtualenv thing that messing | 08:55 |
marios|ruck | ykarel: no i don't think its related to your patch | 08:55 |
marios|ruck | ykarel: something else | 08:55 |
ykarel | ok let me know if u find something, else i think i will revise it to move forward | 08:55 |
marios|ruck | ykarel: e.g weshay|ruck also posted that one * https://review.opendev.org/#/c/707469/1 | 08:55 |
marios|ruck | ykarel: revise what | 08:55 |
ykarel | marios|ruck, revise https://review.opendev.org/#/c/706572/ | 08:56 |
ykarel | such that it goes green | 08:56 |
marios|ruck | ykarel: is that blocking you? | 08:57 |
ykarel | marios|ruck, other centos8 work relies on this one | 08:57 |
ykarel | this is the base job, which is stuck currently | 08:57 |
marios|ruck | ykarel: this is the check job | 08:58 |
ykarel | if u planning to work on it, u can go for the fixes | 08:58 |
marios|ruck | ykarel: what is blocked on that today though | 08:58 |
ykarel | marios|ruck, so parallel work going on promotion jobs? | 08:58 |
marios|ruck | ykarel: we merged the promotion definition a while ago | 08:59 |
marios|ruck | ykarel: but i don't think wired up yet | 08:59 |
marios|ruck | ykarel: there are other things coming still | 08:59 |
ykarel | yup it's not running | 08:59 |
marios|ruck | ykarel: like reworking get-hash | 08:59 |
ykarel | main idea is check jobs keep running and start catching issues upstream | 08:59 |
marios|ruck | ykarel: sure, i will update that in a bit | 09:00 |
ykarel | marios|ruck, ack Thanks | 09:00 |
*** dsneddon has joined #oooq | 09:08 | |
*** dsneddon has quit IRC | 09:13 | |
*** marios|ruck has quit IRC | 09:16 | |
*** marios has joined #oooq | 09:19 | |
*** marios is now known as marios|ruck | 09:19 | |
*** chem has joined #oooq | 09:20 | |
dpawlik | marios|ruck, could you tell me what is an email for pcci ? | 09:20 |
dpawlik | please | 09:20 |
marios|ruck | dpawlik: sure sec | 09:20 |
marios|ruck | dpawlik: see pvt | 09:21 |
chandankumar | dpawlik, we cleaned up yesterday na? | 09:21 |
*** bogdando has joined #oooq | 09:21 | |
chandankumar | dpawlik, we can remove fs02 images | 09:22 |
chandankumar | for rhel-8, once we had promotion, we can keep last 5 days of images before promotion date | 09:22 |
dpawlik | marios|ruck, thx | 09:23 |
dpawlik | chandankumar, I don't see more free space there | 09:24 |
dpawlik | chandankumar, /dev/vdb1 1.0T 628G 396G 62% /var/www/rcm-guest | 09:24 |
dpawlik | its 20GB more thany yesterday | 09:24 |
dpawlik | lets say, for me 90 days is enough to cleanup, it will remove over 250GB | 09:25 |
chandankumar | dpawlik, I reduced it to 20 days | 09:25 |
dpawlik | k, cleaning... | 09:26 |
chandankumar | sshnaidm|afk, https://review.rdoproject.org/r/#/c/24924/ needs +w | 09:26 |
chandankumar | dpawlik, can you take a look at this error https://logserver.rdoproject.org/25/24925/1/check/tripleo-ci-rhel-8-ovb-3ctlr_1comp-featureset001/53066c6/job-output.txt | 09:27 |
chandankumar | dpawlik, http://images-vexxhost.rdoproject.org:4433/rcm-guest/images/redhat8/master/rdo_trunk/bcbb1aed4b669a90d4717b10a0b85e6093d42270_f990613b/overcloud-full.tar.md5 image does not exists | 09:28 |
chandankumar | dpawlik, I was testing pull here https://review.rdoproject.org/r/#/c/24925/ | 09:28 |
*** derekh has joined #oooq | 09:29 | |
dpawlik | chandankumar, them, it can be problem with http | 09:31 |
dpawlik | chandankumar, priv | 09:31 |
dpawlik | marios|ruck, chandankumar: please check and vote in free time https://review.rdoproject.org/r/#/c/24926/. If both agree to set 20 days, I will change. later I will write an email | 09:36 |
chandankumar | dpawlik, does the default port for ssh rsyncing image is 22 na? or is it something else? | 09:41 |
chandankumar | dpawlik, https://review.rdoproject.org/r/24915 please have a look | 09:41 |
chandankumar | for image server | 09:41 |
*** apetrich has joined #oooq | 09:43 | |
marios|ruck | ack dpawlik added to next reviews list might be later/tomorrow | 09:44 |
dpawlik | chandankumar, port 22 is correct for sync | 09:46 |
dpawlik | marios|ruck, eeee, but it really depends if I need to attach new volume or not xD | 09:47 |
dpawlik | chandankumar, let me know when tests finished | 09:56 |
dpawlik | chandankumar, I will try to create rewirte rule for redirection | 09:56 |
chandankumar | dpawlik, yes watching here https://review.rdoproject.org/zuul/stream/51ef66054b3d47c7af77c3ab72b58002?logfile=console.log | 09:57 |
chandankumar | fetch images worked with https | 09:58 |
dpawlik | cool | 09:58 |
