I would imagine, from my own experience developing/designing in Alpine, that the containers are somewhat... 'distinct' from support, but they can be assisted with, provided questions, again, are kept within the confines of the running application;

• I had a number of times where the features I was implementing and enhancing required advanced assistance of the engineering team, and only once or twice did contention arise from it being a Docker container, and the contention was resolvable by drawing attention back to the running app; For the most part I think it was the engineer's inclination to satisfy their curiosity vs discarding something as unsupportable, when you get that far up the support chain.
  ~I may have been a rare opportunity that they could interact with me, instead of just instructing.

That said of course, when you get that far up (the support chain), it is unlikely they will discard something as unsupportable, because they know an awful lot more tricks than the agents do, you just need to take each step at a time and work through the issues experienced WITH the staff you are handling at that time.

As far as your question, you will find SIGNIFICANT abundance of performance available on your machine if you choose the Alpine option; I initially developed Ubuntu containers, but I prefer machines which operate with the lowest feature set necessary, and when I moved to Alpine, I could operate Confluence + Jira + Bitbucket on the same hardware that I had one app running on previously;  As one of the prerequisites for my own work was cost reduction, that was a big incentive to complete development of Alpine Docker containers, but naturally with less research time, Ubuntu may be more practical.
Some notes on Docker in general;
SOURCE: https://hub.docker.com/r/atlassian/confluence-server
With the Dockerfile in mind, I created my own and terminated them with CMD ["/opt/atlassian/atlassian-jira-8.0.1/bin/catalina.sh", "run"], but that is a matter of choice on your part.  And while I am no longer developing Docker containers for Jira, as we prefer a more flexible delivery methodology, I can assist you if I and yourself have the time.

Good luck!

Thanks Kyle. It makes me more confident to know that support should not be a problem as long as it stays within the functions of the application.

My last decision is wether to use the Ubuntu Docker image or the Alpine one. Atlassian does provide a Confluence-server Docker image based on Alpine on their DockerHub account. I just would like the confirmation that they support it like they do the Ubuntu image.

I developed, deployed and administered Jira/Confluence Atlassian containers from 2015-2019.

PROVIDED you keep your support requests away from the infrastructure, and handle it based on functional requisites you won't have any issues; If you start to request help with Kubernetes and/or Docker itself you will find yourself in a rough place.
Ultimately, as an engineer/System administrator, I found support could facilitate all of my requests as I was developing and implementing the servers I used.  Docker did provide the benefit of requiring very little maintenance, as well, though each upgrade required development.

Noting that Atlassian has their own page on dockerhub; https://hub.docker.com/u/atlassian/ It speaks for itself that Docker is a platform they are able to facilitate their platform on.

Just like they can't repair a server or a mainframe which their software is running on, as I said, just keep it away from the technical aspects of the platform, and you will find it more than adequate.

As for your secondary inquiry as to Docker images based on Alpine; This is likely not in Atlassian's pipeline, in any easily recreated way, as yet, but it is incredibly easy to develop, and I can help you out if you need it.  There is also a developer working at Atlassian that they create testing infrastructure for which builds an Alpine image; I didn't bother looking into it since I already had mine fully functional, but it is definitely possible, and you can inquire with support with an open case to see if they can connect you, provided you do it in a collegial way.

Hi,

We are preparing to install Confluence in Kubernetes in production. When available, we use docker images based on Alpine Linux.

Therefore I would like to use atlassian/confluence-server:6.15.3-alpine-adoptopenjdk8.  However I would like a confirmation from Atlassian that images based on Alpine are supported in Production.

What is the status current status on this?

Thank you

Hi Adam,

Can we have the Known issues: Alpine Linux removed from the supported platforms page then? It hasn't been an issue for at least a year now.
provided an appropriate base image was selected to build Confluence on top of (https://hub.docker.com/r/anapsix/alpine-java/), Confluence is installable and operable on Alpine.

Ah, yes... I don't want to point fingers, but I would imagine that page was updated by someone unfamiliar with what Docker is, so yes it should definitely be updated on their supported platforms page, because I have definitely proven that wrong, as has Confluence itself with the link I provided earlier.

Documentation can be convoluted, I just opened a ticket on that page, feel free to open one yourself referencing this ticket if you wish.

My issue is that the Dockerfile clearly uses Alpine, but then their own Supported Platforms page explicitly mentions that Alpine cannot be used. So I hope that the Supported Platforms page is dated and needs to be revised.

Keep in mind that this issue was opened in May of 2017, a great deal has occurred in Docker and tech in general in that time frame.

The Dockerfile found in Atlassian's docker hub is presently v6.11.0 (the most recent version of Confluence), I would imagine its' been validated by their team and it operates as expected.

That seems to me to be as official as Dockerfiles come. What is it you require?