Multi-Server Management

Overview

OllaMan's multi-server feature allows you to connect to multiple Ollama instances, whether they're on your local machine, remote servers, or cloud instances. This is especially useful for teams, managing different hardware setups, or accessing models on various machines.

Adding Servers

Open Settings

Click the Settings icon in the sidebar (bottom-left).

Settings Navigation

Navigate to Servers

In the Settings menu, click on the Servers tab.

Click Add Server

Click the "Add Server" button at the top of the Servers page.

Enter Server Details

Fill in the server information:

Server Name: A friendly name to identify this server (e.g., "Office Server", "Home GPU", "Cloud Instance")

Server URL: The full Ollama server address

Format: http://hostname:port
Examples:
- http://192.168.1.100:11434 (local network)
- http://server.example.com (domain)
- https://ollama.mycompany.com (HTTPS)

Basic Auth (Optional):

Username: If the server requires Basic Auth
Password: The corresponding password

Test Connection

Click the "Test Connection" button to verify the server is reachable.

You'll see:

Connected: Green indicator with server version
Connection Failed: Red indicator with error details

Save Server

If the connection test succeeds, click "Save" to add the server to your list.

Switching Between Servers

Locate Server Selector

At the top of the sidebar, you'll see a dropdown showing the currently active server.

Server Selector

Click on the server selector to see all available servers.

Select Server

Click on any server name to switch to it.

The application will:

Disconnect from the current server
Connect to the selected server
Reload the model list
Update all statistics

Verify Connection

The server selector will show:

Server name
Connection status (green dot = connected)
Server version (on hover)

Automatic Refresh

When you switch servers, OllaMan automatically refreshes:

Local models list
Running models
Dashboard statistics
Available resources

Managing Servers

Edit Server

Open Servers Settings

Navigate to Settings → Servers

Find the Server

Locate the server you want to edit in the servers list.

Click Edit

Click the Edit button (pencil icon) next to the server.

Modify Details

Change any of the server details:

Server name
URL
Authentication credentials

Test and Save

Click "Test Connection" to verify changes, then "Save".

Delete Server

Open Servers Settings

Navigate to Settings → Servers

Find the Server

Locate the server you want to remove.

Click Delete

Click the Delete button (trash icon) next to the server.

Confirm Deletion

Confirm you want to remove this server configuration.

Warning

Deleting a server removes only the connection configuration. Models and data on the actual server remain untouched.

Server Connection States

Connected

Indicator: Green dot

Means:

Successfully connected to the server
Can list and manage models
Can start chats
Real-time statistics available

Disconnected

Indicator: Red dot

Means:

Cannot reach the server
Models list unavailable
Cannot download or chat

Common causes:

Server is offline
Network issues
Incorrect URL
Firewall blocking connection

Testing

Indicator: Yellow/orange dot (pulsing)

Means:

Currently checking connection
Measuring latency
Verifying server version

Use Cases

Home and Work Setup

Scenario: Different models at home (powerful GPU) and office (CPU only)

Solution:

Add "Home Server" pointing to http://192.168.1.50:11434
Add "Office Server" pointing to local http://localhost:11434
Switch based on location

Team Collaboration

Scenario: Development team sharing a powerful remote server

Solution:

Set up a cloud instance with GPUs
Configure authentication
All team members add the server
Everyone accesses the same models

Multi-Cloud Deployment

Scenario: Models distributed across different cloud providers

Solution:

Add "AWS Server", "Azure Server", "GCP Server"
Switch between them based on:
- Cost optimization
- Geographic location
- Model availability
- Performance requirements