Hi, I post this question also in this channel, as I am not sure if this is a incorrect use from my side (using the Terraform provider) or actually an issue in the Rancher UI

Hi everyone,

Hi everyone, I am trying to connect with nerdctl login to Nexus Repository Manager, and I am having problems connecting, I attach the evidence, please helpme

Any idea, how to manually restart kubelet,kube-proxy,calico,rke2 in window worker node

hi, if I create a nodeport service in rancher, how can i access it? I tried deploying an nginx container with a nodeport service, then I do curl to the api server and the external port but i can't reach it

is rancher kind of masking the k8s api server?

Hi,

Hi, I am trying to run Prometheus on a WSL for Windows 11 (build 22621) to monitor an application, but it does not seem to have any visibility to the outside world (i.e. localhost) My docker compose looks like this:

version: '3.7'
services:
  prometheus:
    container_name: prometheus
    image: prom/prometheus
    restart: always
    ports:
      - 10090:9090
    volumes:
      - ./src/prometheus.yml:/etc/prometheus/prometheus.yml

The scrape config section for Prometheus is:

- job_name: "prometheus"
    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.
    static_configs:
      - targets: ["localhost:9090"]

- job_name: "Windows-Exporter"
    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.
    scheme: http
    static_configs:
      - targets: ["host.docker.internal:9182"]

I followed the faq section to add a firewall rule for Rancher Desktop to use

host.docker.internal

, but keep getting the

context deadline exceeded

(i.e. timeout while accessing site) error for the

Windows-Exporter

job. Any idea what else I could be missing?

I was wondering what kind of automated tooling is used to integrate test scenarios as seen for example in this issue? https://github.com/rancher/rancher/issues/40215 Just being curious how these items are tracked 🙂

Hi all, I have a question regarding the installation of kube-vip https://github.com/kubernetes/kubeadm/blob/main/docs/ha-considerations.md#kube-vip What is the role of the network interface here "--interface"? and what happens If I don't provide it As in similar solution "MetalLB" it doesn't require me to provide any network interface. TIA

Hey team - Anyone got RKE2 working with a private ECR registry? My nodes are not on AWS.

Hi, modified labels on nodes are reverting back after node restart. The initial label

<http://topology.kubernetes.io/zone=|topology.kubernetes.io/zone=>"blue"

was added during install with this:

curl -fL <https://rancher.local/system-agent-install.sh> | \
      sudo  sh -s - --server <https://rancher.local> \
      --label '<http://cattle.io/os=linux|cattle.io/os=linux>' \
      --token cluster_token \
      --worker \ 
      --node-name inventory_hostname \
      --label <http://topology.kubernetes.io/region=%22europe%22|topology.kubernetes.io/region="europe"> \
      --label <http://topology.kubernetes.io/zone=%22blue%22|topology.kubernetes.io/zone="blue">

Now I need to change the value of the label to

<http://topology.kubernetes.io/zone=|topology.kubernetes.io/zone=>"red"

I did this using UI but after node restarts the label changes back to original value. Is there a way to make the change permanent?

Is there any documentation about how many pods per system deployment are supported? What do others have these set to?

Hi there. If the latest RKE setup brings a vulnerable/obsolete/whatever nginx ingress controller should I replace it with a recent nginx deployment or must I always stay with the Rancher-released images?

Hi all, what are the best practices for backing up k8S related ressources in RKE2 clusters running on VSphere ?

👋 Hi everyone! I have installed rancher desktop in my mac. Purpose is to run below command

sudo docker run \
    --detach --rm \
    --name lensesio \
    -p 2181:2181 \
    -p 3030:3030 \
    -p 8081-8083:8081-8083 \
    -p 9581-9585:9581-9585 \
    -p 9092:9092 \
    -e ADV_HOST=127.0.0.1 \
    -e RUN_TESTS=0 \
    lensesio/fast-data-dev:latest

but i found out in order to run above command i need to have disable "enable kubernates" from the rancher desktop screen and enable the dockered(moby) option. When i do this i get the below error, can you please help here

Hello everyone, hope you are all doing well. I have a rancher server that control mutliples RKE child clusters, the child clusters are in a lan but can connect to the rancher server. We have prometheus running on each local RKE child cluster but I cannot visualize the metrics with grafana since grafana cannot reach it. I am looking at ways to have the prometheus service on the local child cluster available on the rancher-server is there a way to have a service from a child clustser exposed on the main server permanently ? Either dirrect via a port or a path ? I guess it is similar of doing a kubectl port forward that give access to a svc on a child cluster but I want something more permanent . I think I saw some documentation on that a while ago but cannot remember.

