Mount Additional Agent Node
Mounting ZebClient Filesystem on Linux Systems
The ZebClient filesystem can be mounted on any Linux-based system, providing a versatile solution for expanding your computational cluster or facilitating file access from different machines within your network infrastructure. Utilizing this feature streamlines the process of file transfer, such as uploading data sets or retrieving computational models.
Utilization Scenarios:
Cluster Expansion: Enhance the capabilities of your computing cluster by integrating additional Linux machines, thereby increasing your resource pool and computational throughput.
Data Exchange: Facilitate the exchange of files, such as machine learning models or extensive research datasets, between workstations and high-performance computing clusters.
Optimizing Data Transfer:
When using the ZebClient filesystem, you can take full advantage of your network's bandwidth. This ensures efficient upload and download speeds, enabling:
Swift upload of research data to the cluster
Rapid retrieval of trained models to local workstations
Prerequisites
A ZebClient agent node is installed and is running on machine that runs
zc-cli
Confirm that the new machine's public address is whitelisted in AWS portal or manually added
Adding a New Agent Node to the ZebClient Cluster
Expand your ZebClient Cluster by seamlessly adding a new agent node. Follow these steps to integrate the new agent node into your existing setup.
This can be performed on cloud-based nodes as well as on-premise machines as long as inter-machine connectivity is confirmed by successful ping responses between all relevant devices.
Step 1: Access ZebClient CLI and Add a Remote Agent Node
Access the ZebClient CLI, choose the option to add a remote agent node, and configure the new agent node.
Step 2: Configure the New Agent Node
Before proceeding, you will be prompted to copy the SHA key to the machine where a new agent will run to allow secure and seamless communication.
This needs to be executed by the user
with permission to use the sudo command.
Example:
Once pasted you can continue on machine running zc-cli
.
Enter the required information:
IP address of the new agent that is accessible by the machine running
zc-cli
:hostname
Username of the new agent:
user
Same user that was used to create the SHA key, permission to use the sudo command is required.Alias for the new agent:
agent1
This can be any name you desire. When deployment finishes successfully you will be able to access the remote node by using the command 'ssh <agent-alias>' from the machine used for ZebClient Start Deployment.
With the necessary data provided, you will be prompted to proceed with your ZebClient deployment:
From this moment zc-cli
will carry on automatically until finished.
Step 3: Verify Connectivity and Copy Files
If you press Y and proceed with the deployment zc-cli
will verify connectivity and obtain the necessary files.
Next,zebclient
user will be created if it doesn't exist on the new agent node and configuration will be automatically tested:
Automated Installation of the new agent on the designated machine.
Once finished you will be brought back to zc-cli
menu.
Final Note
Congratulations! You have successfully added a new agent node to your ZebClient Cluster.
You can access the new remote agent by using the command 'ssh <agent-alias>' from the machine used for ZebClient Start Deployment.
You should be able to access your files immediately in the mount point location (/zcfs
) and use it with your applications.
If you encounter any issues, refer to the troubleshooting section in the Zebware documentation or contact the Zebware support for assistance.
Last updated