update 8.1 LDAP Worker Fails to Start After Switching to Dev Server: "no ldap provider defined"

[andreasrevdal]

I started using the dev-server version based on the recommendation in this comment: #11182 (comment)

Since then, none of my outposts work anymore. They keep restarting with the following log:

{"error":"no ldap provider defined","event":"Failed to run server","level":"panic","timestamp":"2024-09-04T14:27:34Z"}

However, I have an LDAP provider configured, as shown in the screenshots below:

This setup worked perfectly before the update to the dev-server and before the 8.0 update. It seems like the LDAP provider is not being recognized properly by the outposts after the change.

Steps to Reproduce:

• Switch to the dev-server image.
• Restart the outposts.
• Observe the logs indicating "no ldap provider defined" even though a provider is configured.

Expected Behavior: The LDAP worker should start successfully, recognizing the defined LDAP provider.

Environment:

• Authentik version: dev-server
• LDAP provider configuration: Confirmed as correctly set.

Any help resolving this issue would be greatly appreciated!

[BeryJu]

Can you re-save the outpost with the LDAP app selected?

[andreasrevdal]

Can you re-save the outpost with the LDAP app selected?

Yes i can re-save with the application, still broken. I deleted it, remade it and went back to main 8.0 but still error there. And still error in dev

[BeryJu]

Can you send an HTTP Get request to http://authentik.company/api/v3/outposts/ldap/ with the Header Authorization: Bearer %s where %s is the AUTHENTIK_TOKEN the outpost uses?

[andreasrevdal]

Can you send an HTTP Get request to http://authentik.company/api/v3/outposts/ldap/ with the Header Authorization: Bearer %s where %s is the AUTHENTIK_TOKEN the outpost uses?

This is what i get from HTTP Get:

Response data: {'pagination': {'next': 0, 'previous': 0, 'count': 0, 'current': 1, 'total_pages': 1, 'start_index': 0, 'end_index': 0}, 'results': []}

This is docker logs: {"event":"Loaded config","level":"debug","path":"inbuilt-default","timestamp":"2024-09-04T18:16:02Z"} {"event":"Loaded config from environment","level":"debug","timestamp":"2024-09-04T18:16:02Z"} {"event":"not enabling debug server, set AUTHENTIK_DEBUG to true to enable it.","level":"info","logger":"authentik.go_debugger","timestamp":"2024-09-04T18:16:02Z"} {"event":"Successfully connected websocket","level":"info","logger":"authentik.outpost.ak-ws","outpost":"27b4af0c-4be4-4d89-98f4-5177517898fe","timestamp":"2024-09-04T18:16:03Z"} {"error":"no ldap provider defined","event":"Failed to run server","level":"panic","timestamp":"2024-09-04T18:16:04Z"} {"event":"finished shutdown","level":"info","logger":"authentik.outpost.ak-api-controller","timestamp":"2024-09-04T18:16:04Z"} panic: (*logrus.Entry) 0xc0001d60e0

and as you can see in this picture it keeps restarting over and over every 2 seconds:

[andreasrevdal]

This is also the same case with the 8.1 update. Still error

[OzSho]

I'm also encountering a recurring issue with the latest version (8.1) of the application. The LDAP container keeps restarting, and the logs show the following error: ak-outpost-ldap | {"event":"Successfully connected websocket","level":"info","logger":"authentik.outpost.ak-ws","outpost":"1581fcea-f002-4a40-a57a-fa33f1f752f0","timestamp":"2024-09-09T20:34:47Z"} ak-outpost-ldap | {"error":"no ldap provider defined","event":"Failed to run server","level":"panic","timestamp":"2024-09-09T20:34:47Z"} ak-outpost-ldap | {"event":"finished shutdown","level":"info","logger":"authentik.outpost.ak-api-controller","timestamp":"2024-09-09T20:34:47Z"} ak-outpost-ldap | panic: (*logrus.Entry) 0xc0004001c0 ... ak-outpost-ldap exited with code 2

This issue occurs regardless of whether the container is built automatically by the server or manually—both methods lead to the same result.This problem persists from 8.0,dev, and now 8.1, apart from deleting the volumes i deleted and reconfigured anything anew but it didn't solve the problem.

the container fails due to an undefined LDAP provider while there is one. I would greatly appreciate any help in resolving this issue. I'm happy to provide additional logs or information if needed.

[BeryJu]

One potential workaround for this would be to create a new Outpost in authentik and assign the same applications/providers and use that token with the container. I'm still not quite sure what causes this issue. You could also try to re-run the outpost_token_ensurer system task, which will ensure that all permissions are correctly set.