Server Operations¶

Upgrade, resize, and migrate your cloud server.

Quick Reference¶

Command	Description
`dango remote upgrade`	Upgrade Dango to a newer version
`dango remote resize`	Resize the server (DigitalOcean only)
`dango remote migrate`	Migrate to a new server (DigitalOcean only)

Upgrading Dango¶

Upgrade the Dango installation on your server to a newer version:

dango remote upgrade

This upgrades to the latest version on PyPI. To install a specific version:

dango remote upgrade --version 1.2.3

Upgrade Workflow¶

Version check — compares installed version with target version
Pre-upgrade backup — creates a full backup (unless --skip-backup)
Stop services — stops dango-web and Metabase
Install — runs pip install getdango==<version> in the server's venv
Migrations — applies any database migrations for the new version
Docker rebuild — rebuilds Metabase container if needed
Start services — restarts dango-web and Metabase
Health check — verifies the server is responding

If the installed version already matches the target, the upgrade is a no-op.

Options¶

Option	Description
`--version`	Specific version to install (e.g., `1.2.3`). Must match semantic versioning. Default: latest on PyPI
`--yes`, `-y`	Skip confirmation prompt
`--skip-backup`	Skip the pre-upgrade backup

Failed Upgrade¶

If the health check fails after an upgrade:

# Roll back to the pre-upgrade backup
dango remote rollback

# Check what went wrong
dango remote logs

Installing from Git¶

For pre-release testing, install directly from a Git branch via SSH:

dango remote ssh

Then on the server:

cd /srv/dango
source venv/bin/activate
pip install git+https://github.com/getdango/dango.git@v1

Warning

Installing from Git bypasses the normal upgrade workflow (no backup, no migrations, no health check). Only use this for testing. Run dango remote upgrade to return to a stable release.

Resizing the Server¶

Change the server's CPU, RAM, and disk allocation.

DigitalOceanBYOS

View Current Size and Available Tiers¶

Run without arguments to see your current tier and available options:

dango remote resize

Current: Standard (s-2vcpu-4gb) — 2 vCPU, 4 GB RAM, 80 GB disk, $24/mo

Available tiers:
Tier          Slug            vCPUs  RAM    Disk    Price
Standard      s-2vcpu-4gb     2      4 GB   80 GB   $24/mo
Performance   s-4vcpu-8gb     4      8 GB   160 GB  $48/mo

Run: dango remote resize <SIZE_SLUG>

Resize¶

dango remote resize s-4vcpu-8gb

Resize plan:
  Current: Standard (s-2vcpu-4gb) — $24/mo
  New:     Performance (s-4vcpu-8gb) — $48/mo

Proceed? [y/N]: y

Resizing...
  ✓ Pre-resize backup created
  ✓ Powered off droplet
  ✓ Resize complete
  ✓ Powered on droplet
  ✓ dbt profiles.yml regenerated (threads=4, memory_limit=2GB)
  ✓ Health check passed

Resize complete. New size: s-4vcpu-8gb

What happens:

Creates a pre-resize backup
Powers off the droplet (1–3 minutes downtime)
Resizes the droplet via the DigitalOcean API
Powers on the droplet
Regenerates dbt/profiles.yml to match the new hardware:
- threads = number of vCPUs
- memory_limit = RAM ÷ 4 (e.g., 8 GB RAM → 2GB memory limit)
Verifies health

Disk Never Shrinks

DigitalOcean does not support reducing disk size. If you resize from Performance (160 GB) to Standard, the disk stays at 160 GB. Only CPU and RAM change.

Option	Description
`size`	Size slug (optional — omit to view current info)
`--yes`, `-y`	Skip confirmation prompt

Resize is not available for BYOS deployments. To change your server's resources, resize through your hosting provider's console, then update the dbt profile via SSH:

dango remote ssh

Then edit /srv/dango/project/dbt/profiles.yml to match your new hardware:

Set threads to the number of vCPUs
Set memory_limit to RAM ÷ 4 (e.g., 8 GB RAM → 2GB)

# Example for a 4 vCPU / 8 GB RAM server
my_project:
  target: dev
  outputs:
    dev:
      type: duckdb
      path: /srv/dango/project/data/warehouse.duckdb
      schema: main
      threads: 4
      settings:
        memory_limit: 2GB
        threads: 4

Migrating to a New Server¶

Move your deployment to a new server with a different size or region.

DigitalOceanBYOS

dango remote migrate --size s-4vcpu-8gb

Optionally change region:

dango remote migrate --size s-4vcpu-8gb --region fra1

Migration Workflow¶

Backup — creates a backup of the current server
Upload — uploads backup to DigitalOcean Spaces
Provision — creates a new droplet with the specified size and region
Setup — installs Dango on the new server
Copy secrets — transfers .env and credentials to the new server
Download — downloads backup from Spaces to the new server
Restore — restores data on the new server
Domain — configures domain on the new server (if set)
Firewall — sets up firewall rules on the new server
Health check — verifies the new server is responding
Destroy old — removes the old server after verification

Downtime: 5–15 minutes depending on data size and network speed.

Prerequisites¶

DigitalOcean Spaces must be configured (used for data transfer between servers)
Spaces credentials must be set in the remote .env (see Environment Variables)

DNS Update Required

If you have a custom domain configured, you must update the DNS A record to point to the new server's IP address after migration completes. Dango will remind you of this.

Option	Description
`--size`	Size slug for the new server (required)
`--region`	Region slug for the new server (default: same as current)
`--yes`, `-y`	Skip confirmation prompt

Migration is not available for BYOS deployments. To migrate manually:

Download a backup: dango remote backup download <NAME>
Set up Dango on the new server following the BYOS guide
Copy the backup to the new server and restore

Troubleshooting¶

Upgrade health check failed¶

After a failed upgrade, the server may not be responding:

# Check logs for errors
dango remote logs

# Roll back to pre-upgrade state
dango remote rollback

# If rollback also fails, SSH in directly
dango remote ssh
sudo systemctl restart dango-web
sudo docker compose -f /srv/dango/project/docker-compose.yml up -d

Resize timeout¶

If the droplet takes too long to power back on after resize:

Check the DigitalOcean console for droplet status
Try powering on manually from the DO console
Once running, verify health: dango remote status

Migration Spaces error¶

Error: Spaces not configured

Migration requires DigitalOcean Spaces for data transfer. Set up credentials:

dango remote env set SPACES_ACCESS_KEY=DO00EXAMPLE123
dango remote env set SPACES_SECRET_KEY=wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLE
dango remote env set SPACES_REGION=nyc3
dango remote env set SPACES_BUCKET=my-dango-backups

Disk space warning¶

If dango remote status shows high disk usage:

Remove old dlt pipeline state: dango remote ssh → rm -rf /srv/dango/project/.dlt/pipelines/*/
Consider resizing to a larger disk
Download and remove old backups: dango remote ssh → check /srv/dango/backups/deploy/

pip install failed on server¶

If pip install fails during upgrade (e.g., dependency conflict):

dango remote ssh
cd /srv/dango && source venv/bin/activate
pip install getdango==<version> --force-reinstall

If the venv is corrupted, recreate it:

rm -rf /srv/dango/venv
python3 -m venv /srv/dango/venv
source /srv/dango/venv/bin/activate
pip install getdango==<version>

Then restart services:

sudo systemctl restart dango-web

Next Steps¶

Backups — configure automated backups
Remote Management — push changes and manage the server
Destroy & Cleanup — tear down a deployment