Automatically redirected to "undefined user" page after logging in

Same behaviour (going to the undefined page) if using /login as suggested.

1 Like

ok. thanks. if thatā€™s the case, then iā€™ve got no additional ideas.

i tried various browsers, various language settings, browser password remember on and off, e-mail vs iNat ID to log in, the various login pages, etc., and i just canā€™t reproduce what you guys are encountering. it doesnā€™t look like your accounts are set up in any strange ways either as far as i can tellā€¦ so i donā€™t see any other obvious ways to approach this for now.

There has to be some kind of a change in the last 4 to 5 days driving it as it has only taken place very recently for me.

the problem is that if no one else can consistently reproduce the problem besides you and chris971, thereā€™s not really a good way to debug the issue. it would just be scanning through code and guessing where the problem might be. just doing a very high-level scan of the changes over the last week, thereā€™s nothing specific that would seem like it should trigger this kind of problem.

1 Like

And so, no ā€œundefined userā€ this morning!

  • My bookmark was set to https://www.inaturalist.org/
  • I have been directed to the unlogged iNaturalist homepage and I followed the ā€œconnexionā€ link
  • It brought me to https://www.inaturalist.org/login where I filled my details in.
  • And I finally arrived to my regular profile page, and not the ā€œundefined userā€ one.

I still got it 5 minutes ago. I tried not selecting the ā€˜remember meā€™ option, but it still happened, so I would assume it is not related to that.

Is this still happening for anyone? Visiting /login or /users/sign_in should not make a difference. Where you get taken after sign in depends on what page you visited previously and your session. /users/undefined looks like some kind of JS request is being made for a user that doesnā€™t exist and thatā€™s populating the session with that URL as the last one you visited, but since I canā€™t replicate the problem itā€™s hard to hunt down, as @pisum mentioned.