chandankumar | dpawlik, once it finishes, we will test the rhel8 push mechanism then we can merge the patches anytime | 09:58 |
dpawlik | chandankumar, so now it will not require to download something from the images-vexxhost server or? | 10:03 |
marios|ruck | dpawlik: ack | 10:03 |
marios|ruck | dpawlik: main concern is the deletion needs to be restricted a bit more please? | 10:04 |
dpawlik | marios|ruck, ok | 10:04 |
*** dtantsur|afk is now known as dtantsur | 10:05 | |
zbr | it seems that snap does a good job bringing pypy on centos :) https://snapcraft.io/install/pypy/centos -- tested succesully. | 10:06 |
zbr | raised https://github.com/pypa/virtualenv/issues/1605 | 10:37 |
*** udesale has quit IRC | 11:02 | |
*** dsneddon has joined #oooq | 11:09 | |
*** dsneddon has quit IRC | 11:15 | |
chandankumar | sshnaidm|afk, can we merge this https://review.rdoproject.org/r/#/c/24915/ so that I can push test patch of pushing the image | 11:18 |
chandankumar | ? | 11:18 |
*** dtantsur is now known as dtantsur|brb | 11:22 | |
sshnaidm|afk | chandankumar, USE_IMAGE_SERVER="${USE_IMAGE_SERVER}" ?? | 11:22 |
sshnaidm|afk | chandankumar, did you mean to set default there? | 11:22 |
sshnaidm|afk | USE_IMAGE_SERVER="${USE_IMAGE_SERVER:-false}" ? | 11:23 |
*** sshnaidm|afk is now known as sshnaidm|brb | 11:25 | |
chandankumar | yes | 11:25 |
chandankumar | sshnaidm|brb, but that value should be setted by ansible var | 11:26 |
chandankumar | not via shell var | 11:26 |
*** hamzy_ is now known as hamzy | 11:26 | |
chandankumar | better I will rename it use_vexxhost_server | 11:26 |
*** dsneddon has joined #oooq | 11:30 | |
zbr | i am now working with tox/virtualenv maintainer to avoid current issues we have, if everything goes fine we should have new releases that work on our outdates systems. | 11:31 |
*** marios|ruck is now known as marios|ruck|call | 11:32 | |
weshay|ruck | marios|ruck|call, https://github.com/weshayutin/tripleo-critical-bugs | 11:44 |
chandankumar | weshay|ruck, https://softwarefactory-project.io/etherpad/p/images_migration | 11:53 |
marios|ruck|call | https://review.rdoproject.org/r/#/c/24926/ | 11:54 |
chandankumar | weshay|ruck, marios|ruck|call https://softwarefactory-project.io/etherpad/p/images_migration | 11:55 |
marios|ruck|call | thanks chandankumar | 11:55 |
marios|ruck|call | weshay|ruck: 13:54 < marios|ruck|call> https://review.rdoproject.org/r/#/c/24926/ | 11:55 |
*** jpena is now known as jpena|lunch | 11:56 | |
*** sshnaidm|brb is now known as sshnaidm | 11:56 | |
sshnaidm | chandankumar, it's not about the name | 11:57 |
sshnaidm | chandankumar, you assign variable to itself | 11:58 |
*** amoralej is now known as amoralej|lunch | 12:00 | |
weshay|ruck | pojadhav, hey | 12:01 |
*** marios|ruck|call is now known as marios|ruck | 12:01 | |
chandankumar | sshnaidm, ok, then updated with false default value | 12:03 |
pojadhav | weshay|ruck, hello | 12:04 |
weshay|ruck | zbr, can we establish a job like tihs one.. https://review.rdoproject.org/r/#/c/24916/ | 12:04 |
weshay|ruck | that runs through the install-deps? | 12:04 |
dpawlik | marios|ruck, https://review.rdoproject.org/r/#/c/24926/ => added regex | 12:05 |
dpawlik | cc weshay|ruck ^^ | 12:05 |
marios|ruck | dpawlik: ack will check | 12:08 |
zbr | weshay|ruck: yes but we already test with containers, is a bit of resource hog, but obviously that easy to do. | 12:10 |
chandankumar | weshay|ruck, dpawlik I will back in 1 and half hour | 12:11 |
zbr | for the moment I am trying to add minimal documentation regarding rdo nodesets, see https://review.rdoproject.org/r/#/c/24919/ | 12:11 |
dpawlik | okey chandankumar | 12:11 |
zbr | there are too many of them without clear purpose documented, we cannot add jobs testing them unless their purpose is documented. | 12:12 |
zbr | weshay|ruck: help me merge the "seeds" of documenting nodes, and I will add the install-deps.sh jobs. | 12:12 |
zbr | i need to know which nodes we are supposed to use/test, that is why.... | 12:13 |
weshay|ruck | bhagyashris, ping.. 1-1 | 12:17 |
bhagyashris | weshay|ruck, joining... | 12:19 |
*** ykarel is now known as ykarel|afk | 12:19 | |
*** derekh has quit IRC | 12:23 | |
*** ccamacho has quit IRC | 12:25 | |
*** rfolco has joined #oooq | 12:25 | |
marios|ruck | sshnaidm: arxcruz: thanks for your votes. sshnaidm please see my comments, you already +2 but I replied to all your concerns hope that is OK. Only thing i updated in v23 is explained there https://review.rdoproject.org/r/#/c/24807/23/ci-scripts/container-push/roles/containers-promote/tasks/check_set_ppc_manifests.yml@27 | 12:28 |
