Kibana throws encryptionKey missing although it is present?

Elastic Stack Kibana
,
1

Why does Kibana throw this exception:

A new encryption key is generated for saved objects each time you start Kibana. Without a persistent key, you cannot delete or modify rules after Kibana restarts. To set a persistent key, add the xpack.encryptedSavedObjects.encryptionKey setting with any text value of 32 or more characters to the kibana.yml file

although we've specified the following in our docker-compose file:
xpack.encryptedSavedObjects.encryptionKey: <redacted value>

The setting is specified under environment for the Kibana container. Our encryption key is 48 characters long. We're running v8.10.x of the stack. The environment settings looks like:

    environment:
      HTTPS_PROXY: ''
      HTTP_PROXY: ''
      KIBANA_SYSTEM_PASSWORD: ${KIBANA_SYSTEM_PASSWORD:-}
      http_proxy: ''
      https_proxy: ''
      xpack.encryptedSavedObjects.encryptionKey: <redacted>

I'm puzzled why this is happening. Any clues anybody?

2

The environment variable is wrong, it needs to be XPACK_ENCRYPTEDSAVEDOBJECTS_ENCRYPTIONKEY

1 Like
3

I think it's important to clarify that I'm running in a (relatively) "confined" environment (no Internet connection). Although the issue you pointed out is 100% valid I'm still receiving the same error after I amended my compose file.

The Kibana log looks like below:

image
image1270×1119 302 KB

and my Kibana environment looks like:

image
image2272×223 24.3 KB

So I'm still stranded (although I was very happy initially that you pointed out my embarrassing mistake) and I think the issue is because I'm "locked in". Is it possible for me to download the necessary artifacts and make them available to my docker compose stack? If it's possible to download them I may put them into our own repo and get them from there if it's possible to re-route the artifact requests by declaring another repo base url...

4

Can you share your entire docker compose?

I have the same environment variable on my compose without any issues.

5

Hi Leandro!

Hmmmm... I'm currently a little puzzled but I've deployed my ansible playbook more than ten times now and I must agree with you. No problem! I didn't change anything except the environment setting you pointed out and had the issue I mentioned but the recent deployments have worked flawlessly. I simply must have done something else but can't remember what :slight_smile: I'll monitor this for a while and see if this "problem" reappears. Many thanks for your support!

Cheers

6

Hi Leandro!

I've "rolled out" my playbook now numerous times and it seems to work perfectly fine. Frankly I don't know why I had the issue earlier and am sorry I stirred up some unnecessary dust.

Cheers