| *** goldyfruit_ has joined #openstack-containers | 00:03 | |
| *** goldyfruit_ has quit IRC | 00:56 | |
| guilhermesp | yeah I was going to do that brtknr but just saw the the PR is in the gate already :) | 01:02 |
|---|---|---|
| openstackgerrit | Merged openstack/magnum master: k8s_atomic: Run all syscontainer with podman https://review.opendev.org/685749 | 01:10 |
| *** ykarel|away has joined #openstack-containers | 01:45 | |
| *** goldyfruit_ has joined #openstack-containers | 02:05 | |
| *** ricolin has joined #openstack-containers | 02:21 | |
| *** goldyfruit_ has quit IRC | 02:35 | |
| *** ramishra has joined #openstack-containers | 02:37 | |
| guilhermesp | brtknr: do we expect that, with that change, we will keep compatibility with later k8s versions? https://github.com/openstack/magnum/commit/3674b3617a770bd71d09e23137ff96f90eb1241a | 02:55 |
| guilhermesp | just asking because we already have cluster templates set to deploy v1.14 and v1.15 and some existent clusters running | 02:55 |
| guilhermesp | uowww just applied that commit | 03:38 |
| guilhermesp | v1.16.0 created sucessfully and faster then usually I used to see | 03:38 |
| guilhermesp | 7 minutes agains an average of 15 minutes | 03:39 |
| *** ramishra has quit IRC | 03:43 | |
| *** udesale has joined #openstack-containers | 03:55 | |
| *** ykarel|away has quit IRC | 04:01 | |
| *** ykarel|away has joined #openstack-containers | 04:22 | |
| *** ykarel|away is now known as ykarel | 04:22 | |
| guilhermesp | and conformance ok https://github.com/cncf/k8s-conformance/pull/742 | 04:50 |
| *** ramishra has joined #openstack-containers | 04:59 | |
| guilhermesp | and yeah, don't take into account what I asked above about compatibility, it clearly works with older version as faster as the cluster deployment with v1.16 | 04:59 |
| *** dave-mccowan has quit IRC | 05:12 | |
| *** namrata has joined #openstack-containers | 06:39 | |
| *** pcaruana has joined #openstack-containers | 07:10 | |
| brtknr | guilhermesp: sorry I was sleeping! Thanks for testing, good to hear it’s faster, it’s a single hyperkube image vs multiple docker images previously.... and they’re all published by google so must be good :) with regards to api compatibility, I assume we will need to make necessary adaptations in v1.17 but should be good for 1.16 | 07:26 |
| brtknr | Because they’re google images, i bet v1.16.1 is already available too | 07:26 |
| *** namrata has quit IRC | 07:31 | |
| *** ivve has joined #openstack-containers | 07:41 | |
| *** ykarel is now known as ykarel|lunch | 07:44 | |
| *** ramishra_ has joined #openstack-containers | 08:27 | |
| *** ramishra has quit IRC | 08:29 | |
| *** ykarel|lunch is now known as ykarel | 08:34 | |
| brtknr | flwang: strigazi: meeting in 20 mins? | 08:38 |
| *** flwang1 has joined #openstack-containers | 08:43 | |
| strigazi | yes | 08:43 |
| flwang1 | hello | 08:43 |
| brtknr | hi both :) | 08:45 |
| brtknr | guilhermesp: dioguerra: you guys also there? | 08:45 |
| brtknr | also ttsiouts but he's offline on irc | 08:46 |
| brtknr | jakeyip: ^ | 08:46 |
| brtknr | might write an irc bot to notify all the people active in the channel in the past week | 08:47 |
| brtknr | :P | 08:47 |
| flwang1 | :) | 08:51 |
| jakeyip | I'm here | 08:55 |
| strigazi | flwang1: you start the meeting? | 08:59 |
| flwang1 | #startmeeting magnum | 08:59 |
