dex VS rancher

Dex: https://dexidp.io

Recently I looked into having a relatively simple SSO setup for my homelab. My main objective is that I could easily login with Google or GitHub auth. At my previous job I used both JetBrains Hub [1] and Keycloak but I found both of them a bit of a PITA to setup.

JetBrains Hub was really, really easy to get going. As was my previous experience with them. The only thing that annoyed me was the lack of a latest tag on their Docker registry. Don't get me wrong, pinned versions are great, but for my personal use I mostly just want to update all my Docker containers in one go.

On the other hand I found Keycloak very cumbersome to get going. It was pretty easy in dev mode, but I stumbled to get it going in production. AFAIK it had something to do with the wildcard Let's Encrypt cert that I tried to use. But after a couple of hours, I just gave up.

I finally went with Dex [2]. I had previously put it off because of the lack of documentation, but in the end it was extremely easy to setup. It just required some basic YAML, a SQLite database and a (sub)domain. I combined Dex with the excellent OAuth2 Proxy and a custom Nginx (Proxy Manager) template for an easy two line SSO configuration on all of my internal services.

In addition to this setup, I also added Cloudflare Access and WAF outside of my home to add some security. I only want to add some CrowdSec to get a little more insights.

1. https://www.jetbrains.com/hub/

2. https://dexidp.io/

3. https://github.com/oauth2-proxy/oauth2-proxy

3. https://github.com/alex3305/unraid-docker-templates

Another great option is to use https://github.com/dexidp/dex in an authentication setup. In your app, you federate the authentication to dex using OAuth2. Dex then has a pluggable architecture with built-in connectors for many established identity providers using a variety of protocols: Among others OAuth2, SAML 2 but also GitHub, Google, Gitea and so forth.

dex is well-known: https://github.com/dexidp/dex

We used keycloak for openid identity provider as well. It is fine to setup keycloak once. But it is painful share the setup with other engineers.

For local development, we end up using dex (https://dexidp.io). When we need support group/role, we use dex and glauth(https://glauth.github.io). Both dex and glauth can be configured with yaml files. We just created a few yaml files and a docker compose file, every engineer can be brought up the whole environment in a few seconds.

Also https://www.authelia.com and https://github.com/goauthentik/authentik look pretty promising, if you need more advanced features from them.

I added it as an available option (flake) in Dex: https://github.com/dexidp/dex

Did something happen to the Apache 2 rancher? https://github.com/rancher/rancher/blob/v2.7.5/LICENSE RKE2 is similarly Apache 2: https://github.com/rancher/rke2/blob/v1.26.7%2Brke2r1/LICENS...

I follow the 4 ABCD steps bellow, but the first pod deployment never ends. What's wrong in it? Logs and result screens are at the end. Detailed configuration can be found here.

CVE-2023-22651 is rated 9.9/10 : https://github.com/rancher/rancher/security/advisories/GHSA-6m9f-pj6w-w87g

Depends. When I came into my last company I immediately noticed the lack of reproducible environments. Brought this up a few times and was met with some resistance because "we didn't have the capacity"... Until prod went down and it took us 23 hours to bring it back up due to spaghetti terraform.

For the latest releases of rancher: https://github.com/rancher/rancher/releases When is Rancher 2.7.1 going to be released? The Rancher support matrix for 2.7.1 shows k8s v1.24.6 as the highest supported version and Azure will drop AKS v1.24 in a few months... Should this be a concern for us? What could happen if we create our cluster with Rancher for an unsupported K8s version? 1.25 for example. - Rancher 2.7.2 just got released including support for 1.25. I have however tested running unsupported versions before, unless there is major deprecations in the kubernetes API it is fine in my experience. If we move to AKS imported clusters, in case we add node pools, and upgrade the cluster, will those changes be reflected in the Rancher Platform? - Yep! If we face some issues by running an unsupported K8s version on Rancher Launched K8s clusters, is it possible to remove it from Rancher, do the stuff we need, and then import it into the platform? - Yes, however be careful and do testing before doing in prod. From top of mind: Remove cluster from rancher (if imported), if rancher created you might want to revoke ranchers SA key for the cluster first (so it can't remove it). Delete the cattle-system namespace, and any other cattle-* namespaces you don't want to keep. And do your thing. It looks like AKS is faster than Rancher regarding supported Kubernetes versions... We would like to know if Rancher will always be on track with AKS regarding the removal of K8s version support and new versions. - In my experience yes. (Been using rancher on all three clouds for a 4 years now). What are exactly the big differences between imported AKS and Rancher-launched AKS? What should we look at, and what issues can we face when using one or another? - The main difference is that rancher will not be able to upgrade the cluster for you. You will have to do that yourself.

variable "rancher" { type = object({ namespace = string version = string branch = string chart_set = list(object({ name = string value = string })) }) default = { namespace = "cattle-system" # There is a bug with destroying the cloud credentials in version 2.6.9 until 2.7.1 and will be fixed in next release 2.7.2. # See https://github.com/rancher/rancher/issues/39300 version = "2.7.0" branch = "stable" chart_set = [ { name = "replicas" value = 3 }, { name = "ingress.ingressClassName" value = "nginx-external" }, { name = "ingress.tls.source" value = "rancher" }, # There is a bug with the uninstallation of Rancher due to missing priorityClassName of rancher-webhook # The priorityClassName need to be set # See https://github.com/rancher/rancher/issues/40935 { name = "priorityClassName" value = "system-node-critical" } ] } description = "Rancher Helm chart properties." }

When I searched DuckDuckGo instead, the 12th link actually had the real answer. It's in this issue on Rancher's GitHub. Turns out the Rancher admin needs to be in all of the Keycloak groups they want to have show up in the auto-populated picklist in Rancher. Being a Keycloak admin and even creating the groups isn't good enough. Frustratingly, the "caveat" note the Rancher guy is pointing to that says this is only present in the guide to setting up Keycloak for SAML, but apparently this is also true for OIDC.

Explicitly set TLS 1.3 in Rancher, though it could be a bug in Rancher: https://github.com/rancher/rancher/issues/35654

Thanks. Yeah looks like this might work: https://github.com/rancher/rancher/releases/tag/v2.7.2-rc3