| *** ericho has quit IRC | 00:09 | |
| *** sthussey has quit IRC | 00:26 | |
| openstackgerrit | Matt McEuen proposed openstack/openstack-helm-infra master: Fix jq quoting issue when waiting for nodes https://review.openstack.org/574499 | 00:47 |
|---|---|---|
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Gate: dont deploy hwe kernel by default https://review.openstack.org/574381 | 01:03 |
| *** unicell has quit IRC | 01:58 | |
| *** julim has joined #openstack-helm | 02:36 | |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Deployments: Use current kubernetes deployment version https://review.openstack.org/574517 | 02:51 |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Deployments: Use current kubernetes deployment version https://review.openstack.org/574519 | 02:53 |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Daemonsets: Use current kubernetes daemonset api version https://review.openstack.org/574525 | 03:02 |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm master: Daemonsets: Use current kubernetes daemonset api version https://review.openstack.org/574526 | 03:04 |
| *** sofioni has joined #openstack-helm | 03:06 | |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Prometheus: update function to live in correct location https://review.openstack.org/574528 | 03:08 |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Fluent-Logging: update functions to live in correct locations https://review.openstack.org/574529 | 03:14 |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Nagios: update functions to live in correct locations https://review.openstack.org/574531 | 03:15 |
| *** zhangfei has joined #openstack-helm | 03:16 | |
| *** piotrrr has quit IRC | 03:21 | |
| *** piotrrr has joined #openstack-helm | 03:27 | |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Deployments: Use current kubernetes deployment version https://review.openstack.org/574517 | 03:34 |
| *** sofioni has quit IRC | 03:41 | |
| *** ianychoi has joined #openstack-helm | 03:51 | |
| *** yamamoto has joined #openstack-helm | 04:08 | |
| *** sofioni has joined #openstack-helm | 04:46 | |
| *** zhangfei has quit IRC | 04:57 | |
| *** MarkBaker has quit IRC | 05:07 | |
| *** zhangfei has joined #openstack-helm | 05:12 | |
| *** unicell has joined #openstack-helm | 05:35 | |
| *** unicell has quit IRC | 05:36 | |
| *** unicell has joined #openstack-helm | 05:41 | |
| *** jgu has joined #openstack-helm | 05:48 | |
| *** zhangfei has quit IRC | 06:06 | |
| *** zhangfei has joined #openstack-helm | 06:19 | |
| *** cNilesh has joined #openstack-helm | 06:52 | |
| *** unicell has quit IRC | 06:53 | |
| *** gkadam has joined #openstack-helm | 07:15 | |
| *** jgu has quit IRC | 07:18 | |
| *** mdih has quit IRC | 07:32 | |
| *** gagehugo_ has joined #openstack-helm | 08:31 | |
| *** gagehugo has quit IRC | 08:32 | |
| *** mdih has joined #openstack-helm | 08:35 | |
| *** cNilesh_ has joined #openstack-helm | 09:49 | |
| *** cNilesh has quit IRC | 09:49 | |
| *** julim has quit IRC | 10:50 | |
| *** cristicalin has joined #openstack-helm | 12:09 | |
| *** roman_g has joined #openstack-helm | 12:11 | |
| *** julim has joined #openstack-helm | 12:30 | |
| *** julim has quit IRC | 13:13 | |
| *** julim has joined #openstack-helm | 13:15 | |
| *** kman has joined #openstack-helm | 13:16 | |
| *** cristicalin has quit IRC | 13:19 | |
| *** felipemonteiro_ has joined #openstack-helm | 13:23 | |
| *** felipemonteiro_ has quit IRC | 13:29 | |
| *** yamamoto has quit IRC | 13:36 | |
| *** yamamoto has joined #openstack-helm | 13:37 | |
| *** aaronsheffield has joined #openstack-helm | 13:39 | |
| *** jdandrea has joined #openstack-helm | 13:51 | |
| *** sthussey has joined #openstack-helm | 14:09 | |
