I’m using UA 1.1.1 hosted in IIS (WS2019). I’ve two AppPools/Websites, one for the Automation itself, one for the Dashbaord. The automation is working fine, I can acces it with the commands like Get-UAJob including the configured app token.
But the universal dashboard throws some errors after the authentication was processed:
The specified log file is empty or has no log entry while the issue happens. In the chrome debugging tools I can see, that I have some 401 for the following call:
Any idea how to fix that?
Hmmm. Not sure what’s going on but I have a couple debugging steps we can take.
From the browser window that currently has this error, press F12. Then click the Network tab and see if there was a request made to /api/sessionAppToken. That should be returning the session app token to the browser.
That should list the token that was returned from the server. Then we should be able to use that token with the cmdlets to see if we are experiencing the same problem there.
Yes, I have a request against /api/sessionAppToken, status code 200 and in the body I can see a long encoded string, should be the app token. Immediately after this request, two request against the following api are performed, both 401. The token which is passed in the header “uatoken” is the same token returned by the /api/sessionAppToken before.
If I type the following command in the “Console” tab of the developper tools, I get an error:
By the way, I have some WebSocket handshare errors before:
Forgot to say, the app token returned by /api/sessionAppToken also don’t work with the cmdlets -> I get a 401 Unauthorized.
Thanks. That’s good to know. Can you also send me the line that you are using to generate the UDAppTokens in your authorization policy?
One other thing I was thinking of. Could you do a Get-UAAppToken with the system token to see the tokens that are available? I’d be curious if we have some bad tokens in the system and it’s using one of those.
Okay, it was something wrong in the database, maybe the AppTokens as you’ve said. Deleting the database and start over with a new empty one did resolve the problem. If it will happen again, I will update this ticket.