marios|ruck | panda: please recheck that one https://review.rdoproject.org/r/#/c/24807/ when you next have some time thank you | 12:29 |
weshay|ruck | ysandeep, ping when you have a sec.. openstack-cert | 12:29 |
ysandeep | weshay|ruck, pong , I am free to talk now | 12:29 |
sshnaidm | marios|ruck, I still don't understand why 2 same tasks run: https://review.rdoproject.org/r/#/c/24807/21/ci-scripts/infra-setup/roles/promoter/molecule/container-push/converge.yml@169 | 12:32 |
marios|ruck | sshnaidm: i replied there... they have different conditional | 12:32 |
marios|ruck | sshnaidm: but you already +2 so please don't block for that :D | 12:32 |
sshnaidm | marios|ruck, yeah, but it doesn't explain why do you run 2 same tasks :) | 12:32 |
sshnaidm | marios|ruck, either to combine all conditionals, or remove them | 12:33 |
*** dsneddon has quit IRC | 12:33 | |
sshnaidm | marios|ruck, this is confusing tbh for understanding the workflow | 12:33 |
marios|ruck | sshnaidm: please make a clear request on the review. for clarity you are referring to the molecule converge https://review.rdoproject.org/r/#/c/24807/23/ci-scripts/infra-setup/roles/promoter/molecule/container-push/converge.yml right? | 12:34 |
*** ysandeep is now known as ysandeep|brb | 12:34 | |
sshnaidm | marios|ruck, task "Check that the containers are pushed correctly to all target registries" | 12:34 |
weshay|ruck | ysandeep|brb, ah.. missed ya.. let's chat tomorrow | 12:35 |
marios|ruck | sshnaidm: my main problem is i don't know how to do what you want, exactly. the first one deals with the case of x86 only containers | 12:35 |
marios|ruck | sshnaidm: the second one deals with x86 and ppc | 12:36 |
marios|ruck | sshnaidm: if i remove those the molecul scenario fails | 12:36 |
marios|ruck | sshnaidm: note this is molecule converge not the main code | 12:36 |
sshnaidm | marios|ruck, maybe "when: (not ppc_manifests and '"ppc" not in image') or ppc_manifests" ? | 12:36 |
marios|ruck | sshnaidm: thanks for checking anyway please comment on there if you feel strongly enough to block on it. i will change it but grateful for suggestions | 12:36 |
*** ysandeep|brb is now known as ysandeep | 12:37 | |
sshnaidm | marios|ruck, I don't understand why you run exactly same tasks under different conditions, maybe this is more clear | 12:37 |
ysandeep | weshay|ruck, okay | 12:37 |
sshnaidm | marios|ruck, or these tasks are different? I looked at them for a few minutes trying to find a difference :) | 12:37 |
marios|ruck | sshnaidm: they are different in the list of containers they check for | 12:38 |
marios|ruck | sshnaidm: and that is expressed in the when: | 12:38 |
rfolco | zbr, quick question... how can I modify tox.ini to run non-delegated test ? https://review.rdoproject.org/r/#/c/24762/22/tox.ini@82 | 12:38 |
marios|ruck | sshnaidm: one is with ppc enabled... the other is ppc disabled so check only x86 containers | 12:38 |
marios|ruck | sshnaidm: we want only one of them to run based on ppc_manifests | 12:38 |
sshnaidm | marios|ruck, is my condition above solving it? | 12:39 |
sshnaidm | I see the same in both tasks: loop: "{{ target_registries | map(attribute='host') | product(containers) | map('join', '/') | list }}" | 12:39 |
zbr | rfolco: sure, in fact that tox.ini file has a bug, the filters should be part of {posargs:HERE} not outside. | 12:40 |
zbr | in order to allow you to run molecule with whatever args you want. | 12:40 |
weshay|ruck | amoralej|lunch, https://review.opendev.org/#/q/owner:%22Dan+Sneddon+%253Cdan%2540dansneddon.com%253E%22 | 12:41 |
zbr | rfolco: do you want me to update the review to fix that? | 12:41 |
weshay|ruck | amoralej|lunch, sorry | 12:41 |
weshay|ruck | wrong guy | 12:41 |
weshay|ruck | akahat, https://review.opendev.org/#/q/owner:%22Dan+Sneddon+%253Cdan%2540dansneddon.com%253E%22 | 12:41 |
rfolco | zbr, would you update the review please ? | 12:41 |
sshnaidm | marios|ruck, nvrmind, I don't block, but I'd be glad to have better explanation or design of this later, it's more logical issue then technical for me | 12:41 |
rfolco | zbr, I am spinning wheels | 12:41 |
marios|ruck | sshnaidm: honestly thanks for looking at that, i am trying to understand if your conditional covers all cases | 12:42 |
marios|ruck | sshnaidm: but then i don't see the value of adding somethign which is really not clear/quick to understand what is happening, to save like 5 lines task, which in fact, it is in the molecule scenario not even the main code | 12:42 |
sshnaidm | marios|ruck, it's just a combination of these two conditions | 12:42 |
sshnaidm | marios|ruck, well, what is really not clear is 2 same tasks :) | 12:43 |
marios|ruck | sshnaidm: i appreciate that you take so much effort to review the code and thanks for the discussion .. imo i'd rather have it more explicit than use a conditional like when: (not ppc_manifests and '"ppc" not in image') or ppc_manifests | 12:44 |
marios|ruck | sshnaidm: makes it easier for stupid people like me :) | 12:44 |
marios|ruck | sshnaidm: thanks for not blocking on it | 12:44 |
zbr | rfolco: done | 12:48 |
marios|ruck | rfolco: i replied there fyi thanks for looking https://review.opendev.org/#/c/701937/16/roles/build-containers/templates/build.sh.j2 | 12:50 |
zbr | tox 3.14.4 was just relases few minutes ago, with some fixes from me, like forcing install of six>1.14 | 12:51 |
rfolco | zbr, thank you very much :) | 12:52 |
*** ykarel|afk is now known as ykarel | 12:56 | |
*** skramaja has quit IRC | 12:56 | |
rfolco | planning meeting starting at https://meet.google.com/tbr-rgjz-cjc?authuser=1 | 13:00 |
rfolco | weshay|ruck, chandankumar zbr marios|ruck sshnaidm arxcruz panda ^ | 13:00 |
*** rlandy has joined #oooq | 13:00 | |
rfolco | agenda https://hackmd.io/IhMCTNMBSF6xtqiEd9Z0Kw?both | 13:02 |
weshay|ruck | chandankumar, need you in planning sir | 13:03 |
*** marios|ruck is now known as marios|ruck|call | 13:06 | |
sshnaidm | rfolco, sprint till Wed 2020-04-04 ? | 13:17 |
*** rascasoft has joined #oooq | 13:18 | |
chandankumar | weshay|ruck, I am there | 13:18 |
*** ykarel is now known as ykarel|afk | 13:18 | |
*** ccamacho has joined #oooq | 13:19 | |
weshay|ruck | chandankumar, thanks | 13:19 |
*** jpena|lunch is now known as jpena | 13:29 | |
*** amoralej|lunch is now known as amoralej | 13:32 | |
rlandy | chandankumar: if your still around after the meeting, I can sync with you re: rhel 8.1 stuff | 13:34 |
chandankumar | rlandy, may be monday if works for you? | 13:34 |
rlandy | chandankumar: ack | 13:34 |
*** Goneri has quit IRC | 13:40 | |
*** dsneddon has joined #oooq | 13:41 | |
chandankumar | rlandy, please merge this https://review.rdoproject.org/r/#/c/24915/ | 13:44 |
chandankumar | dpawlik, sshnaidm master review to test movement to new image server for rhel8 https://review.rdoproject.org/r/24929 | 13:56 |
chandankumar | if that works, we are good to go, merge the patches | 13:57 |
chandankumar | it will test both pull and push | 13:57 |
dpawlik | chandankumar, ack | 13:58 |
marios|ruck|call | ykarel|afk: zbr: please see https://review.opendev.org/#/c/706572/21/roles/build-containers/tasks/venv_setup.yml@1 when you have a sec, not sure how to workaround for six there | 13:59 |
marios|ruck|call | zbr: ykarel|afk: is that sane https://review.opendev.org/#/c/706572/22/roles/build-containers/tasks/venv_setup.yml :D | 14:06 |
zbr | wait for new release of virtualenv today, i am working on it | 14:17 |
*** dtantsur|brb is now known as dtantsur | 14:24 | |
panda | chandankumar: I had to mute you. | 14:25 |
chandankumar | panda, sorry | 14:25 |
chandankumar | panda, created the user story https://tree.taiga.io/project/tripleo-ci-board/us/1535 for rdo tox molecule | 14:26 |
chandankumar | panda, added additional us for extended promoter work https://tree.taiga.io/project/tripleo-ci-board/us/1536?milestone=252270 | 14:28 |
*** udesale has joined #oooq | 14:29 | |
*** Goneri has joined #oooq | 14:31 | |
chandankumar | dpawlik, is it ok to move to new image server tomorrow, I am still testing it | 14:41 |
chandankumar | ? | 14:41 |
chandankumar | or better monday | 14:41 |
dpawlik | Evil is on friday deployment xD | 14:43 |
dpawlik | evil comes* on friday deployment | 14:43 |
dpawlik | so lets say on monday it will be better | 14:43 |
chandankumar | dpawlik, works for me | 14:45 |
chandankumar | weshay|ruck, marios|ruck|call ^^ we will be doing final image server migration on Monday, we need more testing, FYI! | 14:46 |
dpawlik | cc jpena ^^ | 14:47 |
marios|ruck|call | ack chandankumar | 14:47 |
*** TrevorV has joined #oooq | 14:55 | |
*** dtantsur is now known as dtantsur|afk | 14:58 | |
chandankumar | rfolco, weshay|ruck, need to leave now | 14:59 |
chandankumar | see ya | 14:59 |
*** chandankumar is now known as raukadah | 14:59 | |
ykarel|afk | marios|ruck|call, you need to do before - name: Install deps from upper-constraints task | 15:02 |
