*** ianychoi has quit IRC | 00:17 | |
*** ianychoi has joined #openstack-helm | 00:18 | |
*** sthussey has quit IRC | 00:20 | |
*** jistr has quit IRC | 01:00 | |
*** jistr has joined #openstack-helm | 01:00 | |
*** cheng1 has joined #openstack-helm | 01:26 | |
*** cheng1 has quit IRC | 03:02 | |
*** chenyb4 has joined #openstack-helm | 03:03 | |
*** kranthikirang has quit IRC | 03:06 | |
*** cheng1 has joined #openstack-helm | 03:08 | |
chenyb4 | Hi, guys! Use helm deploy openstack, Change mariadb deploy from openstack namespace to database namespace, then deploy keystone appear error “Resolving dependency Service mariadb in namespace database failed: endpoints “mariadb” is forbidden: User “system:serviceaccount:openstack:keystone-db-init” cannot get resource “endpoints” in API group “” in the namespace “database” .” | 03:10 |
---|---|---|
chenyb4 | How can i chanage? | 03:10 |
chenyb4 | Pelease give me a help, th | 03:10 |
*** cheng1 has quit IRC | 03:10 | |
*** cheng1 has joined #openstack-helm | 03:13 | |
*** cheng1_ has joined #openstack-helm | 03:17 | |
*** cheng1 has quit IRC | 03:19 | |
portdirect | chenyb4: I think there is a bug in the service account function for the init container across namespace with services :( | 03:20 |
portdirect | This function will need updated: https://github.com/openstack/openstack-helm-infra/blob/master/helm-toolkit/templates/snippets/_kubernetes_pod_rbac_serviceaccount.tpl | 03:22 |
chenyb4 | portdirect, I don't known how can i change. init container dependency "DEPENDENCY_SERVICE: database:mariadb", but db-init role or rolebinding not create finish. | 03:23 |
portdirect | It's late here atm, but we can get a look at it tomorrow morning (usa) | 03:23 |
portdirect | For now if you disable rbac in the cluster that would let you test other aspects of interaction across namespaces | 03:24 |
portdirect | But obviously not a viable solution for more than just playing around :( | 03:24 |
chenyb4 | portdirect, ok, thanks | 03:25 |
*** cfriesen has quit IRC | 03:34 | |
*** cheng1_ has quit IRC | 03:35 | |
*** jaypipes_ has quit IRC | 04:02 | |
*** jaypipes_ has joined #openstack-helm | 04:02 | |
*** cheng1_ has joined #openstack-helm | 04:14 | |
*** cheng1_ has quit IRC | 04:16 | |
*** cheng1 has joined #openstack-helm | 04:20 | |
*** cheng1 has quit IRC | 04:33 | |
*** cheng1 has joined #openstack-helm | 04:35 | |
openstackgerrit | Gage Hugo proposed openstack/openstack-helm master: Add wait.resources.type to armada manifest https://review.openstack.org/648503 | 04:57 |
*** chenyb4 has quit IRC | 05:05 | |
*** jamesgu has quit IRC | 06:10 | |
*** chenyb4 has joined #openstack-helm | 06:16 | |
*** Darcidride has joined #openstack-helm | 06:21 | |
openstackgerrit | chinasubbareddy mallavarapu proposed openstack/openstack-helm-infra master: WIP] Test - ceph-osd udev fixes https://review.openstack.org/645460 | 06:22 |
*** jsuchome has joined #openstack-helm | 06:49 | |
*** Darcidride has quit IRC | 06:59 | |
*** dpawlik has joined #openstack-helm | 07:17 | |
*** pgaxatte has joined #openstack-helm | 07:36 | |
*** witek has joined #openstack-helm | 08:04 | |
*** Darcidride has joined #openstack-helm | 08:08 | |
openstackgerrit | chinasubbareddy mallavarapu proposed openstack/openstack-helm-infra master: [ceph-doc] move monolithic ceph-osd chart to multiple ceph-osd charts https://review.openstack.org/647810 | 08:59 |
evrardjp | morning | 09:00 |
*** itxaka has joined #openstack-helm | 09:02 | |
*** itxaka has quit IRC | 09:06 | |
*** dimitris_ has joined #openstack-helm | 09:18 | |
*** alisanhaji has joined #openstack-helm | 09:21 | |
*** cheng1 has quit IRC | 09:50 | |
*** cheng1 has joined #openstack-helm | 09:51 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-helm-images master: Fix inconsistent image name in libvirt https://review.openstack.org/648657 | 09:59 |
evrardjp | I have noticed the image for libvirt was not consistenty named, so this fixes it ^ | 10:00 |
evrardjp | I added the cores reviewers (Not sure which of you have the credentials) to this review, so that you can eventually clean up things on the docker hub | 10:01 |
jsuchome | evrardjp: why is libvirt image using this naming scheme, it's confusing? | 10:02 |
evrardjp | it's a bug :) | 10:02 |
evrardjp | As you can see, 648657 is fixing it. | 10:02 |
jsuchome | that's not what i mean | 10:02 |
evrardjp | When things will run, all the change_ will be cleaned | 10:02 |
evrardjp | but the other images are not removed | 10:03 |