| openstack | Meeting started Wed Oct 9 08:59:33 2019 UTC and is due to finish in 60 minutes. The chair is flwang1. Information about MeetBot at http://wiki.debian.org/MeetBot. | 08:59 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 08:59 |
| *** openstack changes topic to " (Meeting topic: magnum)" | 08:59 | |
| openstack | The meeting name has been set to 'magnum' | 08:59 |
| flwang1 | #topic roll call | 08:59 |
| *** openstack changes topic to "roll call (Meeting topic: magnum)" | 08:59 | |
| flwang1 | o/ | 08:59 |
| strigazi | o/ | 08:59 |
| flwang1 | brtknr: ? | 09:01 |
| brtknr | oo/ | 09:01 |
| flwang1 | #topic fcos driver | 09:02 |
| *** openstack changes topic to "fcos driver (Meeting topic: magnum)" | 09:02 | |
| flwang1 | strigazi: do you have give us an update? | 09:02 |
| flwang1 | s/have/wanna | 09:02 |
| strigazi | the patch is in working state, I want to only add UTs | 09:02 |
| jakeyip | o/ | 09:02 |
| strigazi | works with 1.15.x and 1.16.x | 09:02 |
| brtknr | strigazi: when I tested it yesterday, I was havving issues with the $ssh_cmd | 09:03 |
| strigazi | The only thing I want to check before merging is selinux | 09:03 |
| strigazi | brtknr: works for me | 09:03 |
| strigazi | brtknr: can you be more specific? | 09:03 |
| brtknr | why do we use $ssh_cmd in some places and not others | 09:03 |
| strigazi | is this the problem? | 09:04 |
| brtknr | + ssh -F /srv/magnum/.ssh/config root@localhost openssl genrsa -out /etc/kubernetes/certs/kubelet.key | 09:04 |
| brtknr | ssh: connect to host 127.0.0.1 port 22: Connection refused | 09:04 |
| strigazi | sshd down? | 09:04 |
| brtknr | this is in the worker node after which it fails to join the cluster | 09:04 |
| brtknr | when I tried the command manually, it worked | 09:05 |
| brtknr | also when i restarted the heat-container-agent manually, it joined the cluster | 09:05 |
| strigazi | I have After=sshd.service | 09:05 |
| strigazi | I can add requires too | 09:05 |
| brtknr | wondering if heat-container-agent is racing against sshd | 09:05 |
| strigazi | transient error? I never saw this after adding After=sshd.service | 09:06 |
| flwang1 | strigazi: it would be nice to have After=sshd.service | 09:06 |
| strigazi | we do have it!!! | 09:06 |
| strigazi | https://review.opendev.org/#/c/678458/8/magnum/drivers/k8s_fedora_coreos_v1/templates/user_data.json@75 | 09:06 |
| flwang1 | then good | 09:06 |
| flwang1 | i haven't got time to test your new patch | 09:07 |
| strigazi | hm, no it only in configure-agent-env.service | 09:07 |
| strigazi | but | 09:07 |
| strigazi | heat-container-agent.service has After=network-online.target configure-agent-env.service | 09:08 |
| strigazi | and configure-agent-env.service has After=sshd.service | 09:08 |
| flwang1 | so it would work | 09:08 |
| strigazi | I'll add Requires as well | 09:08 |
| strigazi | anyway, that is a detail | 09:09 |
| flwang1 | strigazi: re UT, i'm happy with having limited UT for this new driver given the gate is closing | 09:09 |
| strigazi | the important parts are two | 09:09 |
| strigazi | one, the iginition patch is no working :( | 09:10 |
| flwang1 | ??? | 09:10 |
| flwang1 | your mean my heat patch? | 09:10 |
| strigazi | I missed testing the last PS | 09:10 |
| strigazi | yes | 09:10 |
| flwang1 | why? | 09:10 |
| flwang1 | what's the problem? | 09:10 |
| strigazi | https://review.opendev.org/#/c/683723/5..13/heat/engine/clients/os/nova.py@454 | 09:10 |
| flwang1 | is it a regression issue? | 09:10 |
