Hi team, I’m trying to provision downstream cluster using docker commands and I got some errors. This error is from Rancher UI while trying to finish the provisioning process. Failed to apply the ServiceAccount needed for job execution: Post "https://10.2.28.126:6443/apis/rbac.authorization.k8s.io/v1/clusterrolebindings?timeout=30s": Forbidden There are no errors on the containers running de agents.

Hi All - I need to generate API keys to access Rancher. Is it possible to authenticate to Rancher using an Azure AD Service Principle instead of an actual user account?

good day, please have a look at my thread https://rancher-users.slack.com/archives/C3ASABBD1/p1679568407311599

Hi All I was using docker desktop and replaced with Rancher desktop

One difference i am seeing is container in docker desktop starts in root whereas here its not and container start is failing with permission denied when chown is part of container start script

how do yall currently do basic alerting 'pod stuck, node down, etc' in rancher? The oldschool default alerts havnt worked for awhile on newer versions

Does anyone know if there is a separate Harvester HCI slack channel?

I am using rke installation through the command line ... I did rke config ... now how can I add a new master in the existing kubeconfig for k8s with certificate etc …kubeconfig was generated initially which had only one master and now I want to add another 2 masters ..... I can see inside rke config 3 masters are there but in kube config only one node with a certificate etc…

The company I work for is trying to upgrade our k8s cluster. Unfortunately, our infra is cut off from the outside world so all sourcing has to be done manually. When trying to source the rancher/rke-tools image, which is a dependency on the rancher rke install, our Aquasec scan is picking up the CVE-2019-5736 vulnerability in the image due to the version of Docker in that image. So it's being blocked from entering our estate. Could somebody explain how this image is used during the cluster creation, i.e is it ephemerally spun up to perform some action and then stopped, or does it remain running to perform further tasks within the cluster? We were unsure how to mitigate this vulnerability unless we understand its use in the cluster. We're aware we can update our Deployment yaml to ensure runNotAsRoot to true, but unsure of the best practice as this would be dependent on the usage of the images in the cluster. need some urgent help

how to choose istio, the rancher integrated one or install it from official site

Anyone have any opinions on the best way to automate RKE downstream cluster creation? I'm using TF Rancher2 Provider and I just have so many consistency issues. I can't get a clean build and destroy to work. I'm at a point where creating a cluster using node templates all works, but destroy has issues where node templates don't delete properly or cloud credential key will be deleted somehow, but TF will fail to delete it because it doesn't exist. I'm also a little confused how to customize docker on the RKE nodes, things like mirror and dockerhub credentials that require a docker restart. Do I build a prebuilt AMI (in AWS) with everything loaded, or is there a way to do this with the node template configuration?

Error: Error removing Cloud Credential: Bad response statusCode [500]. Status [500 Internal Server Error]. Body: [baseType=error, code=ServerError, message=Index with name by-cloud-cred does not exist] from [https://rancher-ha-jw-mgmt.bylightsdc.bylight.com/v3/cloudCredentials/cattle-global-data:cc-28pdw]

I have standalone rancher (docker) installed and have successfully imported DigitalOcean Managed Kubernetes (rancher is in a remote location accessible via https). The ingress controller is traefik and Cert-manager is installed. Everything looks good (can deploy pods and access from web) except for Fleet. I am getting Wait Check-in. What is missing and what can I do to get fleet going? Thanks

this is my article regarding LB in on premises, it’s using rancher actually and in house node watcher to written in rust https://kevinchou888.medium.com/in-house-k8s-load-balancer-architecture-on-premises-55d3b23867b9

Hello all 🙂 Im setting up a External Nginx Load Balancer and following this docs - https://ranchermanager.docs.rancher.com/how-to-guides/new-user-guides/infrastructure-setup/nginx-load-balancer But seams the syntax is incorrect cuz i get “”proxy_pass” directive is not allowed here ” in the http block can anyone help with the nginx conf

Is there any reason when I login to Rancher, I always have to enter my details twice?

My company has an internal Artifactory instance which we use to remote/cache packages from the internet. I’m trying to configure a remote repository for rpm.rancher.io, but I can’t seem to get it to work. Works fine if I configure yum to use the repository and pull RPMs from the internet so I think the RPM repo is valid. Anyone ever do this or successfully mirror rpm.rancher.io?

I am writing my own cluster driver, and am trying to find out how cloud credentials are sent to the driver. Does anyone have information around cloud credentials and cluster drivers? Thank you

I want to use on-prem nfs storage as PV, what is the general approach with Rancher? I built a cluster with RKE2, but when I tried to create a StorageClass, there was no NFS option in Provisioner. In this case, should I manually set up nfs-client-provisioner?

Hi Everyone, I update rancher monitor to latest version (v100.2.0+up40.1.2) but I am having a new issue that when i type grafana.company.com in browser with that /k8s/clusters/local/api/v1/namespaces/cattle-monitoring-system/services/httprancher monitoring grafana80/proxy/?orgId=1 this one is also adding any idea what where I need to update so it will not come in url then

Once I've developed the extension according to the guide(https://rancher.github.io/dashboard/extensions/extensions-getting-started), how can I install it programmatically on a clean Rancher installation?

Anyone else having problems moving clusters between workspaces in Fleet/Continuous Delivery? I'm trying to move a cluster but when clicking Apply nothing happens. There are no errors in the Browser Console, Network tab or from the fleet pods

Hello, after enabling google Authentication on the rancher, it allows only the domain users with GSuite to authenticate, how do you enable the Gmail users to log in to the Rancher?

Hi all, i am trying to use cluster autoscaler with provider "rancher". As per the github documentation, it says that i need token with the following privileges only: • Get/Update of the

<http://clusters.provisioning.cattle.io|clusters.provisioning.cattle.io>

resource to autoscale • List of

<http://machines.cluster.x-k8s.io|machines.cluster.x-k8s.io>

in the namespace of the cluster resource Ref: https://github.com/kubernetes/autoscaler/tree/master/cluster-autoscaler/cloudprovider/rancher But, the only way autoscaler works is with the token of full admin user. Nothing less. Any ideas please?

I beg you, can someone please answer this question. I am running Rancher 2.6.11. The entire deployment is automated via Terraform Rancher2 provider. Should I be migrating my rancher2_catalog resources to rancher2_catalog_v2 and my rancher2_app resources to rancher2_app_v2? or are these just more choices? I cannot find a clear answer.

Seeing as how Catalogs are listed under Legacy along with Alerts, CIS Scans, Notifiers and Projects, Its safe to assume I should be migrating to Repos / rancher2_catalog_v2

Anyone run into issues with Worker nodes on VSphere having an issue with

kube-proxy

not coming back up properly with a kubeconfig file not found error in

/etc/kubernetes

? Seems to happen every time to our worker nodes now if they power down and its been a big annoyance lately. This is on Rancher

v2.6.10

.

Is it possible to limit the login for rancher to certain subnets or ip addresses?

is anyone familiar with replacing the rancher web cert with cert manager and using lets encrypt i cant find any clear documentation.

wanted to know will k3s is good far more loads, or do we need to move k8s ?