jsuchome | I mean why use "latest-" as a prefix when nobody else is doing it? | 10:03 |
openstackgerrit | Merged openstack/openstack-helm master: Move adding extra dashboard to the proper place https://review.openstack.org/645547 | 10:03 |
evrardjp | jsuchome: all the images are using latest | 10:03 |
evrardjp | in osh-images | 10:03 |
jsuchome | " yet libvirt is doing the other way around." - this is what I'm wondering, why? | 10:03 |
evrardjp | because of a bug | 10:04 |
evrardjp | when I re-wrote this, I seem to have omitted libvirt | 10:04 |
evrardjp | and I re-wrote this when the convention was set | 10:05 |
*** chenyb4 has quit IRC | 10:06 | |
evrardjp | jsuchome: did that answer yoru question? | 10:06 |
evrardjp | Not really sure :( | 10:06 |
jsuchome | oh, so you are saying it is _correct_ to use "latest-" as a prefix, and you are fixing it for libvirt so it does it as well | 10:07 |
jsuchome | I think the commit message is bit confusing | 10:07 |
jsuchome | posted a comment | 10:08 |
evrardjp | ok | 10:08 |
evrardjp | thanks! | 10:08 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-helm-images master: Fix inconsistent image name in libvirt https://review.openstack.org/648657 | 10:10 |
evrardjp | jsuchome: could you tell me if that's clearer ? ^ | 10:10 |
jsuchome | yes, thanks | 10:16 |
openstackgerrit | Itxaka Serrano Garcia proposed openstack/openstack-helm master: Allow more generic overrides for keystone https://review.openstack.org/647403 | 10:20 |
openstackgerrit | Itxaka Serrano Garcia proposed openstack/openstack-helm master: Allow more generic overrides for horizon https://review.openstack.org/645141 | 10:21 |
openstackgerrit | Itxaka Serrano Garcia proposed openstack/openstack-helm master: Add heat dashboard if available to horizon https://review.openstack.org/645543 | 10:28 |
*** lemko has joined #openstack-helm | 10:50 | |
*** ohryhorov has left #openstack-helm | 11:11 | |
*** ohryhorov has joined #openstack-helm | 11:11 | |
*** jaypipes_ is now known as leakypipes | 12:30 | |
*** cheng1 has quit IRC | 12:39 | |
*** cheng1 has joined #openstack-helm | 12:40 | |
*** cheng1 has quit IRC | 12:49 | |
*** aaronsheffield has joined #openstack-helm | 12:54 | |
openstackgerrit | Rahul Khiyani proposed openstack/openstack-helm-infra master: ceph-rgw: Add pod/container security context https://review.openstack.org/648554 | 12:54 |
*** howell has joined #openstack-helm | 13:20 | |
*** cheng1 has joined #openstack-helm | 13:27 | |
*** cfriesen has joined #openstack-helm | 13:46 | |
*** cheng1 has quit IRC | 13:46 | |
*** cheng1 has joined #openstack-helm | 13:46 | |
*** kranthikirang has joined #openstack-helm | 13:56 | |
openstackgerrit | Angie Wang proposed openstack/openstack-helm master: Add Panko Chart https://review.openstack.org/641144 | 14:03 |
openstackgerrit | Angie Wang proposed openstack/openstack-helm master: Add Aodh Chart https://review.openstack.org/641518 | 14:03 |
openstackgerrit | Angie Wang proposed openstack/openstack-helm master: Ceilometer chart: add the ability to publish events to panko https://review.openstack.org/641519 | 14:03 |
evrardjp | portdirect: I/we have a few questions about some images ... how did you build this: docker.io/port/ceph-config-helper:v1.10.3 ?What is nova_{spice|novnc}proxy_assets for (if necessary we can dig into the chart)? | 14:03 |
*** jamesgu has joined #openstack-helm | 14:07 | |
sgrasley | It looks like neutron is defaulting to flat external/provider networking. does helm neutron support vlan external/provider networking? Are there example overrides to use vlan?? | 14:09 |
evrardjp | that would be good for docs | 14:16 |
evrardjp | sgrasley: https://github.com/openstack/openstack-helm/tree/master/tools/overrides/backends/networking has multiple provider networks | 14:17 |
evrardjp | mechanisms* | 14:17 |
evrardjp | on top of being a more complex sr-iov though | 14:17 |
sgrasley | everardjp: Thanks for the pointer. I didn't think to look in the tools | 14:21 |
portdirect | ^ yeah - this is one of the areas of our docs we really need to work on | 14:21 |
portdirect | thx evrardjp for the link | 14:21 |
portdirect | if you need any more help sgrasley i can try and get some of the authors of the neutron chart in here, or on the ml | 14:22 |
sgrasley | Thanks! | 14:22 |
portdirect | but for the most part, things are fairly unopinionated, its just a case of transposing the oslo conf to our yaml representation of it | 14:23 |