| strigazi | /var/lib/os-collect-config/local-data must be a direcroty | 09:11 |
| strigazi | not a file | 09:11 |
| strigazi | see https://github.com/openstack/os-collect-config/blob/master/os_collect_config/local.py#L70 | 09:11 |
| flwang1 | shit | 09:11 |
| flwang1 | do you have a fix ? | 09:12 |
| strigazi | not a proper one | 09:12 |
| strigazi | if we write a file in /var/lib/os-collect-config/local-data our agent doesn't work still | 09:12 |
| *** ttsiouts has joined #openstack-containers | 09:13 | |
| strigazi | but at least os-collect-config is not throwing an exception | 09:13 |
| ttsiouts | o/ sorry I was late.. | 09:13 |
| strigazi | my fix was to just copy the file to /var/lib/cloud/data/cfn-init-data" | 09:14 |
| strigazi | I don't know why the local collector is not working | 09:14 |
| flwang1 | can we propose a patch to still use the two 'old' file paths? | 09:14 |
| strigazi | yes a milllion times | 09:14 |
| strigazi | or the heat team can help us using the local collector which will require us to patch our agent | 09:16 |
| strigazi | so action is to patch heat again with the old files? | 09:16 |
| flwang1 | ok, i will take this | 09:16 |
| flwang1 | i will try the current way first | 09:17 |
| strigazi | FYI, FCOS team is happy to patch ignition as well | 09:17 |
| strigazi | I'm testing that too | 09:17 |
| strigazi | they wrote the patch aleary | 09:17 |
| flwang1 | ok, let's do it in parallel | 09:17 |
| strigazi | they wrote the patch already | 09:17 |
| flwang1 | just in case | 09:17 |
| strigazi | yeap | 09:17 |
| strigazi | and the second part for fcos driver is: | 09:18 |
| strigazi | will we enable selinux? conformance pass (I think), I'll post the result to be sure | 09:18 |
| flwang1 | if we can enable selinux, why not? | 09:18 |
| strigazi | I don't know how it will work with the storage plugins | 09:19 |
| strigazi | flwang1: you can test cinder if you use it | 09:19 |
| strigazi | not sure if cinder works with 1.16 | 09:20 |
| flwang1 | strigazi: you mean test cinder with 1.16 with selinxu enabled? or they are 2 different cases | 09:20 |
| strigazi | one | 09:21 |
| strigazi | test cinder with 1.16 with selinxux enabled | 09:21 |
| strigazi | does it work with 1.15? | 09:21 |
| flwang1 | iirc, in 1.16, k8s won't use the build-in cinder driver, right? | 09:21 |
| strigazi | excellent :) | 09:22 |
| strigazi | so nothing to test? | 09:22 |
| strigazi | will we use csi? | 09:22 |
| flwang1 | i don't know, i need to confirm with CPO team | 09:22 |
| flwang1 | i will let you guys know | 09:22 |
| strigazi | we get off-topic, but leave a comment in the review for cinder | 09:23 |
| strigazi | we go with selinux if conformance passes. | 09:23 |
| strigazi | that's it for me | 09:23 |
| flwang1 | cool, thanks | 09:23 |
| strigazi | next? NGs? | 09:25 |
| flwang1 | ttsiouts: ng? | 09:26 |
| ttsiouts | flwang1: strigazi: sure | 09:26 |
| flwang1 | i can see there are still several patches for NG, are we still going to get them in train? | 09:26 |
| strigazi | we need them, they are fixes really | 09:27 |
| ttsiouts | flwang1: yes | 09:27 |
| ttsiouts | flwang1: the NG upgrade is the main thing I wanted to talk about | 09:27 |
| flwang1 | ok | 09:27 |
| ttsiouts | the WIP patch https://review.opendev.org/#/c/686733/ does what it is supposed to do | 09:28 |
| ttsiouts | it upgrades the given NG | 09:29 |
| ttsiouts | the user has to be really careful as to what labels to use | 09:30 |
