I’m trying to add GitHub authentication onto my BinderHub. It’s something I’ve successfully managed before, but for some reason it’s giving me difficulty this time around.
<jupyterhub-ip>/hub/login keeps giving me a 403: Forbidden error. Config is below.
config: BinderHub: # connect image registry and set image prefix use_registry: true image_prefix: <docker-org>/<prefix>- # jupyterhub IP address hub_url: http://<jupyterhub-ip> # enable authentication auth_enabled: true jupyterhub: hub: services: binder: oauth_redirect_uri: "http://<binder-ip>/oauth_callback" oauth_client_id: "binder-oauth-client-test" extraConfig: hub_extra: | c.JupyterHub.redirect_to_server = False binder: | from kubespawner import KubeSpawner class BinderSpawner(KubeSpawner): def start(self): if 'image' in self.user_options: # binder service sets the image spec via user options self.image = self.user_options['image'] return super().start() c.JupyterHub.spawner_class = BinderSpawner singleuser: # to make notebook servers aware of hub cmd: jupyterhub-singleuser auth: type: github github: clientId: "xxxx" clientSecret: "xxxx" callbackUrl: "http://<jupyterhub-ip>/hub/oauth_callback"
I’m using BinderHub Helm Chart version
0.2.0-7b2c4f8. I tried to go back to version
0.2.0-3b53fce which was the last version I had authentication working on, but the JupyterHub pod kept giving
CrashLoopBackOff. I’ve tried remaking the OAuth app on GitHub as well. The only other difference from the last time I had this working is that my Kubernetes cluster is now deployed onto an Azure Virtual Machine Scale Set (VMSS) which allows autoscaling, but it’s a “preview” feature.
Thanks in advance for any advice! It’s possible I’m a victim of copypasta and blind to the issue, so even just a fresh set of eyes on the config would be much appreciated.