ykarel|afk | anywayhere in the job should work | 15:02 |
ykarel|afk | zbr, new release would work with older six? | 15:02 |
ykarel|afk | new release of virtualenv | 15:02 |
*** ykarel|afk is now known as ykarel | 15:02 | |
zbr | ykarel|afk: yep, this is what i was doing. | 15:03 |
ykarel | zbr, okk if it works with older six, then it would be good | 15:03 |
weshay|ruck | https://projects.engineering.redhat.com/browse/OSPPRODCH-17 | 15:03 |
zbr | ykarel: see https://github.com/pypa/virtualenv/pull/1606 | 15:04 |
* ykarel looks | 15:04 | |
ykarel | zbr, ack from idea wise lgtm | 15:05 |
*** marios|ruck|call is now known as marios|ruck | 15:05 | |
ykarel | with this hard dep on six >=1.12 is removed | 15:05 |
marios|ruck | ykarel: thanks will do | 15:06 |
marios|ruck | ykarel: gimme few mins just came of long call | 15:06 |
ykarel | marios|ruck, ack take ur time | 15:06 |
marios|ruck | ykarel: was trying to do explicit install but failed | 15:06 |
zbr | ykarel: yep. tox is already released, now only virtualenv remaining. | 15:06 |
ykarel | marios|ruck, explicit mean? | 15:06 |
marios|ruck | ykarel: https://review.opendev.org/#/c/706572/23/roles/build-containers/tasks/venv_setup.yml | 15:06 |
ykarel | hmm permission denied | 15:07 |
ykarel | use become, as need to override system six | 15:07 |
zbr | ykarel: also read the last private comment on https://bugzilla.redhat.com/show_bug.cgi?id=1790174 | 15:07 |
openstack | bugzilla.redhat.com bug 1790174 in python-virtualenv "virtualenv activate chokes when run from bash scripts in strict mode due to missing PS1" [Medium,New] - Assigned to python-maint | 15:07 |
zbr | this time I was able to make a patch to virtualenv, but I doubt we can continue like this. | 15:08 |
* rlandy will brb - need to buy some cough medicine | 15:09 | |
*** rlandy is now known as rlandy|brb | 15:10 | |
ykarel | yup this is how rhel workflow is | 15:12 |
ykarel | they need to keep it running for years | 15:12 |
ykarel | also they mentioned python3 native venv don't have issue | 15:14 |
marios|ruck | dpawlik: thanks i commented lgtm @ https://review.rdoproject.org/r/#/c/24926/2/roles/rdo-infra/image-server/templates/purge-images.sh.j2 | 15:20 |
marios|ruck | weshay|ruck: when you have time check ^ purge | 15:21 |
marios|ruck | dpawlik: i don't have +2 there | 15:21 |
marios|ruck | ykarel: zbr: so wdyt? should I add 'become' there... or '--user' for https://review.opendev.org/#/c/706572/23/roles/build-containers/tasks/venv_setup.yml ? | 15:21 |
marios|ruck | zbr: or should i just wait? | 15:21 |
marios|ruck | zbr: :) | 15:22 |
zbr | marios|ruck: neither, wait for my fix. | 15:22 |
weshay|ruck | dpawlik-- | 15:22 |
marios|ruck | zbr: :D ok going to remove that explicit task now go back to the mergable version | 15:23 |
marios|ruck | ykarel: ^ | 15:23 |
ykarel | anything will work for me, but NOTE getting virtualenv release in nodepool images is itself will take good time | 15:24 |
dpawlik | weshay|ruck, should I send an email about migration or you will do that :> ? | 15:24 |
ykarel | so if u can wait ok for me | 15:24 |
*** rlandy|brb is now known as rlandy | 15:26 | |
*** derekh has joined #oooq | 15:27 | |
marios|ruck | ykarel: let's see ;) https://review.opendev.org/#/c/706572/25/roles/build-containers/tasks/venv_setup.yml | 15:28 |
marios|ruck | ykarel: if that works we can merge it and then go into witness protection from zbr hit squad(s) | 15:28 |
ykarel | marios|ruck, ack, not sure if --user will work | 15:29 |
ykarel | but yes let's see | 15:29 |
marios|ruck | ykarel: me neither lets see :D | 15:29 |
zbr | marios|ruck: when I hit, I hit workflow! ;) | 15:31 |
marios|ruck | lol | 15:31 |
*** marios|ruck is now known as john_smith | 15:31 | |
*** john_smith is now known as marios|ruck | 15:32 | |
*** saneax has quit IRC | 15:34 | |
marios|ruck | ykarel: installed six ok... now doing upper-constraints.... | 15:47 |
marios|ruck | http://zuul.openstack.org/stream/85ee66be34a043e0bb1b8e86f1e0ca62?logfile=console.log | 15:47 |
rlandy | rfolco: hi - can we chat re:task 1436? | 15:47 |
raukadah | dpawlik, https://logserver.rdoproject.org/29/24929/2/check/periodic-tripleo-rhel-8-buildimage-overcloud-full-master/1f67355/job-output.txt | 15:51 |
raukadah | dpawlik, 2020-02-13 09:59:43.921899 | primary | Warning: Permanently added 'images.rdoproject.org,38.145.33.168' (ECDSA) to the list of known hosts. | 15:52 |
raukadah | 2020-02-13 09:59:44.165090 | primary | rcm-uploader@images.rdoproject.org: Permission denied (publickey). | 15:52 |
raukadah | 2020-02-13 09:59:44.165630 | primary | rsync: connection unexpectedly closed (0 bytes received so far) [sender] | 15:52 |
raukadah | 2020-02-13 09:59:44.165654 | primary | rsync error: unexplained error (code 255) at io.c(226) [sender=3.1.3] | 15:52 |
raukadah | 2020-02-13 09:59:44.171338 | primary | cp: cannot stat '/home/zuul/*repo_setup*log': No such file or directory | 15:52 |
dpawlik | raukadah, because you are trying to reach the old server | 15:52 |
dpawlik | the new server temp name is | 15:52 |
dpawlik | images-vexxhost.rdoproject.org | 15:52 |
dpawlik | in the future we will change that to images.rdoproject.org | 15:53 |
raukadah | ah let me fix that | 15:53 |
raukadah | dpawlik, https://review.rdoproject.org/r/24933 can you cross check again? | 15:55 |
*** Trevor_V has joined #oooq | 15:55 | |
raukadah | sshnaidm, rlandy https://review.rdoproject.org/r/24933 please merge once dpawlik +1 | 15:55 |
dpawlik | raukadah, you have "space" | 15:56 |
marios|ruck | ykarel: looks like it works... it is building containers now at least | 15:58 |
marios|ruck | ykarel: i will split that into a review ontop of yours (just six install) then the check jobs on top of that | 15:58 |
marios|ruck | zbr: ^ | 15:58 |
*** TrevorV has quit IRC | 15:58 | |
sshnaidm | raukadah, what does it fix? | 16:00 |
*** jmasud has quit IRC | 16:02 | |
*** jmasud has joined #oooq | 16:02 | |
rfolco | rlandy, sorry back from lunch | 16:03 |
rlandy | rfolco: k - will ping you in a few | 16:04 |
rfolco | rlandy, ok ping me when you are free | 16:04 |
*** jpena is now known as jpena|off | 16:06 | |
*** jmasud has quit IRC | 16:07 | |
*** udesale_ has joined #oooq | 16:14 | |
*** udesale has quit IRC | 16:14 | |
*** tosky has quit IRC | 16:21 | |
rlandy | rfolco: ping | 16:28 |
raukadah | sshnaidm, it is trying to use old image server | 16:28 |
rlandy | rfolco: I am just trying to figure out where to start here | 16:28 |
rlandy | there is no card for creating jobs | 16:29 |
raukadah | sshnaidm, current image server name is images-vexxhost.rdoproject.org | 16:29 |
rlandy | rfolco: but nvm - I'll just start and work it out | 16:29 |
sshnaidm | raukadah, ack, didn't it migrate already? I thought it will be the same name | 16:30 |
raukadah | sshnaidm, currently, we need to make sure new server is working fine, then we decomossion old one and populate dns on new on | 16:31 |
raukadah | as a fallback we are using images-vexxhost.rdoproject.org | 16:31 |
rfolco | rlandy, we need to replicate the work done for compute/keystone | 16:31 |
rfolco | rlandy, but the promotion names should be changed, so first lets get this merged https://tree.taiga.io/project/tripleo-ci-board/task/1450?kanban-status=1447276 | 16:32 |
rlandy | rfolco: not mergeable :) | 16:32 |
rlandy | anyways no worries | 16:33 |
rfolco | rlandy, yeah, need to rebase that one, just to say the other pipelines should follow this one | 16:33 |
rfolco | rlandy, name ocnvention | 16:33 |
rlandy | k | 16:33 |
sshnaidm | raukadah, ok, so no need to change link in jobs afaiu | 16:34 |
raukadah | dpawlik, ^^ | 16:34 |
sshnaidm | rlandy, are you ok with https://review.rdoproject.org/r/#/c/24933/ now? | 16:35 |
raukadah | sshnaidm, dpawlik will populate redirect to images.rdoproject.org in future as per last discussion once we move to new server | 16:35 |
sshnaidm | raukadah, cool, where are such discussions take place btw? | 16:35 |
raukadah | sshnaidm, https://softwarefactory-project.io/etherpad/p/images_migration | 16:36 |
mjturek | are overcloud and ipa images remaining as centos7 for the foreseeable future? | 16:38 |
*** ykarel is now known as ykarel|away | 16:38 | |
*** udesale_ has quit IRC | 16:40 | |
*** tosky has joined #oooq | 16:43 | |
sshnaidm | raukadah, etherpad.. | 16:44 |
sshnaidm | raukadah, in addition to taiga and jira :D | 16:44 |
dpawlik | raukadah, looks good :) | 16:46 |
*** marios|ruck is now known as marios|out | 16:48 | |
*** tesseract has quit IRC | 16:59 | |
*** marios|out has quit IRC | 17:10 | |
rlandy | sshnaidm: yep - +2'ed it | 17:16 |
*** bogdando has quit IRC | 17:17 | |
*** jpena|off is now known as jpena | 17:27 | |
*** jmasud has joined #oooq | 17:30 | |
rlandy | rfolco: ok if I edit https://review.rdoproject.org/r/#/c/24512/? | 17:37 |
rfolco | rlandy, I am ok even if you take it over | 17:38 |
rfolco | rlandy, yes go for it | 17:38 |
rlandy | rfolco: cool - thanks | 17:38 |
