OpenDistro Docker Compose - Kibana stuck Loading

Hello all,

I am trying to set up the Elastic Stack + OD Plugins through the docker compose install instructions provided here: https://opendistro.github.io/for-elasticsearch-docs/docs/install/docker/#start-a-cluster

All the images boot up fine and report no errors, however when going to Kibana it is stuck on the “Loading Elastic” page.

Again, this is with the provided sample docker compose files without any changes.

This has also occurred to me when installing the alerting plugin standalone on another Elastic stack deployment.

Thanks all for any help.

Are you starting it up in daemon mode? If not you will have logs on your screen, these should provide some kind of information about what the problem is. I have used this docker compose file and it works fine for me.

Thanks for the reply Tony. No I am not starting it in daemon mode, no errors are coming up. I’m running on a 2016 Macbook pro for work.

I decided to run the docker compose on my personal machine and it worked fine there, it was a Windows box with WSL 2 for the docker backend. So I’m guessing there’s a issue of not enough resources allocated to the Docker VM on my Mac perhaps. I am going to experiment with that and will post what I did just in case someone else has this issue.

I run it on a 2018 Macbook Pro and it works fine. You have updated everything you need in brew for it to run?

Everything in brew is updated. Changing the resources available to the Docker Engine fixed my issue.

Before I had 2 CPU’s, 6144 MB of ram allocated.

Now I have 4 CPU’s, 8192 MB of ram allocated.

Whereas before it would just hang on the “Loading Elastic” splash screen, it booted basically instantly. I’m guessing it was the number of CPU’s? Seems a bit strange.

1 Like

Possibly the jvm needing more memory for heap.

This is a known issue. Elasticsearch in general needs a lot of hardware resources to successfully run. You can read more about it here: https://opendistro.github.io/for-elasticsearch-docs/docs/troubleshoot/