You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
kubectl oidc-login get-token -h | grep listen-address
--listen-address strings [authcode] Address to bind to the local server. If multiple addresses are set, it will try binding in order (default [127.0.0.1:8000,127.0.0.1:18000])
Since default is 127.0.0.1:8000, this could (and did for me) collide with my browser's cache. I had recently run this command:
docker run --platform linux/amd64 --publish 8000:8080 ghcr.io/mermaid-js/mermaid-live-editor
and my browser cached the mermaid editor rather than hitting the webserver for login. I'd request the default be changed to something other than 8000 to avoid future occurrences for others.
To reproduce
run kubectl oidc-login get-token --force-refresh --oidc-issuer-url=...
have browser open, but http://localhost:8000 is cached as something else, so oidc web server never hit and console hangs
Your environment
OS: macOS
kubelogin version: e.g.v1.28.0
kubectl version: e.g. v1.29.5
The text was updated successfully, but these errors were encountered:
There might be a different issue at play - binding to / listening on 127.0.0.1:8000 can succeed even if something else has taken :8000 (i.e. a wildcard IP, which includes IPv6) - this is totally fine, but there is inconsistency with the bind IP of 127.0.0.1 and opening the browser to localhost which can be a different IP address (possibly depending on OS or ordering in /etc/hosts) - a fix would be to open a browser on 127.0.0.1:$PORT instead of localhost:$PORT.
I worked around the issue by changing localhost:8000 in my browser (which was some other Docker-based app) to 127.0.0.1:8000 (and again after the redirect back)
Describe the issue
Since default is 127.0.0.1:8000, this could (and did for me) collide with my browser's cache. I had recently run this command:
and my browser cached the mermaid editor rather than hitting the webserver for login. I'd request the default be changed to something other than 8000 to avoid future occurrences for others.
To reproduce
kubectl oidc-login get-token --force-refresh --oidc-issuer-url=...
Your environment
The text was updated successfully, but these errors were encountered: