Wednesday, 2018-01-03

*** larainema has joined #openstack-helm00:52
*** ljjjustin has joined #openstack-helm01:00
*** lamt` has quit IRC01:10
*** mdih has joined #openstack-helm01:18
*** yangyapeng has quit IRC01:51
*** yangyapeng has joined #openstack-helm01:52
*** mdih has quit IRC02:17
*** mdih has joined #openstack-helm05:03
*** yamamoto has joined #openstack-helm05:10
openstackgerritRenis Makadia proposed openstack/openstack-helm-infra master: Loopback device setup/configuration - Setup loopback devices using Ansible - Ensure devices are available after reboot  https://review.openstack.org/53030205:43
jayahnsrwilkers: have you had any problem with fluent-bit w/ ealsticsearch output? for example, garbage like .......""took""errors""took""errors""took""errors""took""errors""took""errors""took""errors""took""errors""took""errors""......06:16
openstackgerritRenis Makadia proposed openstack/openstack-helm-infra master: Loopback device setup/configuration - Setup loopback devices using Ansible - Ensure devices are available after reboot  https://review.openstack.org/53030206:26
*** coolsvap has joined #openstack-helm06:35
openstackgerritRenis Makadia proposed openstack/openstack-helm-infra master: Loopback device setup/configuration  https://review.openstack.org/53030206:40
*** yamamoto has quit IRC06:49
*** yamamoto has joined #openstack-helm07:38
*** yamamoto has quit IRC07:49
openstackgerritRenis Makadia proposed openstack/openstack-helm-infra master: Loopback device setup/configuration  https://review.openstack.org/53030208:37
*** krypto has joined #openstack-helm08:42
*** yamamoto has joined #openstack-helm08:49
*** yamamoto has quit IRC08:57
*** xek has quit IRC08:57
*** xek has joined #openstack-helm08:57
*** xek has left #openstack-helm08:58
*** MarkBaker has quit IRC09:42
*** MarkBaker has joined #openstack-helm09:43
*** yamamoto has joined #openstack-helm09:53
*** ljjjustin has quit IRC10:18
*** haint has quit IRC11:56
*** yamamoto has quit IRC12:24
*** yangyapeng has quit IRC12:45
*** yangyapeng has joined #openstack-helm12:46
*** yangyapeng has quit IRC12:50
*** yamamoto has joined #openstack-helm12:56
*** tlam has joined #openstack-helm13:02
*** yangyapeng has joined #openstack-helm13:12
*** yangyapeng has quit IRC13:17
*** yangyapeng has joined #openstack-helm13:20
*** julim has joined #openstack-helm13:24
*** yamamoto has joined #openstack-helm13:28
*** yamamoto has quit IRC13:33
*** jaypipes has joined #openstack-helm13:35
*** larainema has left #openstack-helm14:02
osh-chatbot4<abhishekjaiswal> can any help me with this Kubernetes Node DNS Resolution¶  my `/etc/resolv.conf` is like  nameserver 192.168.1.10 nameserver 10.233.0.3 nameserver # search pramati.com openstack.svc.cluster.local svc.cluster.local cluster.local options ndots:5 timeout:1 attempts:1  where `10.233.0.3` is my kube dns14:28
*** yamamoto has joined #openstack-helm14:28
osh-chatbot4<abhishekjaiswal> is that correct ?14:28
*** yamamoto has quit IRC14:32
*** yamamoto has joined #openstack-helm14:46
*** yamamoto has quit IRC14:46
portdirectThat looks backward14:49
portdirectIt will work down the list with those options set14:50
portdirectSo put the k8s nameserver at the top ;)14:50
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Add Grafana chart to OSH infra  https://review.openstack.org/52360514:54
*** coolsvap has quit IRC15:25
*** d|k has quit IRC15:37
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Add elasticsearch exporter  https://review.openstack.org/52363415:38
*** yamamoto has joined #openstack-helm15:46
*** yamamoto has quit IRC15:54
*** XueFeng has quit IRC15:57
osh-chatbot4<abhishekjaiswal> @portdirect15:57
osh-chatbot4<abhishekjaiswal> @portdirect i had added those and resolv.conf.d/base file so it is coming autmatically15:58
