how come k3d uses nginx as proxy into k3s? why not use traefic which k3s uses as default?

for local development, are you using k8s volumes for source code? is there a better solution?

If it's code that supports hot reloading then yes. There are also tools like

air

to make that work for compiled languages ( Go in this case). Otherwise there are e.g. tilt or skaffold for dev loops. In acorn (acorn.io) we have a built-in dev mode even.

I have restarted the host, so server and agents containers have been restarted, but, i got the error in agent containers:

Waiting to retrieve agent configuration; server is not ready: failed to retrieve configuration from server: <https://127.0.0.1:6444/v1-k3s/config>: 401 Unauthorized

, now, server is ready and agents is not ready in k3d cluster, so what should i do for resolving this problem?

When I run: kubectl get pods -n kube-system I don't see the api server, controller manager listed I do see coredns listed why is this?

I ran into an issue with DNS not working inside a container. I found issue: https://github.com/k3d-io/k3d/issues/209 The last statement in this bug raises an issue with whether

K3D_FIX_DNS=1

is still required/currently works what's the current status on this issue?

❓ hello, quick question: is

k3d-dind

still a thing? I found mentions in older articles (2021) and 1yo image in Rancher dockerhub. I am interested in controlling cluster spin up/down with docker-compose lifecycle. so

k3d-dind

seems like a perfect option for me.

Hi everyone, does k3d support in-place K8s upgrades, specifically via Rancher? I ask because when adding a k3d-created cluster to Rancher, it shows up as a regular k3s cluster which seemingly can be upgraded. (I’m not sure about the mechanism, but I believe Rancher then replaces the k3s binaries within the containers.) Should I expect trouble when doing this? I ask because I don’t see any references to upgrades (in-place or otherwise) in the docs. Thanks!

@wooden-coat-97755 has left the channel

Hello, I am trying to start latest available

k3d:sha1-1afe-dind

image. Getting same issue as in this thread ,

ip: can't find device 'ip_tables'

Author solved it by building own dockerimage with

ip_tables

installed, but it seems not very maintainable for me. Github issues does not mention it. What am I doing wrong? WSL.

k3d

binary worked perfectly.

❔ Hello, what would be the best option to create cluster upon start of k3d dind container? I have config file for new cluster, and docker-compose that starts

<http://ghcr.io/k3d-io/k3d:5.5.1-dind|ghcr.io/k3d-io/k3d:5.5.1-dind>

container, and configfile directory is volume-mounted to container. I'd like to achieve smth like official postgres container that runs everything in

docker-entrypoint-initdb.d

directory after db engine start (e.g. to have some dataset populated). So I'd like

k3d

to create my cluster after start, without doing custom image based on

k3d-dind

. • Overwriting

command

or

entrypoint

dos not look like the best option, (would mean having combo of calling default entrypoint plus cluster creation command), though seems easily doable. • Another idea would be some one-off container calling creation command, but

k3d

does not seem to expose some callable gimmicks out of container, and running

docker exec <k3d_container> /bin/k3d create --config <..>

from other privileged container Any thoughts?

I followed the steps here to install k3d in airgap env, but it did not work. Any one got k3d cluster installation working in airgapped env?

<https://github.com/k3d-io/k3d/discussions/685>

🆕 https://github.com/k3d-io/k3d/releases/tag/v5.5.2 🆕

What's the recommended way to access a container running in k3d as the root user?

I have a surprising DNS issue: I created a cluster using k3d 5.5.1 with

K3D_FIX_DNS=1

and otherwise default networking settings. The

resolv.conf

points to the docker network gateway, as expected (in this case

192.168.30.129

). The issue: After restarting the docker service (`systemctl restart docker`;

live-restore

is enabled in docker’s

daemon.json

– not sure if relevant) DNS resolution no longer works (

nslookup: write to '192.168.30.129': Connection refused \n ;; connection timed out; no servers could be reached

). It starts working again after restarting the container (

docker stop/start …

or

k3d node stop/start …

).

On page: https://github.com/k3d-io/k3d/blob/main/CONTRIBUTING.md I clicked on: The link in: Get an Overview of the k3d project in the documentation: k3d.io/internals/project and got a 404 error! Can someone look at this and let us know the correct page to look at?

@wide-oil-9550 has left the channel

Anyone have any pointers on using devcontainers with k3d for service development for a service which needs access to services already running in cluster?

