[GH-ISSUE #3853] `initial-data.conf` is only processed once during first startup #1339

New issue

Open

opened 2026-03-07 21:02:37 +03:00 by kerem · 1 comment

kerem commented

2026-03-07 21:02:37 +03:00

Owner

Originally created by @mohan-s-gopal on GitHub (Oct 28, 2025).
Original GitHub issue: https://github.com/dbeaver/cloudbeaver/issues/3853

Is your feature request related to a problem? Please describe.
CloudBeaver's initial-data.conf is only processed during the initial startup when the database is empty (configuration mode). After the first initialization, any changes to this configuration file are ignored because teams are persisted in the database and never re-synchronized.

This creates a significant operational challenge for:

Kubernetes/Helm deployments - Cannot manage teams declaratively via ConfigMaps
GitOps workflows - Cannot version control team definitions

For organizations deploying CloudBeaver in Kubernetes with Infrastructure-as-Code practices:

Teams are defined in initial-data.conf ConfigMap

After first deployment, teams are added to MySQL
Updating the ConfigMap has no effect
Teams must be managed manually via UI or direct database manipulation
No declarative way to sync desired state → actual state

Describe the solution you'd like
I would like CloudBeaver to support declarative team management through configuration, similar to how Kubernetes resources work. Teams should be syncable from initial-data.conf even after the initial setup.

Option 1: Continuous Sync (Preferred)
Option 2: Sync on Startup, read initial-data.conf on every startup, not just first time:
Compare desired state vs database state
Apply differences automatically

Describe alternatives you've considered
API Endpoint for Sync

Additional context
Add any other context or screenshots about the feature request here.

Originally created by @mohan-s-gopal on GitHub (Oct 28, 2025). Original GitHub issue: https://github.com/dbeaver/cloudbeaver/issues/3853 **Is your feature request related to a problem? Please describe.** CloudBeaver's initial-data.conf is only processed during the initial startup when the database is empty (configuration mode). After the first initialization, any changes to this configuration file are ignored because teams are persisted in the database and never re-synchronized. This creates a significant operational challenge for: - Kubernetes/Helm deployments - Cannot manage teams declaratively via ConfigMaps - GitOps workflows - Cannot version control team definitions For organizations deploying CloudBeaver in Kubernetes with Infrastructure-as-Code practices: Teams are defined in initial-data.conf ConfigMap 1. After first deployment, teams are added to MySQL 2. Updating the ConfigMap has no effect 3. Teams must be managed manually via UI or direct database manipulation 4. No declarative way to sync desired state → actual state **Describe the solution you'd like** I would like CloudBeaver to support **declarative team management** through configuration, similar to how Kubernetes resources work. Teams should be syncable from `initial-data.conf` even after the initial setup. Option 1: Continuous Sync (Preferred) Option 2: Sync on Startup, read initial-data.conf on every startup, not just first time: Compare desired state vs database state Apply differences automatically **Describe alternatives you've considered** API Endpoint for Sync **Additional context** Add any other context or screenshots about the feature request here.