Installing an Agent

You can also install an agent in your organization to use one agent in multiple tenants. Learn more here.

System Requirements

Name Value
Minimum RAM 500 MB
Recommended RAM 1GB
Supported Operating System Types 64 bit
Supported Windows Server Versions 2008 R2 SP1 & later
Support Linux Variants View Here

After you have selected your agent details, you can download the agent. The Agent installer is a single Powershell Script that will download the necessary run times, the newest version of the agent and it will also pre-configure your agent to connect to this tenant.

The install script will automatically be installed as a Local System account. To change the account settings, you will need to modify the service. (You can read about modifying the service in the next section.)

Prerequisites:

  • After you’ve downloaded the script, make sure to run the script with administrator privileges.
  • Windows users will need to ensure they have Powershell 4 at a minimum installed.
  • Linux users will need to install Powershell Core for their platform.

Initial Steps for All Platforms

From the homepage, click on the Tenant name in the top-right corner. From the resulting dropdown menu, select Agents.

Agent in Settings

On the Agents page, select Setup Agent.

Setup Agent

Click on your chosen platform from the tabs at the top of the page. This is the environment where the agent will be deployed.

Agent Details for All Platforms

Provide a name for the agent. This should be unique and easily identifies this agent. It may only contain letters, numbers, dashes, and dots.

Choose an Expiry Date. You can also click on the calendar on the right of the field to choose from a date selector.

Provide a description for the agent.

On the next page, choose to add networking configuration:

  • You can choose to add your own client IPv4 address at the top-right of the table.
  • You can also add firewall rules. You can add a range of IPv4 addresses.
    • Provide a name for the firewall rule, and then choose a start and end IPv4 address.

If applied, the agent can be accessed from these IPv4 addresses.

Next, review your agent details and then click on Create Agent to create the agent.

Installation Instructions for Windows and Linux

Follow the instructions on the next screen to install and configure your agent on your selected environment.

You can then follow the prompts in the installation wizard for your chosen platform.

The following sections contain more information for other platforms.

Windows

  1. Open PowerShell as an Administrator in the folder that contains the new Loome agent installer.
  2. Copy and run the command, which you can copy from Loome once you have chosen a platform, into your (Administrator) PowerShell Window.

If other Loome agent installer files are present in this folder, you must run the newest file. The best way to avoid any confusion is to remove all installer files before downloading a new agent.

Linux

  1. Open an elevated shell, and then run PowerShell Core.
  2. Copy and run the command from Loome.

Silent Install

  • You can also use the script with the silent install option. It will proceed with the agent installation by selecting the default options for the prompts. It will choose “Y” for confirming the installation of the agent, “N” for installing SnowSQL (for use with Snowflake Cloud Data Warehouse), and “N” for using Microsoft Genomics.
    • To use silent install, select the checkbox beneath the install script.
    • Copy the script and paste it into your Administrator PowerShell installation window and run the script.

Run the Script

After you have run the script, it will download the newest version of the agent whilst removing older versions for the tenant. (If you are upgrading an agent, the script will remove the previous version.)

You may be prompted to install additional components during the installation. These are optional components. (If you have used silent install, these prompts will be installed as described above.) If you decide you want to install an optional component later, you can run the installation script again.

It will start the service and complete installation.

You are then ready to use this agent.

If the agent is unhealthy in Loome, first refresh the page and if it is still unhealthy please restart the service.

Modifying the Service

First, open up Services on your computer.

Find the agent you just installed with the same name listed in Services. It will be loome-agent followed by your unique agent code.

Once you’ve found your agent, open up the log on window.

Change your account log on here

Change the account here.

Once you save, the agent will now use this account.

Installation Locations

Loome Integrate Agents are installed in a standard directory, based on a unique ID. All agent installs can be found in %ProgramData%\LoomeAgent on Windows and /.loomeagent/ on Linux. There is no need to take note of this directory when updating the agent, as the install script will automatically detect if an agent is installed for the related tenant.

Running Without a Service

Whilst we highly recommend you run the agent as a service, you are able to run it as a console application you startup when necessary. Agents can be run as a console application using the Run.ps1 powershell script located in their install directory.

Windows Considerations

If the agent is running as a service, you will need to provide an account with “Logon as a Service” functionality.

When providing the username and password credentials for the service, ensure that you include the domain as part of the username. Eg: “Company\User” or “DeviceName\LocalUser”

Agent services can be viewed in the Windows Services manager under the name loome-agent-[unique identifier for the agent] (eg:loome-agent-b0...-f6e78).