| flwang1 | ttsiouts: ok, user can upgrade a specific ng with the current upgrade command, right? | 09:30 |
| ttsiouts | flwang1: exactly | 09:30 |
| ttsiouts | flwang1: by defining --nodegroup <nodegroup> | 09:30 |
| flwang1 | what do you mean 'what labels to use'? | 09:30 |
| ttsiouts | flwang1: for example if the availability zone label is in the cluster template, and it is different than the one set in the NG | 09:31 |
| ttsiouts | flwang1: it will cause the nodes of the NG to be rebuilt | 09:32 |
| brtknr | ttsiouts: i thought all other labels get ignored apart from kube_tag? | 09:32 |
| strigazi | the heat_container_agent_tag can be destructive too | 09:32 |
| flwang1 | brtknr: it should be, that's why i asked | 09:32 |
| ttsiouts | hmm.. I tried to remove the labels that could cause such things here: https://review.opendev.org/#/c/686733/1/magnum/drivers/k8s_fedora_atomic_v1/driver.py@121 | 09:33 |
| ttsiouts | flwang1: brtknr: I am not sure they are ignored. But I can test again | 09:34 |
| flwang1 | hmm... i'd like to test as well to understand it better | 09:34 |
| strigazi | the happy path scenarios work great | 09:35 |
| ttsiouts | strigazi: yes | 09:35 |
| strigazi | corner cases like changing one of the labels ttsiouts linked in the patch can be descrtuctive | 09:35 |
| strigazi | or a bit rebuildy | 09:35 |
| ttsiouts | yes this is why I wanted to raise this here. | 09:36 |
| ttsiouts | so we are on the same page | 09:36 |
| flwang1 | ok | 09:36 |
| flwang1 | thanks | 09:36 |
| flwang1 | anything else? | 09:38 |
| brtknr | I would rather prevent upgrade taking place if the specific subset of labels do not match | 09:38 |
| strigazi | and we need these for tain | 09:38 |
| strigazi | and we need these for train | 09:38 |
| brtknr | e.g. force those specific labels to match | 09:38 |
| brtknr | ttsiouts: ^ | 09:38 |
| strigazi | for U yes | 09:38 |
| strigazi | for T a bit late? | 09:38 |
| ttsiouts | brtknr: the logic is that the non-default NGs get upgraded to match the cluster | 09:40 |
| ttsiouts | brtknr: there is a validation in the api that checks that the user provided the same CT as the cluster has (only for non-default NGs) | 09:41 |
| brtknr | strigazi: Yes, I am happy to develop it further later... perhaps add accompanying notes for bits that are "hacks" for this release... we dont want code to silently ignore... | 09:41 |
| ttsiouts | brtknr: If we have one CT the labels cannot match to all cluster NGs | 09:42 |
| brtknr | ttsiouts: i meant, matching label during upgrades | 09:43 |
| brtknr | you have CT-A-v1 which has AZ-A | 09:43 |
| brtknr | then you want that to upgrade nodegroup from CT-A-v1 to CT-A-v2 which has AZ-B | 09:44 |
| brtknr | I'd rather the upgrade wasnt allowed in this situation | 09:44 |
| brtknr | as it might be error on part of the user... | 09:44 |
| strigazi | this doesn't make sense | 09:44 |
| strigazi | I have ng-1-az-A and ng-1-az-B | 09:45 |
| strigazi | I won't to upgrade both | 09:45 |
| strigazi | I want to upgrade both | 09:45 |
| strigazi | the CT in the cluster can be only one. | 09:45 |
| strigazi | the rule above would never allow me to upgrade one of the two NGs | 09:46 |
| ttsiouts | brtknr: check here: https://review.opendev.org/#/c/686733/1/magnum/api/controllers/v1/cluster_actions.py@153 | 09:47 |
| strigazi | for U we will put it in the NG spec with details. | 09:47 |
| strigazi | for T we can say that only kube_tag will be taken into account | 09:47 |