dpawlik | weshay|ruck, https://review.rdoproject.org/r/#/c/24926/ => why -1? | 17:50 |
rlandy | rfolco: sorry to bother you again ... | 17:50 |
rfolco | rlandy, not a bother | 17:51 |
rlandy | rfolco: but all the promote-consistent-to-component-ci-testing job can be centso 8 only | 17:51 |
rlandy | there is no centos 7 left here | 17:51 |
rlandy | or we need both job definitions? | 17:51 |
rlandy | ie: component pipeline is going to move to centos8 with no centos7 surviving, correct?? | 17:52 |
rlandy | weshay|ruck: raukadah: ^^ do I have that right? | 17:52 |
rfolco | rlandy, there is no centos7 component promotion anymore. | 17:53 |
rlandy | rfolco: k | 17:53 |
rfolco | rlandy, dlrn stopped working w/ centos7 | 17:53 |
rfolco | for components | 17:53 |
rlandy | fine - I am removing those definitions | 17:53 |
rfolco | great thanks for updating it | 17:54 |
*** derekh has quit IRC | 17:59 | |
*** jmasud has quit IRC | 18:09 | |
*** amoralej is now known as amoralej|off | 18:12 | |
*** rascasoft has quit IRC | 18:24 | |
weshay|ruck | dpawlik, don't remove any file that is symlinked | 18:24 |
weshay|ruck | dpawlik, there is already a tool that does this | 18:24 |
weshay|ruck | david simard has it | 18:25 |
weshay|ruck | dpawlik, go talk to him | 18:25 |
*** jpena is now known as jpena|off | 18:31 | |
rlandy | question about centos8 parenting | 18:40 |
rlandy | do we want to parent centos8 off centos7 or just start clean definitions? | 18:41 |
sshnaidm | rlandy, why not to parent? | 18:42 |
rlandy | sshnaidm: well if centos7 goes away | 18:43 |
rlandy | then our base is non-existent | 18:43 |
rfolco | zbr, still around? | 18:43 |
sshnaidm | rlandy, yeah, good point | 18:43 |
rlandy | sshnaidm: open question ... will add you to patches so you can comment and we can discuss on monday | 18:44 |
sshnaidm | rlandy, I think rhel8 is parenting from c7 too | 18:44 |
rlandy | sshnaidm: yeah ... | 18:44 |
rlandy | I think centos8 should be our new base going forward but not worth rewriting old job definitions | 18:44 |
rlandy | so I would build a new set of base jobs | 18:45 |
rlandy | anyways - will put forward to the team on monday | 18:45 |
sshnaidm | rlandy, ack | 18:45 |
*** holser has quit IRC | 18:59 | |
rlandy | lol - chicken and egg dependencies | 19:02 |
rlandy | woohoo ... zuul is finally happy ... https://review.rdoproject.org/r/#/c/24512/ | 19:37 |
weshay|ruck | zbr what was the issue opened on molecule that requires 3.0a5? | 19:50 |
rfolco | weshay|ruck, zbr o/ interested ^ | 19:58 |
rfolco | coz I still struggle with molecule | 19:58 |
rlandy | rfolco: could I testproject the centos 8 standalone jobs - or not piushed containers yet? | 20:02 |
weshay|ruck | rlandy, we have not | 20:02 |
rlandy | ok | 20:02 |
weshay|ruck | rlandy, we need to land https://review.rdoproject.org/r/#/c/24762/ | 20:02 |
rlandy | keystone/compute pipeline cards in review - waiting for ^^ | 20:02 |
weshay|ruck | ya.. that's busted on everything https://review.rdoproject.org/zuul/builds?job_name=rdo-tox-molecule | 20:03 |
weshay|ruck | rfolco, it's failing on the destroy? | 20:04 |
rfolco | its because ci-scripts/infra-setup/roles/promoter/molecule/default/playbook.yml does not exist | 20:05 |
rfolco | there is a converge.yml, so not sure how to make molecule see it instead of playbook.yml | 20:05 |
rfolco | we could can create a playbook.yml and include ? I really don't know what we need to do | 20:06 |
rfolco | weshay|ruck, trying to skip it in tox.ini w/ a code comment... lets see | 20:08 |
rfolco | rlandy, not keystone anymore :( | 20:22 |
rfolco | https://trunk.rdoproject.org/centos8-master/component/security/ | 20:22 |
rfolco | weshay|ruck, rlandy zbr: I got a green job on rdo-tox-molecule by ignoring the promoter not-delegated test... | 20:25 |
rfolco | https://review.rdoproject.org/r/#/c/24762/ | 20:25 |
rlandy | lol | 20:25 |
rlandy | fixing | 20:25 |
weshay|ruck | hrm | 20:25 |
rfolco | https://review.rdoproject.org/r/#/c/24762/26/tox.ini@82 | 20:25 |
weshay|ruck | rfolco, but that tox config is in base | 20:25 |
weshay|ruck | right | 20:25 |
rfolco | we still need to fix the broken tests in roles/promoter (not delegated ones) | 20:26 |
weshay|ruck | so we'd be ignoring that test on any change | 20:26 |
rfolco | just to show my get-hash is working :) | 20:26 |
weshay|ruck | rfolco, so there are legit tests failing? or just the teardown is failing | 20:26 |
rfolco | weshay|ruck, I am not suggesting we merge as is | 20:26 |
rfolco | weshay|ruck, its molecule problem I guess | 20:27 |