| *** eeiden has joined #openstack-helm | 14:15 | |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: PVC monitoring: Add alerting rules and service check for PVCs https://review.openstack.org/574316 | 14:28 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: Add missing node selectors to chart templates https://review.openstack.org/574384 | 14:28 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: WIP: Add prometheus exporter manifest template to HTK https://review.openstack.org/574469 | 14:28 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: Elasticsearch S3 repo https://review.openstack.org/559417 | 14:33 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: Elasticsearch S3 repo https://review.openstack.org/559417 | 14:33 |
| *** kman has quit IRC | 14:45 | |
| *** jgu has joined #openstack-helm | 14:52 | |
| mattmceuen | Here's the agenda for our meeting coming up in 5min: https://etherpad.openstack.org/p/openstack-helm-meeting-2018-06-12 | 14:55 |
| mattmceuen | meeting time! | 15:00 |
| openstackgerrit | Merged openstack/openstack-helm-infra master: Gate: dont deploy hwe kernel by default https://review.openstack.org/574381 | 15:23 |
| *** zhangfei has quit IRC | 15:25 | |
| *** yamamoto has quit IRC | 15:39 | |
| *** yamamoto has joined #openstack-helm | 15:40 | |
| *** yamamoto has quit IRC | 15:45 | |
| rwellum | cfriesen: to continue your question from meeting: "speaking of that...I've asked on the main channel a few times about how to configure the nova-compute chart to handle heterogeneous compute nodes, multiple sizes of hugepages, PCI devices, dedicated CPUs, etc. Can anyone point me at discussions/documentation on how this is supposed to be done?" | 16:01 |
| cfriesen | rwellum: thanks...was just going to copy that over. We currently configure our compute nodes with a certain amount of CPUs and RAM reserved for the host, and it's configurable per-compute-node how much is reserved. | 16:02 |
| cfriesen | we also do a lot of instances with dedicated CPUs, we support configurable amounts of 2MB and 1GB hugepages per compute node | 16:03 |
| cfriesen | and we do a lot of PCI/SRIOV | 16:03 |
| rwellum | Same cfriesen - I need these answers too | 16:03 |
| cfriesen | that doesn't seem to translate well into the daemonset model for nova-compute | 16:04 |
| cfriesen | also, in an ideal world it would be nice to have a compute node be able to run non-openstack K8s pods as well | 16:05 |
| *** julim has quit IRC | 16:08 | |
| *** gagehugo_ is now known as gagehugo | 16:09 | |
| *** cNilesh_ has quit IRC | 16:15 | |
| rwellum | cfriesen: +1 yup. So you've asked before and no-one answered? | 16:22 |
| cfriesen | rwellum: yes | 16:24 |
| *** yamamoto has joined #openstack-helm | 16:42 | |
| *** yamamoto has quit IRC | 16:49 | |
| *** julim has joined #openstack-helm | 16:52 | |
| mattmceuen | Sorry for that cfriesen - have you seen this doc yet? https://docs.openstack.org/openstack-helm/latest/devref/node-and-label-specific-configurations.html | 17:00 |
| mattmceuen | Was added by this PS https://review.openstack.org/#/c/524422/ | 17:07 |
| mattmceuen | Let me know if that approach wouldn't work for what you're after | 17:07 |
| cfriesen | on a call, will take a look | 17:13 |
| *** unicell has joined #openstack-helm | 17:17 | |
| *** osh-chatbot has joined #openstack-helm | 17:25 | |
| anticw | osh-chatbot: ping | 17:26 |
| osh-chatbot | <cw> anticw: pong | 17:34 |
| portdirect | oh - chatbock is back | 17:35 |
| portdirect | *bot | 17:35 |
| portdirect | thx cw :) | 17:35 |
| anticw | portdirect: np, i think we'll move this in a little while to somewhere else, i had a plan to do that but the "move to place" would have also been somewhat affected by recent glitches so requires a bit more thought | 17:36 |
| *** collision has joined #openstack-helm | 17:36 | |
| collision | Hi all | 17:37 |
| collision | What should be my value for kubernetes_network_default_device in /opt/openstack-helm-infra/tools/gate/devel/multinode-vars.yaml ? | 17:37 |
| collision | for multinode installation | 17:37 |