| strigazi | sounds reasonable? | 09:48 |
| brtknr | im happy with only kube_tag for T | 09:48 |
| strigazi | so we add only kube_tag in https://review.opendev.org/#/c/686733 ? | 09:49 |
| strigazi | ttsiouts: brtknr flwang1 ^^ | 09:49 |
| flwang1 | i'm ok | 09:49 |
| ttsiouts | this is the safest for now | 09:49 |
| strigazi | and safeest | 09:49 |
| strigazi | and safest | 09:49 |
| brtknr | yes, does it still require stripping things out? | 09:49 |
| brtknr | or would it be better to only select kube_tag during upgrade? | 09:50 |
| ttsiouts | brtknr: we could use the code that checks the validity of the versions and pass only the kube_tag | 09:51 |
| brtknr | it might be better to use "upgradeable" parts rather than "skip these labels" | 09:51 |
| brtknr | it might be better to use "upgradeable labels" rather than "skip these labels" | 09:51 |
| brtknr | for clarity | 09:51 |
| brtknr | we can expand the list of "upgradeable labels" as we are able to upgrade more things | 09:52 |
| flwang1 | we're running out time | 09:52 |
| brtknr | does that make sense? | 09:52 |
| strigazi | let's keep it simple for T | 09:52 |
| strigazi | pass onlt kube_tag, end of story | 09:52 |
| strigazi | pass only kube_tag, end of story | 09:53 |
| strigazi | add it in reno too, and we are clear | 09:53 |
| brtknr | sounds good | 09:53 |
| brtknr | lets start backporting things | 09:54 |
| flwang1 | #topic backport for train | 09:54 |
| *** openstack changes topic to "backport for train (Meeting topic: magnum)" | 09:54 | |
| brtknr | bfv+ng+podman+fcos | 09:54 |
| strigazi | basically current master plus ng and fcos? | 09:55 |
| brtknr | yes | 09:55 |
| strigazi | flwang1: sounds good? | 09:56 |
| flwang1 | yep, good for me | 09:57 |
| strigazi | anything else for the topic/meeting? | 09:58 |
| flwang1 | i'm good | 09:58 |
| flwang1 | actually, i'm sleepy ;) | 09:58 |
| brtknr | thats all | 09:58 |
| strigazi | on time | 09:58 |
| flwang1 | cool, thank you, guys | 09:59 |
| flwang1 | #endmeeting | 09:59 |
| *** openstack changes topic to "OpenStack Containers Team | Meeting: every Wednesday @ 9AM UTC | Agenda: https://etherpad.openstack.org/p/magnum-weekly-meeting" | 09:59 | |
| openstack | Meeting ended Wed Oct 9 09:59:32 2019 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 09:59 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/magnum/2019/magnum.2019-10-09-08.59.html | 09:59 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/magnum/2019/magnum.2019-10-09-08.59.txt | 09:59 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/magnum/2019/magnum.2019-10-09-08.59.log.html | 09:59 |
| brtknr | strigazi: are you planning to fix fcos flannel for T? | 10:00 |
| flwang1 | brtknr: he did i think | 10:00 |
| flwang1 | see the new patchset | 10:00 |
| strigazi | brtknr: I did, plus selinux support | 10:01 |
| strigazi | basically run flannel as root | 10:01 |
| flwang1 | i'm off now, ttyl | 10:01 |
| * strigazi goes for lunch | 10:01 | |
| brtknr | flwang1: sleep well! | 10:01 |
| *** ttsiouts has quit IRC | 10:11 | |
| *** lpetrut has joined #openstack-containers | 10:39 | |
| *** rcernin has quit IRC | 10:48 | |
| *** udesale has quit IRC | 11:22 | |
| *** ramishra_ has quit IRC | 11:30 | |
| *** ttsiouts has joined #openstack-containers | 11:32 | |
| *** dave-mccowan has joined #openstack-containers | 11:49 | |
| *** ramishra has joined #openstack-containers | 11:49 | |