Linux Considerations

“loomeusers” Group

The installer will create a new group called loomeusers on install, this group is used for managing file permissions in the /.loomeagent directory and so it is a requirement that the group is created.

Installing the Agent as a Container

Loome Integrate also supports running the agent using Docker. The official Loome Integrate Agent image is available from the Docker Hub at loomesoftware/agent. Docker Agents are configured entirely through environment variables/secrets so there is no need to mount additional volumes for configuration files. Using the Docker installation screen, you can generate shell commands that will pre-configure the Docker based agent to automatically run and connect to your tenancy.

Once installed, this Docker agent will include PowerShell 7.

To update agents being hosted in Docker, just run the same installation script again.

Whilst Loome Integrate only generates commands for certain installation options, the agent is compatible with any Docker based platform.

Using Docker Run

After you have provided the above agent details, the Docker Run tab provides a generated command for installing the agent using docker run. This command by default will run the agent as a detached background service which will automatically restart with the host PC’s restart.

Whilst the Docker Run command provides all environment variables inline with the -e switch, these environment variables can be used to host the agent using other means such as an envlist or orchestration engines like Docker Swarm and Kubernetes.

Using Azure Container Instances

The Azure Container Instances (ACI) tab provides a generated command for create an ACI resource for the agent.

After you have provided the initial agent details, as described above, you will need to choose where the agent will be deployed and the size of the container.

Container Details

Learn about the prerequisites before installing an agent on an Azure Container Instance to use an agent in Loome Assist.

Provide the details needed for where you want to deploy the agent:

  • Provide a Container Name.
  • Choose an Azure Region from the dropdown.
  • Provide the name of an existing resource group where this container will be provisioned in the Resource Group field.
  • Choose where the Loome agent image should be pulled from in the Docker Image Location dropdown.

Next, optionally choose the size of the container instance. You can change the provided values if you want to modify the resources available to the agent. If you do not change these values, the size of the container instance will be the default size.

  • Provide a number of CPU Cores to allocate to your container. (You can choose between 1 - 8 CPU Cores.)
  • Provide the amount of memory to allocate to your container. (You can choose between 1 - 32 GB.)

Azure Container Configuration

Next, you can specify optional configuration for the agent.

Choose whether you want to:

  • Specify external storage for your container instance.
    • By default, we store files, such as logs, inside the container instance. You can choose to supply an Azure Storage Account, which will allow you to easily retrieve and view log files.
    • Click on the checkbox for Use an Azure Storage Account to specify this.
      • Then provide:
        • ‘Storage Account Name’ - The name of the Azure Storage Account where you want to store agent files.
        • ‘Storage Account Key’ - One of the access keys for your storage account.
        • ‘Fileshare Name’ - The name of the Azure Fileshare where you want to store agent files.
  • Specify Azure Virtual Networking for your Container Instance.
    • If you do not use Azure Virtual Networks you can skip this step.
    • Click on the checkbox for Use Azure Virtual Networking to specify this.
      • You can then:
        • You can supply the names or IDs of an existing Azure Virtual Network and Subnet in Azure Virtual Network ID or Name or Azure Subnet ID or Name
          • If the Azure Virtual Network and Subnet reside in the same Resource Group, you can provide the names of Azure Virtual Network and Subnet.
          • If the Azure Virtual Network and Subnet reside in a different Resource Group, you will need to use the fully qualified Azure Resource IDs of the Azure Virtual Network and Subnet.
  • Assign a User Assigned Identity to your Container Instance.
    • By default, we will enable a System Assigned Identity and you can use this Identity to provide the Loome Agent access to resources in Azure.
    • If you prefer to use a User Assigned Identity, you can provide the fully qualified Azure Resource ID in the Azure Resource ID of the User Identity field.

After you have configured the Agent with a Managed Identity, you will need to provide it with permissions to provision and manage resources within the Subscription. Learn about the roles and permissions you will need to provision here, as well as the reasons these permissions are required.

Azure Container Command

After you have provided these agent details and chosen the optional configuration, you can proceed to the next page.

The command provided needs to be run in an instance of the Azure Cloud Shell or a connected Azure CLI, a valid resource group name is also required for billing purposes.

SQL Server may require a firewall exception for the container. To get a containers public IP address, execute the following command in the Azure Container instance: curl ifconfig.co