| cfriesen | mattmceuen: thanks for that link...looks like it should work for the per-node settings. I assume you'd have to upgrade the helm chart any time any of those things change for any compute node? | 17:42 |
| cfriesen | mattmceuen: I'm still not clear on how you ensure that k8s pods don't run on host CPUs that are intended to be used as dedicated CPUs for openstack guests | 17:42 |
| *** collision has quit IRC | 17:44 | |
| cfriesen | mattmceuen: and I'm not sure how to handle the fact that we can have multiple sizes of hugepages and K8s can only handle one size (last I checked anyway). Do we just keep k8s unaware of hugepages entirely? | 17:49 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: WIP: Use Elasticsearch to gather logs in post-run jobs https://review.openstack.org/573007 | 17:53 |
| openstackgerrit | Tin Lam proposed openstack/openstack-helm-infra master: Remove a check for docker proxy task https://review.openstack.org/574837 | 17:58 |
| *** Romik has joined #openstack-helm | 18:07 | |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm master: Ceph-config-helper: Add rgwadmin python package https://review.openstack.org/573405 | 18:24 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm master: Update Armada LMA manifest to include Elasticsearch S3 Repository https://review.openstack.org/572201 | 18:26 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm master: Neutron: Increase default number of RPC workers https://review.openstack.org/573689 | 18:32 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm master: Developer environment: remove ipcalc from required packages https://review.openstack.org/574085 | 18:32 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: Add missing node selectors to chart templates https://review.openstack.org/574384 | 18:34 |
| *** yamamoto has joined #openstack-helm | 18:47 | |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: WIP: Elasticsearch S3 repo https://review.openstack.org/559417 | 18:54 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: Elasticsearch S3 repo https://review.openstack.org/559417 | 19:00 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm master: Update armada source repository for armada multinode check https://review.openstack.org/572122 | 19:01 |
| *** julim has quit IRC | 19:14 | |
| *** sofioni has quit IRC | 19:35 | |
| *** mdih has quit IRC | 20:08 | |
| *** yamamoto has quit IRC | 20:23 | |
| openstackgerrit | Hee Won Lee proposed openstack/openstack-helm master: Docs: Update ceph documentation https://review.openstack.org/566381 | 20:30 |
| *** aaronsheffield has quit IRC | 20:35 | |
| openstackgerrit | Hee Won Lee proposed openstack/openstack-helm master: Docs: Update ceph documentation https://review.openstack.org/566381 | 20:41 |
| *** gkadam has quit IRC | 20:42 | |
| openstackgerrit | Tin Lam proposed openstack/openstack-helm-infra master: Update toolkit to support ingress TLS https://review.openstack.org/568132 | 20:49 |
| openstackgerrit | Tin Lam proposed openstack/openstack-helm-infra master: Update toolkit to support ingress TLS https://review.openstack.org/568132 | 20:52 |
| *** Romik has quit IRC | 21:12 | |
| *** Romik has joined #openstack-helm | 21:13 | |
| *** Romik has quit IRC | 21:13 | |
| *** Romik has joined #openstack-helm | 21:14 | |
| *** Romik has quit IRC | 21:14 | |
| *** Romik has joined #openstack-helm | 21:14 | |
| *** Romik has quit IRC | 21:15 | |
| *** Romik has joined #openstack-helm | 21:15 | |
| *** Romik has quit IRC | 21:15 | |
| *** Romik has joined #openstack-helm | 21:16 | |
| *** Romik has quit IRC | 21:16 | |
| *** yamamoto has joined #openstack-helm | 21:24 | |
| *** eeiden has quit IRC | 21:30 | |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Deployments: Use current kubernetes deployment version https://review.openstack.org/574517 | 21:38 |
| *** yamamoto has quit IRC | 21:39 | |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Deployments: Use current kubernetes deployment version https://review.openstack.org/574517 | 21:46 |
