Uninstall instructions don't address undoing the swarm

[DiegoBM]

To Reproduce

N/A

Current vs. Expected behavior

Installation automatically creates a swarm and adds the Dokploy node as manager, later on you are given instructions on how to add worker nodes.

Currently Uninstall instructions page doesn't mention how to remove the underlying swarm created upon install.

Uninstall instructions should address how to undo the swarm setup created by the installation process as well as further addition of nodes, if it's not done automatically by some Dokploy functionality, so that all systems (both managers and workers) go back to a completely clean state after uninstallation.

Optionally mention how to uninstall docker, since the installation script can install it automatically, although probably a stretch

Provide environment information

All environments

Which area(s) are affected? (Select all that apply)

Docker

Are you deploying the applications where Dokploy is installed or on a remote server?

Both

Additional context

No response

Will you send a PR to fix it?

Maybe, need help

[dosubot]

Hi @DiegoBM! I'm Dosu and I'm helping the Dokploy team.

You're right — the current uninstall documentation covers removing Dokploy services, volumes, networks, config files, and running Docker prune commands, but it does not include leaving Docker Swarm mode, even though the install script explicitly initializes a swarm with docker swarm init.

To fully undo the swarm setup, the missing steps would be:

For single-node setups:

docker swarm leave --force

For multi-node setups (worker + manager nodes):

# On each worker node first:
docker swarm leave

# Then on the manager node:
docker swarm leave --force

These should ideally be added after the existing "Remove Docker Networks" step (step 3) and before the general Docker cleanup (step 4) in the uninstall docs.

It's also worth noting a related gap flagged in issue #4832: when services/organizations are deleted in Dokploy, only database rows are removed while actual Docker Swarm services keep running. This means any user-deployed applications would also need to be cleaned up manually (e.g., docker service ls to find them, then docker service rm to remove them) before or during the uninstall process.

If you'd like to send a PR, the uninstall docs live in the website repo and the changes would be fairly straightforward — adding a swarm cleanup step and optionally a note about removing user-deployed services first.

To reply, just mention @dosu.

---

Share context across your team and agents. Try Dosu.

[DiegoBM]

@dosubot I have created the pull request 170 trying to address as well the related issue that you mentioned

[dosubot]

Thanks for putting that together, @DiegoBM! I can see PR #170 adds the swarm cleanup steps for both worker nodes and the manager node, which looks like it covers the gap nicely. Including docker service rm $(docker service ls -q) on workers before leaving is a good touch to ensure user-deployed services get cleaned up too. 👍

To reply, just mention @dosu.

---

Share context across your team and agents. Try Dosu.