| *** goldyfruit_ has joined #openstack-containers | 12:12 | |
| *** goldyfruit___ has joined #openstack-containers | 12:14 | |
| *** goldyfruit_ has quit IRC | 12:17 | |
| *** goldyfruit___ has quit IRC | 12:19 | |
| *** ttsiouts has quit IRC | 12:24 | |
| *** ttsiouts has joined #openstack-containers | 12:25 | |
| *** ttsiouts_ has joined #openstack-containers | 12:28 | |
| *** ttsiouts has quit IRC | 12:28 | |
| guilhermesp | ah sorry brtknr I'm now :) | 13:03 |
| *** lsimngar has joined #openstack-containers | 13:06 | |
| *** lsimngar has quit IRC | 13:08 | |
| *** goldyfruit___ has joined #openstack-containers | 13:28 | |
| *** goldyfruit___ has quit IRC | 13:29 | |
| openstackgerrit | Theodoros Tsioutsias proposed openstack/magnum master: Optimize cluster list api https://review.opendev.org/687556 | 13:30 |
| *** pcaruana has quit IRC | 13:31 | |
| *** goldyfruit has joined #openstack-containers | 13:32 | |
| *** pcaruana has joined #openstack-containers | 13:33 | |
| *** ianychoi has quit IRC | 13:41 | |
| *** udesale has joined #openstack-containers | 13:45 | |
| *** sapd1_ has joined #openstack-containers | 13:53 | |
| *** johanssone has quit IRC | 13:56 | |
| *** ivve has quit IRC | 13:56 | |
| *** sapd1 has quit IRC | 13:56 | |
| *** ioni has quit IRC | 13:56 | |
| *** _ioni has joined #openstack-containers | 13:56 | |
| *** ivve has joined #openstack-containers | 13:56 | |
| *** openstackstatus has quit IRC | 13:58 | |
| *** johanssone has joined #openstack-containers | 13:58 | |
| *** _ioni is now known as ioni | 13:59 | |
| *** kgz has quit IRC | 14:00 | |
| *** ioni has left #openstack-containers | 14:00 | |
| *** ioni has joined #openstack-containers | 14:00 | |
| *** kgz has joined #openstack-containers | 14:07 | |
| *** lpetrut has quit IRC | 14:09 | |
| brtknr | strigazi: strange, I can create v1.16.1 clusters but not v1.16.0 | 14:10 |
| brtknr | for coreos | 14:10 |
| brtknr | just repeating in case it was a fluke | 14:10 |
| *** udesale has quit IRC | 14:41 | |
| strigazi | this can't be | 14:48 |
| strigazi | brtknr: try with calico, not flannel, just in case | 14:48 |
| *** udesale has joined #openstack-containers | 14:54 | |
| brtknr | strigazi: https://seashells.io/p/7H5RTJKM | 15:03 |
| brtknr | strigazi: okay second time, calico works fine | 15:03 |
| *** ivve has quit IRC | 15:03 | |
| brtknr | flannel has pods in kube-system namespace stuck in ContainerCreating phase | 15:03 |
| *** lsimngar_ has joined #openstack-containers | 15:04 | |
| brtknr | and this is what I see what I describe them https://seashells.io/p/skmduwdt | 15:04 |
| brtknr | I suspect I'm not using the latest PS | 15:04 |
| brtknr | even though I am... | 15:04 |
| strigazi | brtknr: it is selinux now, before it was kernel module. | 15:04 |
| brtknr | I'm testing sha 20e0803ddf6edb4f7b7419b0cf7d12e5e6614b8b | 15:05 |
| strigazi | I mentioned it this morning, I haven't fixed selinux yet | 15:05 |
| strigazi | I want to find the proper way | 15:05 |
| strigazi | I can send you a patch with something that works | 15:05 |
| strigazi | one sec | 15:05 |
| brtknr | strigazi: Oh right, I thought selinux was fixed, my misunderstanding then | 15:07 |
| *** udesale has quit IRC | 15:14 | |
| *** lsimngar_ has quit IRC | 15:16 | |
| *** ramishra has quit IRC | 15:21 | |
| *** lpetrut has joined #openstack-containers | 15:21 | |
| *** ryn_eq has joined #openstack-containers | 15:22 | |
| *** ykarel is now known as ykarel|afk | 15:26 | |