*** pma78 has quit IRC | 14:30 | |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: WIP: enable compute kit to be deployed from outwidth the cluster https://review.openstack.org/647629 | 14:34 |
openstackgerrit | chinasubbareddy mallavarapu proposed openstack/openstack-helm-infra master: [ceph-doc] move monolithic ceph-osd chart to multiple ceph-osd charts https://review.openstack.org/647810 | 14:51 |
*** jamesgu has quit IRC | 14:54 | |
*** alisanhaji has quit IRC | 14:57 | |
*** dustinspecker has joined #openstack-helm | 14:58 | |
*** dimitris_ has quit IRC | 15:06 | |
*** aagate has joined #openstack-helm | 15:15 | |
*** dpawlik has quit IRC | 15:16 | |
*** nick_kar has quit IRC | 15:18 | |
*** howell has quit IRC | 15:24 | |
*** jamesgu has joined #openstack-helm | 15:25 | |
*** roman_g has quit IRC | 15:26 | |
*** howell has joined #openstack-helm | 15:26 | |
mcfuller | Is there any documentation available to describe how to properly use the run_tempest flag for individual helm charts? Since those tests run immediately after helm install / upgrade, they test a particular service type, but the verifier also seems to expect that the whole cloud has already been deployed. Is this expected? | 15:30 |
mcfuller | For example, in a sequential deployment of OS services, if I enable run_tempest on glance, the test will fail because it's expecting to find a compute service endpoint, but nova won't get deployed until later as part of the compute kit. Is it better to run the standalone tempest chart at the end of a deployment? | 15:38 |
evrardjp | mcfuller: I don't know how each tempest configuration is run, but if we have that in each chart, it would make sense to only scope the appropriate tests of the chart | 15:42 |
evrardjp | by whitelisting only the ones that matter | 15:42 |
evrardjp | IMO | 15:42 |
evrardjp | let me check if there is an override | 15:43 |
evrardjp | https://github.com/openstack/openstack-helm/blob/master/tempest/templates/_helpers.tpl | 15:44 |
evrardjp | https://github.com/openstack/openstack-helm/blob/master/tempest/templates/configmap-etc.yaml#L47-L53 | 15:44 |
portdirect | jayahn ^ | 15:44 |
portdirect | not sure if your arround - but would be great to get some insight into the tempest function within helm test if you are | 15:45 |
mcfuller | evrardjp: the rally_test.sh script that gets called when run_tempest is true determines the service type and then runs tests based on the pattern "tempest.api.<service>*" | 15:45 |
evrardjp | it's even given in example: https://github.com/openstack/openstack-helm/blob/master/tempest/values.yaml#L95-L121 | 15:45 |
evrardjp | so it runs only api testing of service -- what wrong then? | 15:46 |
evrardjp | I guess with an output of the tests that run it would be easier to understand | 15:46 |
mcfuller | So for glance, it passes in the pattern "tempest.api.image*", but then the first thing it does is reach out to keystone service catalog looking for the compute endpoint which doesn't exist yet | 15:46 |
evrardjp | oh that's weird it looks for compute endpoint | 15:47 |
evrardjp | I understand looking for endpoints, but compute endpoint... | 15:47 |
mcfuller | They all seem to do that as a first step with the ones I've tested so far | 15:47 |
evrardjp | interesting | 15:48 |
evrardjp | do you have a run log I can use? | 15:48 |
mcfuller | Hmm...it looks like the only logs I have now are from when I manually ran tests after the whole stack was up and running, which has a different error now :) | 15:51 |
mcfuller | But it's easy to reproduce, so I'll run it again and get logs to you | 15:51 |
evrardjp | so I am backtracking as I am very new to this | 15:51 |
evrardjp | could you tell me what's in confiMap rally-test.sh ? | 15:52 |
evrardjp | for, for example, glance? | 15:52 |
evrardjp | I don't know much about rally, I am not sure if it's a tempest thing | 15:53 |
mcfuller | https://github.com/openstack/openstack-helm-infra/blob/master/helm-toolkit/templates/scripts/_rally_test.sh.tpl | 15:53 |
evrardjp | https://github.com/openstack/openstack-helm-infra/blob/master/helm-toolkit/templates/scripts/_rally_test.sh.tpl#L69-L74 | 15:53 |
evrardjp | haha | 15:53 |
evrardjp | yes indeed that's what's running | 15:54 |
evrardjp | I expect these to run: https://github.com/openstack/tempest/tree/master/tempest/api/image/v2 not anything else that's why a log would be interesitng | 15:55 |
mcfuller | And actually, one problem I ran into with that script is that it's not expecting multiple service types. grepping for 'cinder' returns a service type of 'volumevolumev2volumev3' | 15:56 |
