!HVdFZNeOFWTSSsZOgu:matrix.org

Kubernetes

556 Members
Container orchestration from Google http://kubernetes.io/ - Explore rooms related to Kubernetes using the #kubernetes:shivering-isles.com community!91 Servers

Load older messages


SenderMessageTime
11 Apr 2024
@soma-kurisu:matrix.orgKurisu joined the room.13:26:27
@ravilerochus:matrix.orgravi le rochus joined the room.14:30:14
12 Apr 2024
@psil0cybin:matrix.orgPsil0 changed their profile picture.10:52:44
@leroyshirto:matrix.orgElzonk set a profile picture.12:10:50
15 Apr 2024
@rhouben:vac.systemec.nlRens Houben I'm trying to set up a k8s cluster behind a proxy, which means it has no external internet connectivity that doesn't go via the proxy, and therefore no default routes. How do I tell it to shut the entire fuck up about Apr 15 10:26:22 curly kubelet[14426]: E0415 10:26:22.674635 14426 kubelet_node_status.go:701] "Failed to set some node status fields" err="can't get ip address of node curly. error: no default routes found in \"/proc/net/route\" or \"/proc/net/ipv6_route\"" node="curly" and just use the actual address I specified in the config? 08:33:54
@mortero:matrix.org@mortero:matrix.org joined the room.23:54:23
@mortero:matrix.org@mortero:matrix.org left the room.23:54:37
16 Apr 2024
@jokeyrhyme:matrix.orgjokeyrhyme
In reply to @rhouben:vac.systemec.nl
I'm trying to set up a k8s cluster behind a proxy, which means it has no external internet connectivity that doesn't go via the proxy, and therefore no default routes. How do I tell it to shut the entire fuck up about Apr 15 10:26:22 curly kubelet[14426]: E0415 10:26:22.674635 14426 kubelet_node_status.go:701] "Failed to set some node status fields" err="can't get ip address of node curly. error: no default routes found in \"/proc/net/route\" or \"/proc/net/ipv6_route\"" node="curly" and just use the actual address I specified in the config?
This guide suggests keeping the default route but pointing it at a dummy network interface 
Not sure if that's how most others set up an air-gapped cluster: https://kubernetes.io/blog/2023/10/12/bootstrap-an-air-gapped-cluster-with-kubeadm/
04:20:47
@iamak:matrix.orgएके changed their profile picture.19:17:14
@thompson_plug:matrix.org@thompson_plug:matrix.orgRedacted or Malformed Event23:43:55
17 Apr 2024
@rhouben:vac.systemec.nlRens HoubenSomebody come and boot the spam bot, please.08:08:58
@rhouben:vac.systemec.nlRens Houben Sheogorath: Spambot in the room. 08:09:37
@Criminal:matrix.orgmorguldir (Old) joined the room.08:10:12
@sheogorath:shivering-isles.comSheogorath banned @thompson_plug:matrix.org@thompson_plug:matrix.org (spam).08:43:37
@timoschuetz:matrix.orgTimo joined the room.09:27:40
@rhouben:vac.systemec.nlRens Houben

Is it possible to suppress extraneous message output during kubeadm init ?

I'm trying to automate the entire process in an ansible playbook, and I need some way to extract the control panel join command as well as the worker node join command from the kubeadm init output, or at least the values for --token, --discovery-token-ca-cert-hash and --certificate-key so I can form the command to be used on the various other nodes.

12:32:11
@rhouben:vac.systemec.nlRens Houben
kubeadm init phase upload-certs --upload-certs --skip-headers
[upload-certs] Storing the certificates in Secret "kubeadm-certs" in the "kube-system" Namespace
[upload-certs] Using certificate key:
a2db3{{ omitted }}3fc58ca7b0410cc412

That's closer to what I want, but there seems to be no way to suppress those two [upload-certs] without resorting to grep, which I'd prefer not to.

12:33:17
@sheogorath:shivering-isles.comSheogorathif you want to set things up with Ansible, I guess the best way is to check how kube-spray does it, as they are the official ansible deployment method: https://kubespray.io12:42:03
@sheogorath:shivering-isles.comSheogorath * if you want to set things up with Ansible, I guess the best way is to check how kube-spray does it, as they are the "official" ansible deployment method: https://kubespray.io12:42:16
@rhouben:vac.systemec.nlRens HoubenYeah, but if I used kubespray directly I'd learn nothing about kubernetes deployment myself. :) 12:42:29
@rhouben:vac.systemec.nlRens HoubenThanks for the pointer, I'll give them a read.12:42:34
@sheogorath:shivering-isles.comSheogorathhttps://github.com/kubernetes-sigs/kubespray/blob/4baa2c8704fe4f091e91630809dc189920b5a1a4/roles/kubernetes/control-plane/tasks/kubeadm-setup.yml#L192-L22812:44:32
@rhouben:vac.systemec.nlRens HoubenPerfect, that looks like what I'll need.12:45:32
@rhouben:vac.systemec.nlRens Houben ... Mind, that is essentially just grepping the output, but hey. :) 12:53:05
@sudoer:beeper.comsudoer I've been staring at a failed flux kustomization for like 3 days. Finally figured out I had 2 objects in one file not separated by "---". Kubeconform, yamllint, kustomize build all didn't help me find it, any other sanity checking tools I should be using? 13:55:37
@sheogorath:shivering-isles.comSheogorathshouldn't yamllint's key-duplicates rule have caught that, since you had two kind or apiversion keys in the same document? https://yamllint.readthedocs.io/en/stable/rules.html#module-yamllint.rules.key_duplicates14:05:17
@sheogorath:shivering-isles.comSheogorathOf course an easy way to avoid this issue in first place is to not allow multiple documents in the same file. But that's more of a convention thing, than actually finding the issue.14:09:34
@feeds:integrations.ems.hostFeeds New post in Kubernetes Blog: Kubernetes v1.30: Uwubernetes 21:00:09
@kenrachynski:matrix.orgDoofusCanadensisuwu? really?21:12:30
@sheogorath:shivering-isles.comSheogorathyes, that's a thing!21:20:21

Show newer messages


Back to Room ListRoom Version: 10