Rack View Configuration
Configuring the rack view in the cluster management interface
Rack View Configuration
The rack view in the cluster management interface displays your compute nodes organized by racks and node types. This visual representation helps administrators and users quickly understand the cluster's hardware organization and status.
Editing the Configuration
To configure or update the rack view, edit the infra/node.cfg file in your installation directory.
The path to this file can be customized via the NODEVIEW_CONFIG_PATH environment variable in your .env file:
# Default path
NODEVIEW_CONFIG_PATH="infra/node.cfg"
File Format
The node.cfg file uses JSON format (with support for // line comments) to define rack arrangements and node groupings. Each rack or logical grouping is defined as a key-value pair with the rack name as the key and an object containing node information as the value.
The file also supports an excludedNodes array to hide specific nodes from all dashboard views.
Example structure:
{
"excludedNodes": ["testnode01", "dev001..005"],
"rackLayout": {
"Rack Name": {
"nodes": ["node-range-notation"],
"description": "Description of nodes"
}
}
}
Node Range Notation
For convenience, the configuration supports a range notation for sequential node names:
sc001..056represents nodes sc001, sc002, ..., sc056- Individual nodes can be listed with commas:
sfpga01a, sfpga01i, sfpga01n - Multiple ranges can be combined:
sch001..5, sh001..5
Example Configuration
Here's an example of how the configuration file is structured:
{
"excludedNodes": ["c001..002"],
"rackLayout": {
"Rack 1": {
"nodes": ["c001..56"],
"description": "CPU nodes"
},
"Rack 2": {
"nodes": ["c057..112"],
"description": "CPU nodes"
},
"Rack 3": {
"nodes": ["g001..10"],
"description": "GPU Nodes"
},
"High Memory Nodes": {
"nodes": ["h001..5", "ch001..5"],
"description": "High Memory Nodes"
},
"FPGA Nodes": {
"nodes": ["fpga01a", "fpga01i", "fpga01n"],
"description": "FPGA Nodes"
}
}
}
-
Make your changes following the JSON format shown above.
- Add new racks by creating new key-value pairs
- Modify existing racks by changing their node arrays or descriptions
- Remove racks by deleting their entries
-
Validate your JSON before saving:
- Ensure all brackets and quotes are properly closed
- Check that commas separate entries correctly
- Verify that the file maintains valid JSON structure
After Modifications:
After making any changes to these files, you'll need to rebuild the
application for them to take effect. Run npm run build followed
by npm start (or restart your PM2 process if using PM2).
Troubleshooting
If the rack view fails to load after editing the configuration:
- Check for JSON syntax errors in your configuration file
- Verify that all node ranges are correctly formatted
- Check the application logs for specific errors
- Restore from your backup if necessary