| openstackgerrit | Tin Lam proposed openstack/openstack-helm master: Ingress TLS for keystone https://review.openstack.org/566350 | 21:47 |
| openstackgerrit | Merged openstack/openstack-helm-infra master: PVC monitoring: Add alerting rules and service check for PVCs https://review.openstack.org/574316 | 21:53 |
| *** sofioni has joined #openstack-helm | 22:04 | |
| *** sofioni has quit IRC | 22:09 | |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: WIP: Add prometheus exporter manifest template to HTK https://review.openstack.org/574469 | 22:23 |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: Kube-state-metrics: Update resources in clusterrole https://review.openstack.org/574909 | 22:26 |
| *** collision has joined #openstack-helm | 22:27 | |
| openstackgerrit | Steve Wilkerson proposed openstack/openstack-helm-infra master: WIP: Add prometheus exporter manifest template to HTK https://review.openstack.org/574469 | 22:27 |
| collision | Hello, good evening all. | 22:27 |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Deployments: Use current kubernetes deployment version https://review.openstack.org/574517 | 22:34 |
| openstackgerrit | Pete Birley proposed openstack/openstack-helm-infra master: Use current kubernetes API version https://review.openstack.org/574517 | 22:36 |
| *** jgu_ has joined #openstack-helm | 22:45 | |
| *** jgu has quit IRC | 22:48 | |
| *** cfriesen has quit IRC | 22:49 | |
| collision | I am failing at this stage "make dev-deploy k8s multinode" . Here is the output -> https://pastebin.com/STQ3FPwh . Any idea will be appreciated. | 22:58 |
| *** unicell has quit IRC | 23:17 | |
| collision | I think i am getting wrong interface with this script on installation page; | 23:30 |
| collision | #!/bin/bash set -xe function net_default_iface { sudo ip -4 route list 0/0 | awk '{ print $5; exit }' } cat > /opt/openstack-helm-infra/tools/gate/devel/multinode-vars.yaml <<EOF kubernetes_network_default_device: $(net_default_iface) EOF | 23:30 |
| portdirect | Do you have the same network interfaces on each node? | 23:32 |
| collision | no i dont | 23:33 |
| collision | some are eth0 | 23:33 |
| collision | some are ens3 | 23:33 |
| portdirect | The gate scripts are very simplistic, and may have difficulty coping with that :( | 23:33 |
| portdirect | Can you deploy kubernetes using kubeadm without the playbook? | 23:33 |
| collision | I can but im curious if i can define more than 1 interface into /opt/openstack-helm-infra/tools/gate/devel/multinode-vars.yaml ? | 23:34 |
| portdirect | I'm out atm, but can have a look into how we could cope with your env tomorrow morning | 23:34 |
| portdirect | Nope :( | 23:34 |
| portdirect | Those scripts were built for gating and dev really | 23:35 |
| collision | Thank you :) | 23:35 |
| collision | One more thing in my mind | 23:35 |
| collision | how it will work if i setup a kube cluster with Rancher | 23:35 |
| collision | and install openstack-helm on them ? | 23:35 |
| collision | Do you think its feasable for production ? | 23:36 |
| portdirect | Provided your k8s is 1.9 or above | 23:36 |
| portdirect | Should be fine | 23:36 |
| portdirect | You'll need to ensure ceph-common is deployed on your hosts | 23:36 |
| collision | hmm | 23:37 |
| collision | Is worker nodes mean compute nodes ? | 23:37 |
| portdirect | And that you modify the /etc/resolv.conf to point to the k8s dns service | 23:37 |
| portdirect | Typically, yes | 23:37 |
| portdirect | Anyway better run, in the stalls of a baseball game :) | 23:38 |
| collision | hehe :) | 23:38 |
| *** sthussey has quit IRC | 23:38 | |
| collision | I change ens to docker0 | 23:39 |
| collision | and it seems not failing yet | 23:39 |
| portdirect | That will work to deploy, though your networking between pods will not :( | 23:53 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!