| strigazi | brtknr: the first flannel issue was fixed. Now there is a second one | 15:26 |
| openstackgerrit | Spyros Trigazis proposed openstack/magnum master: [WIP] Support Fedora CoreOS 30 https://review.opendev.org/678458 | 15:29 |
| strigazi | brtknr: this should work ^^ if I don't have any typos | 15:29 |
| strigazi | brtknr: Ill try to make it work without running as root. | 15:30 |
| * strigazi is going home | 15:30 | |
| *** ttsiouts_ has quit IRC | 15:33 | |
| *** ttsiouts has joined #openstack-containers | 15:33 | |
| *** lpetrut has quit IRC | 15:35 | |
| *** ttsiouts has quit IRC | 15:37 | |
| *** openstackgerrit has quit IRC | 15:52 | |
| *** ykarel|afk is now known as ykarel | 16:35 | |
| *** ramishra has joined #openstack-containers | 16:37 | |
| strigazi | brtknr: have you tried what I sent you? it works fine | 16:53 |
| strigazi | brtknr: have you tried this? https://review.opendev.org/#/c/678458/9/magnum/drivers/common/templates/kubernetes/fragments/flannel-service.sh | 16:53 |
| brtknr | strigazi: i have no doubts it will work, as it is allow everything policy... | 16:54 |
| strigazi | brtknr: even before it had NET_ADMIN so not much changed... | 16:56 |
| strigazi | brtknr: you said in atomic it works. I want to point out that in atomic selinux is in permissive. | 16:58 |
| brtknr | strigazi: Hmm I didnt realise | 16:59 |
| brtknr | strigazi: I’m updating my dev stack env to 18.04, will test your change when it’s ready | 17:05 |
| strigazi | ubuntu 18.04? | 17:06 |
| *** ramishra has quit IRC | 17:27 | |
| *** ykarel_ has joined #openstack-containers | 17:32 | |
| *** ykarel_ has quit IRC | 17:33 | |
| *** ykarel has quit IRC | 17:35 | |
| *** ricolin has quit IRC | 17:41 | |
| brtknr | strigazi: 1yes | 17:59 |
| brtknr | strigazi: been using 16.05 so far | 18:00 |
| brtknr | strigazi: been using 16.04 so far | 18:00 |
| strigazi | brtknr: https://github.com/coreos/flannel/issues/1202 | 18:33 |
| *** flwang1 has quit IRC | 18:46 | |
| *** pcaruana has quit IRC | 19:07 | |
| *** openstackgerrit has joined #openstack-containers | 19:53 | |
| openstackgerrit | Spyros Trigazis proposed openstack/magnum master: [WIP] Support Fedora CoreOS 30 https://review.opendev.org/678458 | 19:53 |
| strigazi | brtknr: flwang for flannel this works, although, spc_t mean super privileged container | 19:54 |
| *** pcaruana has joined #openstack-containers | 20:33 | |
| flwang | but the super privileged permission only for the flannel container, right? | 20:40 |
| flwang | strigazi: ^ | 20:40 |
| *** pcaruana has quit IRC | 20:40 | |
| *** lpetrut has joined #openstack-containers | 20:41 | |
| flwang | strigazi: are you around? | 20:51 |
| *** lpetrut has quit IRC | 20:52 | |
| *** lpetrut has joined #openstack-containers | 21:13 | |
| *** lpetrut has quit IRC | 21:30 | |
| *** goldyfruit has quit IRC | 21:32 | |
| *** goldyfruit has joined #openstack-containers | 21:38 | |
| brtknr | flwang: strigazi: i suppose its better than having selinux set to permissive | 21:56 |
| *** rcernin has joined #openstack-containers | 22:14 | |
| brtknr | strigazi: nonetheless it’s working for me | 22:57 |
| brtknr | flwang: is it working for you? | 22:57 |
| *** flwang has quit IRC | 22:57 | |
| *** openstackstatus has joined #openstack-containers | 23:18 | |
| *** ChanServ sets mode: +v openstackstatus | 23:18 | |
| *** goldyfruit has quit IRC | 23:42 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!