weshay|ruck | rfolco, k.. | 20:27 |
rfolco | molecule issue w/ file names converge,yml vs playbook.yml | 20:27 |
rfolco | weshay|ruck, after checking that my get-hash molecule test works, now I can readd the broken test and point people to the test that ran | 20:27 |
rfolco | maybe I can try fixing it | 20:28 |
rfolco | let me spend a bit more on it while people review the patch... | 20:28 |
weshay|ruck | rfolco, /me trying a thing too | 20:29 |
weshay|ruck | rfolco, have a fix | 20:33 |
rfolco | weshay|ruck, great pls use patchset 25 and add your fix | 20:34 |
rfolco | as patchset 26 is the skip hack | 20:34 |
weshay|ruck | rfolco, ya.. was testing w/o it | 20:36 |
weshay|ruck | so do this.. remove the change to tox.ini | 20:36 |
weshay|ruck | rfolco, drop this file in http://pastebin.test.redhat.com/836028 | 20:36 |
weshay|ruck | worked for me | 20:36 |
rfolco | lol | 20:37 |
weshay|ruck | :) | 20:37 |
rfolco | weshay|ruck, hacker! | 20:37 |
weshay|ruck | aye aye | 20:37 |
weshay|ruck | bread and butter | 20:38 |
rfolco | ok, cool | 20:38 |
rfolco | changing | 20:38 |
rlandy | rfolco: please review again ... https://review.rdoproject.org/r/#/c/24938/ https://review.rdoproject.org/r/#/c/24512/ | 20:47 |
rfolco | uber kid at school, will review later or tomorrow | 20:50 |
weshay|ruck | rfolco, worked | 21:00 |
*** jfrancoa has quit IRC | 21:06 | |
rlandy | weshay|ruck: rfolco: left comments on https://review.rdoproject.org/r/#/c/24762/ - I think this excludes the downstream pipeline where the primary distro is rhel | 21:09 |
rlandy | if we need to merge this review to get centos 8 going, I can remove -1 and accept downstream will not work for now | 21:10 |
weshay|ruck | rfolco, oh now rlandy -1's it.. when we get it working | 21:11 |
weshay|ruck | jeeeez | 21:11 |
weshay|ruck | ;) | 21:11 |
rlandy | weshay|ruck: rfolco: ^^ again - not going to block centos 8 on this | 21:11 |
weshay|ruck | centos-8 container jobs here we come.. patches in gate | 21:12 |
rlandy | woohoo | 21:13 |
* weshay|ruck lays down | 21:14 | |
*** jfrancoa has joined #oooq | 21:20 | |
rlandy | panda: why do you specify ansible 2.8 in https://review.rdoproject.org/r/#/c/24676/7/zuul.d/jobs.yaml? | 21:29 |
dpawlik | weshay|ruck, file with -type f doesn't remove symlinks, but maybe Im wrong. David is absent, so I will perform a test tomorrow | 21:30 |
*** dpawlik is now known as dpawlik_off | 21:30 | |
panda | rlandy: consistentcy with the definition I copied from below. | 21:38 |
rlandy | panda: ok - guess that 's legacy anyways | 21:39 |
*** Trevor_V has quit IRC | 21:55 | |
rlandy | weshay|ruck: rfolco: looking at 'build out zuul pipelines' task ... so this task is dependent on all promoted-component jobs? | 22:15 |
* rlandy tries to get the order of development sorted | 22:15 | |
rlandy | also - where is the image build work? | 22:16 |
*** sshnaidm is now known as sshnaidm|afk | 22:21 | |
rlandy | pls see card https://tree.taiga.io/project/tripleo-ci-board/task/1518 | 22:34 |
*** jfrancoa has quit IRC | 22:37 | |
*** jmasud has joined #oooq | 22:47 | |
*** Goneri has quit IRC | 22:51 | |
*** jmasud has quit IRC | 23:06 | |
*** jmasud has joined #oooq | 23:09 | |
weshay|ruck | dpawlik_off, you can't remove any files that ARE symlinked | 23:12 |
weshay|ruck | and david simard has a script on the system to do it already | 23:13 |
*** jmasud has quit IRC | 23:14 | |
weshay|ruck | rlandy, can't we just remove lines 36 and 47 https://review.rdoproject.org/r/#/c/24762/27/ci-scripts/infra-setup/roles/get_hash/tasks/get_hash.yaml@36 | 23:18 |
weshay|ruck | I think ya | 23:18 |
* weshay|ruck udpates | 23:18 | |
weshay|ruck | meh.. I guess it's fine.. now | 23:20 |
weshay|ruck | +2 https://review.rdoproject.org/r/#/c/24762/ | 23:20 |
rlandy | weshay|ruck: it's fine | 23:27 |
weshay|ruck | rlandy, I think distro name is not needed.. but meh | 23:27 |
rlandy | weshay|ruck: I +2'ed it | 23:28 |
weshay|ruck | thanks | 23:28 |
rlandy | I wasn't going to block centos8 over it | 23:28 |
rlandy | weshay|ruck: I left some notes in https://tree.taiga.io/project/tripleo-ci-board/task/1518 - could use a little direction on how to proceed with next reviews | 23:29 |
rlandy | did zbr cover the image build jobs? | 23:29 |
rlandy | your note said 2/15 | 23:29 |
rlandy | anyways .. need to go to other job now .. bbl | 23:30 |
*** rlandy is now known as rlandy|bbl | 23:30 | |
*** jmasud has joined #oooq | 23:48 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!