I am running

k3d cluster stop k3d-mycluster

but I'm getting

FATA[0000] No nodes found for given cluster

. Running

kubectl get nodes

gives:

NAME                     STATUS   ROLES                  AGE   VERSION
k3d-mycluster-server-0   Ready    control-plane,master   23h   v1.27.4+k3s1

I'm trying to restart my k3d cluster as I need to add a node to it

I ran

k3d cluster start mycluster

but upon running

kubectl get nodes

, I'm getting:

E0814 00:10:03.266044  847233 memcache.go:265] couldn't get current server API group list: Get "<http://localhost:8080/api?timeout=32s>": dial tcp [::1]:8080: connect: connection refused
E0814 00:10:03.266545  847233 memcache.go:265] couldn't get current server API group list: Get "<http://localhost:8080/api?timeout=32s>": dial tcp [::1]:8080: connect: connection refused
E0814 00:10:03.267919  847233 memcache.go:265] couldn't get current server API group list: Get "<http://localhost:8080/api?timeout=32s>": dial tcp [::1]:8080: connect: connection refused
E0814 00:10:03.269223  847233 memcache.go:265] couldn't get current server API group list: Get "<http://localhost:8080/api?timeout=32s>": dial tcp [::1]:8080: connect: connection refused
E0814 00:10:03.270505  847233 memcache.go:265] couldn't get current server API group list: Get "<http://localhost:8080/api?timeout=32s>": dial tcp [::1]:8080: connect: connection refused
The connection to the server localhost:8080 was refused - did you specify the right host or port?

🆕 https://github.com/k3d-io/k3d/releases/tag/v5.6.0 🆕

❔ Hello, what's the status of DNS forwarding issue? I am building a cluster in WSL2 with k3d-5.5.2 and getting same DNS not resolving problem 😞 Is

K3D_FIX_DNS=1

still a thing, or it is considered to be fixed?

❓ Hello, not sure if this is a question for

k3d

or

k3s

but there were similar questions in

k3d

github. Corporate root CA propagation to pods. I start

k3d

create cluster using config file. I have

company.pem

which allows to access company's docker registry. I add helm chart to

k3s-manifests-custom

volume and Helm controller starts installing it. Helm chart describes deployment that uses image pulled from company's registry. Ofc it gets

Error: ImagePullBackOff

because of

Head "<https://company-registry:5555/v2/my-image/manifests/my-tag>": tls: failed to verify certificate: x509: certificate signed by unknown authority

In regular local env usually

company.pem

is pulled with curl (from accessible location) and appended to

etc/ssl/certs/ca-certificates.crt

. I tried to map dir containing

company.pem

to

/var/lib/rancher/k3s/server/tls

, to

/etc/ssl/certs

(whole dir as well), pass volume in k3d cluster config to server.. No success. I might be lost in this chain, what would be the correct way? Is image pulled by k3s server container env?

Hi i have a big problem to pull images in my Rencher v 2.7.5 . I want pull images to deploy one deployment from my registry. My registry es Harbor

How would I configure k3d for coredns when the host has the following /etc/resolv.conf?

nameserver 1.2.3.4
options edns0 trust-ad
search <http://my.domain.com|my.domain.com>

error:

[ERROR] plugin/errors: 2 <http://my.domain.com|my.domain.com>. AAAA: read udp 10.11.12.13:35888->8.8.8.8:53: i/o timeout
[ERROR] plugin/errors: 2 <http://my.domain.com|my.domain.com>. A: read udp 10.11.12.13:47750->8.8.8.8:53: i/o timeout
[ERROR] plugin/errors: 2 <http://my.domain.com|my.domain.com>. A: read udp 10.11.12.13:34842->8.8.8.8:53: i/o timeout
[ERROR] plugin/errors: 2 <http://my.domain.com|my.domain.com>. AAAA: read udp 10.11.12.13:45266->8.8.8.8:53: i/o timeout

Hi, we are using rancher for quite some time, was there already the idea to manage k3d Clusters form Rancher?

basically with driver for Docker or k3d

I'm trying to test something that uses k3d with multiple environments. With Rocky 8, I did a default install, swapped podman for docker, & k3d worked fine. When I redid the Rocky 8 install to set it to use the DISA STIG security policy but otherwise the exact same setup, now I can run k3d as root but not as my normal user. As normal user

docker ps

works, but trying to run k3d gets

-bash: /usr/local/bin/k3d: Operation not permitted

. file permissiosn are 755 and SELinux was enforcing on both installs so I don't think it's the SELinux context. I tossed a shell script with an echo command in /usr/local/bin/test.sh and set to 755 and it echoed properly. Any idea what kinda' lockdown is breaking it with normal user permission?

Does anyone have any experience using nerdctl to run k3d? Is it possible, what issues are caused by doing this?

Is there a way to generate a cluster config file from an existing cluster?