Our Oompa Loompas have not been able to get your credentials from OpenID.

[lionelbeard]

Hi,

I'm trying to make taiga works with keycloak. I have 3 instances (taiga + keycloak) deployed on kubernetes:

• instance 1 on cluster k8s 1: WORKS
• instance 2 on cluster k8s 2: DO NOT WORK
• instance 3 on cluster k8s 1: DO NOT WORK

Actually, only the first deployed instance is working, but all new installations failed with same error: "Our Oompa Loompas have not been able to get your credentials from OpenID." after login through keycloak.

Note: option PUBLIC_REGISTER_ENABLED is True on all instances (#22)

If I look into container logs:

• taiga-gateway: "POST /api/v1/auth HTTP/1.1" returns 500 on failed instances, 200 on working instance
• taiga-back: no log, even in debug level
• taiga-event: {\"cmd\":\"auth\",\"data\":{\"token\":null,\"sessionId\":\"af728c0b0c7be9e1f5c61fa2da69ee1cfe49d39e\"} on failed instances, token is null, but it is not null on working instance.

and…… this is all I have.

Is someone can help me to debug this? I spent several days on it, but still not possible to make it work. I've even tried to connect taiga on instance 2 to keycloak on instance 1, but still not working…

Thanks for your help.

[Rdeisenroth]

Any Updates on this? I have the same issue. For me every time i try to log in the taiga-back worker crashes with just a timeout:

[2023-05-08 17:13:38 +0000] [37] [INFO] Booting worker with pid: 37

[2023-05-08 17:15:39 +0000] [1] [CRITICAL] WORKER TIMEOUT (pid:35)

[2023-05-08 17:15:39 +0000] [59] [INFO] Booting worker with pid: 59

[2023-05-13 13:16:16 +0000] [1] [CRITICAL] WORKER TIMEOUT (pid:37)

[2023-05-13 13:16:16 +0000] [67] [INFO] Booting worker with pid: 67

I have set Debug=True yet i still don't get any reasonable error message. In the browser i just get a 502 Bad Request because of that timeout, that i haven't managed to debug. (probably related to #35 )

[drpsyko101]

I had the same issue, but managed to fix it with slight configuration changes:

OPENID_ID_FIELD: "username"
OPENID_USERNAME_FIELD: "username"
OPENID_FULLNAME_FIELD: "first_name"

and for some reason, incorrect email settings might also cause the same issue. In my case, I've deactivated SMTP settings and done those above and it works now. After setting the right SMTP configuration, I've managed to get it running as well.

[Rdeisenroth]

Hmm i retried with and without smtp and with and without setting custom open id fields, same result

[Rdeisenroth]

Okay, after digging a bit further, i found my issue: i had all my docker containers in a "taiga" Network and i made it internal, thus preventing the containers from getting internet Accesss. Found this out by manually overwriting the backend config and setting DEBUG to True.

Now everything but smtp works for me. With SMTP i still get the error and a timeout though...