Database backup fail when database is in worker node

To Reproduce

1. Setup a worker in your swarm
2. Add a label to the worker (eg. docker node update --label-add type=database <node-name>)
3. Setup a postgres database application to the worker (don't deploy yet)
4. Add a constraint to the postgres database application

{
  "Constraints": [
    "node.labels.type == database", // or whatever you label the worker
    "node.role == worker"
  ]
}

5. Deploy the database application
6. Setup database backup
7. Manually trigger the backup

Current vs. Expected behavior

Currently
It throws an error

Initializing backup
[Sat Jan 24 04:02:23 UTC 2026] Starting backup process...
[Sat Jan 24 04:02:23 UTC 2026] Executing backup command...
[Sat Jan 24 04:02:23 UTC 2026] ❌ Error: Container not found

Expected
Database backup created successfully.

Provide environment information

Operating System: Ubuntu 24.04.3 LTS (manager and worker vps)
Dokploy Version: v0.26.5
VPS Provider: Hetzner
What applications/services are you trying to deploy: Database

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

Databases

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

Remote server

Additional context

I have done some digging and the reason this happened is because the backup command code on packages/server/src/utils/backups/postgres.ts

await execAsync(backupCommand, {
	shell: "/bin/bash",
});

is ran on the manager vps. But since the database is in a worker (different vps), it couldn't find the container, hence the error.

Will you send a PR to fix it?

Yes

Start in packages/server/src/utils/backups/postgres.ts, especially the execAsync call used during the backup process. Reproduce the failure with a PostgreSQL database constrained to a worker node, then trace how the backup command is executed. Done means a manually triggered backup creates the database backup successfully on the remote worker.