*** petevg has joined #openstack-helm16:00
*** dims has quit IRC16:09
*** dims has joined #openstack-helm16:10
*** openstackgerrit has quit IRC16:19
*** MarkBaker has quit IRC16:33
*** openstackgerrit has joined #openstack-helm16:34
openstackgerritPete Birley proposed openstack/openstack-helm-infra master: DNM: WIP: Dep checking updates  https://review.openstack.org/53095816:34
*** mdih has quit IRC16:35
*** tlam has quit IRC16:35
*** tlam has joined #openstack-helm16:35
openstackgerritGage Hugo proposed openstack/openstack-helm master: Fix periods / Retrigger docs gate  https://review.openstack.org/53096216:39
osh-chatbot4<kelvin633> I updated OSH to use `ubuntu-binary` kolla 3.0.3 images and it fails on `ERROR neutron PasteAppNotFound: Could not load paste app 'neutron' from /etc/neutron/api-paste.ini`. Is it advisable to use “binary” images, since “source” is the default?16:43
portdirectkelvin633: I'd use the source images16:45
portdirectas the binaries have a mix of uids16:45
portdirectso you'll need to do some work to match them with the security context set in the k8s pods16:46
*** MarkBaker has joined #openstack-helm16:46
osh-chatbot4<kelvin633> oh, okay. I did see a UID mismatch error prior to the above. Is there a plan to make binary images default in the future?16:47
portdirectno16:47
portdirectwe will be moving to LOCI images before the end of the month16:47
osh-chatbot4<kelvin633> LOCI…. ?16:48
portdirectbut we will have a directory of overides for the common images, which should include Kolla-Binary16:48
portdirectLOCI :) https://github.com/openstack/loci16:48
osh-chatbot4<kelvin633> Ok. Let me digest that. Thanks!16:48
*** krypto has quit IRC16:53
osh-chatbot4<kelvin633> @portdirect how long will source images be supported when LOCI becomes default?17:02
*** felipemonteiro has joined #openstack-helm17:06
*** yamamoto has joined #openstack-helm17:16
*** MarkBaker has quit IRC17:16
*** yamamoto has quit IRC17:21
mwynneportdirect: Should GLANCE_BACKEND be radosgw or pvc for the NFS AIO deployment?17:21
*** tlam has quit IRC17:26
*** tlam has joined #openstack-helm17:26
*** yamamoto has joined #openstack-helm17:29
*** yamamoto has quit IRC17:34
openstackgerritMerged openstack/openstack-helm master: Fix periods / Retrigger docs gate  https://review.openstack.org/53096217:39
*** yamamoto has joined #openstack-helm17:59
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Add Grafana chart to OSH infra  https://review.openstack.org/52360518:03
*** yamamoto has quit IRC18:04
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Fix alertmanager serviceaccountname reference  https://review.openstack.org/53098818:09
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Include ingress entries in manifests key for kibana  https://review.openstack.org/53098918:14
*** yamamoto has joined #openstack-helm18:16
*** yamamoto has quit IRC18:16
*** ChanServ sets mode: +r 18:20
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Include ingress entries in manifests key for kibana  https://review.openstack.org/53098918:22
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Add elasticsearch exporter  https://review.openstack.org/52363418:43
*** felipemonteiro has quit IRC18:44
*** yamamoto has joined #openstack-helm18:46
mwynneportdirect: The mtu of my docker bridge is 1450, but eth0 in all the containers is set to 1500. Can I modify that somewhere? I need a smaller MTU for my configuration.18:50
mwynneOr, anyone :)18:53
*** yamamoto has quit IRC18:54
srwilkersmwynne: i think you can specify it via a kubelet flag18:57
srwilkerslet me poke around a minute18:59
srwilkersill admit that's a bit out of the range of my experience, but seems it can be configured in calico via calicoctl or through fenix env variables19:02
srwilkershttps://docs.projectcalico.org/v2.1/usage/configuration/mtu19:02