evrardjp | It's getting late for me, but I am interested in this -- because I will have to learn it eventually | 16:05 |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: WIP: run OSH on azure https://review.openstack.org/648723 | 16:11 |
evrardjp | do you have extra credits on azure? portdirect ;) | 16:32 |
*** jsuchome has quit IRC | 16:33 | |
evrardjp | portdirect: I will add a run OSH on k3s | 16:33 |
evrardjp | 5 less | 16:33 |
*** roman_g has joined #openstack-helm | 16:33 | |
portdirect | Running on k3s is just the same steps as minikube | 16:34 |
*** pgaxatte has quit IRC | 16:36 | |
*** witek has quit IRC | 16:43 | |
openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Mariadb: move to mariabackup as wsrep sst method https://review.openstack.org/648730 | 16:46 |
*** witek has joined #openstack-helm | 16:46 | |
*** witek has quit IRC | 17:00 | |
openstackgerrit | chinasubbareddy mallavarapu proposed openstack/openstack-helm-infra master: [ceph-osd] Fix ceph osd init script to handle udev https://review.openstack.org/644689 | 17:06 |
openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Mariadb: move to mariabackup as wsrep sst method https://review.openstack.org/648730 | 17:08 |
openstackgerrit | Chris Wedgwood proposed openstack/openstack-helm master: Add network policy ingress rule to mariadb https://review.openstack.org/638299 | 17:13 |
*** jistr is now known as jistr|off | 17:45 | |
*** jistr|off is now known as jistr | 17:45 | |
evrardjp | I am mostly joking about it portdirect :) | 17:47 |
evrardjp | portdirect: I wrote a few questions higher in the backlog, if you have time to answer it, it would be great, so that it doesn't go through the void of an irc backlog | 17:47 |
*** lemko has quit IRC | 17:50 | |
openstackgerrit | Merged openstack/openstack-helm master: Add Panko Chart https://review.openstack.org/641144 | 17:55 |
*** Darcidride has quit IRC | 17:57 | |
openstackgerrit | kranthi kiran guttikonda proposed openstack/openstack-helm-infra master: Improves ceph-provisioner storageclasses https://review.openstack.org/636768 | 18:00 |
*** roman_g has quit IRC | 18:34 | |
openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: [WIP] Make network policy template a little more generic https://review.openstack.org/645400 | 19:06 |
*** dpawlik has joined #openstack-helm | 19:20 | |
*** sthussey has joined #openstack-helm | 19:33 | |
*** dpawlik has quit IRC | 19:57 | |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Heat: make ques ha https://review.openstack.org/648765 | 20:01 |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Heat: make ques ha https://review.openstack.org/648765 | 20:02 |
*** dpawlik has joined #openstack-helm | 20:09 | |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Cinder: make rabbitmq queues ha https://review.openstack.org/648768 | 20:10 |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Cinder: make rabbitmq queues ha https://review.openstack.org/648768 | 20:12 |
*** dpawlik has quit IRC | 20:14 | |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Nova: make rabbitmq queues ha https://review.openstack.org/648769 | 20:17 |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Neutron: make rabbitmq queues ha https://review.openstack.org/648770 | 20:17 |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Glance: make rabbitmq queues ha https://review.openstack.org/648771 | 20:17 |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Nova: make rabbitmq queues ha https://review.openstack.org/648769 | 20:18 |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Neutron: make rabbitmq queues ha https://review.openstack.org/648770 | 20:18 |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Glance: make rabbitmq queues ha https://review.openstack.org/648771 | 20:18 |
*** jamesgu has quit IRC | 20:28 | |
*** dustinspecker has quit IRC | 20:32 | |
openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Heat: make rabbitmq queues ha https://review.openstack.org/648765 | 20:42 |
openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: WIP: fix docs build https://review.openstack.org/648780 | 20:59 |
*** howell has quit IRC | 21:32 | |
openstackgerrit | Merged openstack/openstack-helm-images master: Improve performance for building https://review.openstack.org/644537 | 22:15 |
*** sthussey has quit IRC | 23:01 | |
*** aaronsheffield has quit IRC | 23:03 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!