HI, I have an issue with an image

rancher/mirrored-nginx-ingress-controller-defaultbackend:1.5-rancher1

the pull request failed. I used rancher in airgap mode, i need to download all images and this one failed

docker pull rancher/mirrored-nginx-ingress-controller-defaultbackend:1.5-rancher1

4fc7e39c4260: Download complete 4dc5e07c8ca4: Downloading 528B failed commit on ref "manifest-sha256:4dc5e07c8ca4e23bddb3153737d7b8c556e5fb2f29c4558b7cd6e6df99c512c7": commit failed: "manifest-sha256:4dc5e07c8ca4e23bddb3153737d7b8c556e5fb2f29c4558b7cd6e6df99c512c7" failed size validation: 528 != 428: failed precondition

I have a 3-node RKE-HA cluster where Rancher was deployed over 3 years ago. I did keep things updated, so following are the versions: All 3 nodes are RHEL 7 Rancher version 2.7.4 stable RKE version: 1.4.5 kubernetes version 1.24.10 Rancher Backup was configured and backups are happening to S3 bucket... My question is: I want to restore above rancher on RKE2: 3 - node HA RKE2 cluster OS: RHEL 8.7 RKE2 version 1.25.9 Will this work? Is this a good idea? Why I want to do this? I prefer "fresh" OS environment instead of in-place upgrade from RH7 to RH8. Plus this will also allow me to get to RKE2 (hence get rid of docker requirement that RKE has).

Hi guys, I’m using the rancher rest api to deploy via CI/CD but if I deploy a non existent workload it is always creating a headless service together. How I control this? I need to ser the correct information to this service or disable that creation. Any ideas?

Can i use Google Cloud Storage as etcd-s3-endpoint for RKE2 snapshots ?

Hello Everyone, is there any way to stop/cancel a restore of a cluster?

The default Kubernetes API Server for RKE2 uses port 6443. How can I customize this port during installation, such as changing it to 16443？

Hello! We have setup our main rancher RKE2 with helm and the option privateCA=true. Is there a way to propagate the privateCA to child workloads created from the main rancher deployment? We currently seem to have autogenerated CAs in our child clusters. Do we have to manually add the privateCA to each deployment?

Hello guys, I am new to Rancher Desktop. I just installed it according to the manual and after

docker image list

command I receive the following error

error during connect: in the default daemon configuration on Windows, the docker client must be run with elevated privileges to connect: Get "<http://%2F%2F.%2Fpipe%2Fdocker_engine/v1.24/images/json>": open //./pipe/docker_engine: The system cannot find the file specified.

Can you help me? Thanks

Hi guys, trying to create a new Project from the UI and i'm getting this error:

namespaces "p-rxsxd" not found

Every time i press

Create

it generates a new project id and says that it can't find it's namespace Any ideas?

I installed a custom cluster with VMS from the RancherUI. My question is if kubelet and kube-apiserver talk TLS with each other by default?

Can i use Terraform with rancher?

Random off the wall question…. using Ansible to configure rke2 cluster, Following the Rancher STIG for hardening RKE2/Rancher. One of the steps is to patch the rancher service to remove the HTTP port (80). Patching with json seems to only be able to remove by index. So if I ran the json patch in my Ansible, it would remove the same index every time, which I think after 2 runs would mean all the ports are deleted. I’ve been trying to build an ansible json_query (jmespath) to find the index of the http port (if it exists) so I can use the index in another task to delete it. But I can’t seem to figure out the jmespath syntax to merge an array index in. I can’t be the first person trying to do this… is there a better way?

hey guys, i am trying to import a cluster without public registry access. both

desiredAgentImage

and

system-default-registry

are set, but the registration command’s manifest still showing

image: rancher/rancher-agent:v2.7.5-rc4

. anyone has seen the same issue?