mwynnesrwilkers: There's an MTU config in here: calico/templates/configmap-calico-config.yaml19:05
mwynneIs that the relevant config? Any idea?19:06
srwilkersyeah, that appears to be the one19:07
mwynnesrwilkers: I'd assume I need to start from scratch if I've already deployed up to glance before stuff started failing.19:08
mwynneI probably can't just re-deploy calico hey?19:09
srwilkersyeah, configuring that route would require redeploying calico.  the alternative would be to use calicoctl instead.  for any deployments to get the update, they'd need to be restarted.  it's up to you19:09
mwynnesrwilkers: Is `calicoctl config set --raw=felix IpInIpMtu 1480` (or whatever MTU I need), sufficient?19:10
openstackgerritTin Lam proposed openstack/openstack-helm master: Add domain specific driver support  https://review.openstack.org/52984419:11
mwynneis `--raw=felix IpInIpMtu` the flag I need? Any clue?19:11
mwynnesrwilkers: And where can I run that command? :S19:15
osh-chatbot4<kelvin633> @portdirect is there a branch of OSH that uses LOCI?19:24
*** tlam has quit IRC19:33
*** yangyapeng has quit IRC19:44
*** yamamoto has joined #openstack-helm19:56
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Add elasticsearch exporter  https://review.openstack.org/52363419:59
openstackgerritdave kormann proposed openstack/openstack-helm master: Ceph OSD disk targeting  https://review.openstack.org/52750520:00
*** portdirect has quit IRC20:50
*** portdirect has joined #openstack-helm20:50
*** ChanServ sets mode: +o portdirect20:50
openstackgerritSteve Wilkerson proposed openstack/openstack-helm master: WIP - Add docs for LMA services  https://review.openstack.org/52848822:08
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Add rabbitmq-exporter  https://review.openstack.org/52347622:42
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Add elasticsearch exporter  https://review.openstack.org/52363422:46
mwynneportdirect: Is radosgw the wrong GLANCE_BACKEND for NFS AIO?22:53
portdirectyeah :/22:53
portdirectsorry I missed that22:53
portdirectcan you do "pvc"22:53
portdirectbut you have mariadb up?22:54
portdirectif so w000t! I was getting worried22:54
* portdirect hopes he did not just jinx it.22:54
mwynneI ran it before bed last night, and it failed to come up in the default timeout, but it was up and running when I woke up :S22:54
portdirectoh wow22:54
portdirectwell lets take that as a win22:55
mwynneI think I have other issues with IO load or something. I'm seeing much longer deployments with my puppet scripts too. Frustrating stuff :(22:55
mwynneportdirect: Oh, the issue I ran into this morning was with MTU22:56
mwynneI need to change the MTU of the docker bridge, and all the interfaces in the continers, since this is running in an openstack instance with an MTU of 1450 (for vxlan)22:57
mwynneportdirect: Is that something I can change in: calico/templates/configmap-calico-config.yaml?22:58
portdirecti must confess I'm not sure off the top of my head22:58
mwynneRight now eth0 in every container has an MTU of 1500, so glance failed because it couldn't curl the cirros image :(22:58
mwynneThis is killing me haha22:58
openstackgerritRenis Makadia proposed openstack/openstack-helm-infra master: WIP-Loopback device setup/configuration  https://review.openstack.org/53030223:23
openstackgerritPete Birley proposed openstack/openstack-helm master: Dev-Deploy: move to k8s DNS addr for private network  https://review.openstack.org/53103423:27
openstackgerritSteve Wilkerson proposed openstack/openstack-helm-infra master: Add Grafana chart to OSH infra  https://review.openstack.org/52360523:27
openstackgerritPete Birley proposed openstack/openstack-helm master: Dev-Deploy: move to k8s DNS addr for private network  https://review.openstack.org/53103423:27

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!