UiPath Automation CloudTM robots - VM, or cloud robots - VM for short, make it easy to set up the necessary infrastructure to run automations.
We handle all the work behind the scenes and give you a virtual machine on which to run automations. The only thing you have to do is configure the virtual machine and run jobs.
You must create a new machine template of the type Cloud Robots - VM.
Machine templates allow you to configure properties that will apply to all Cloud Robots - VM machines created from it. Then, whenever you need machines, we create as many machines as needed that are exactly like the template.
Go to Tenant > Machines.
In the top-right, click Add Machines and select Cloud Robots - VM.
The Add Cloud Robots - VM page opens:
- Fill in the Name field and, optionally, fill in the Description field for the machine template.
- Select the management type for your machine pool:
- Manual - allows you to create and delete machines, use remote desktop, add custom software, or implement custom configurations on your machines. The changes you make persist into future logins.
- Automatic - we create and delete machines for you. Allows you to log in remotely on these machines for debuging purposes, but changes you make are discontinued at logout.
Once the machine pool is created, machine management details become definitive and you cannot edit them.
- Click Next to move to the next step. The VM & Runtime details page is displayed.
- Under Pool Environment, select if you want to run your robots in a Production or a Test/NonProduction environment. Different RU consumption rates apply depending on your selection. See Robot units for more details.
Note: To use Test/NonProduction environment, you must be on the Enterprise, Pro, or Pro Trial licensing plan.
- From the VM size list, choose the type of machine that has the technical specifications you need. For more information, see What machine size should I choose?
- Under Machine Image, select a machine image from the following lists:
- Under IP Address, select if you want the machine to use a dynamic IP address, or use a static one.
By default, VMs use a dynamic IP address, meaning it acquires a different IP address every time it starts. To use the same IP address, change this setting to Static.
- Select a Machine availability profile for your machines. This allows you to recover robot units by shutting down machines that have been idle for more than the selected duration:
- Balanced – the default profile. Your machines shut down after being idle for more than 30 minutes.
- Fast – your machines shut down after being idle for more than 60 minutes. Adds 500 RUs to the balanced cost profile.
- Always – your machines are not being shut down, regardless of their idle duration. Adds 1,000 RUs to the balanced cost profile.
Note: Your selection immediately updates the robot units consumption, so make sure you have enough robot units to run the selected profile.
The machine availability profile applies to all machines in the pool, regardless of their size.
- The Execution slots per machine field is automatically set to
1and you cannot change it. This means any machine created from this template can run only one job at a time.
- Runtime license: Robot units (RUs) - each machine created from this template requires a number of robot units (RUs). Make sure you have enough robot units assigned to your tenant, otherwise you cannot create machines.
Manually managed machines precede automatically managed pools in terms of RU consumption: if there are not enough robot units for all machines, the available RUs are allocated to manually managed machines.
- If you selected the Automatic option in step 4, input the Max number of VMs you want us to create on your behalf. This impacts the number of robot units required to create the automatic pool.
- Click Next to configure the pool.
(Optional) If you want to set a periodic timeframe for installing push updates on VMs that belong to this template, switch on the Enable maintenance window toggle.
The Maintenance Window section is displayed. Here, you can set a span of time when all machines that belong to this template go into maintenance mode and can receive push updates, such a Windows security updates. For more information, see the FAQs.
a. From the Timezone list, select the timezone to use for scheduling the maintenance window.
b. Under Recurrence, select the frequency for the maintenance window and then fill in the fields for the selected option.
c. Under Additional options, select how long you want the window to last.
The window starts as indicated under Recurrence and ends after the amount of time you select here.
d. If you want to make sure that maintenance is performed as scheduled, select the checkbox Kill running processes when window begins.
(Optional) If you have a VPN gateway already set up for this tenant and you want to allow machines that belong to this template to access company resources that are behind a firewall:
a. Switch on the Enable VPN Integration toggle.
The VPN gateway section is displayed.
b. In the Address space for VNet field, add the reserved IP addresses you received from your network administrator, in CIDR notation with a mask of
Click Finish to create the template.
On the Machines page, a success message is displayed along the top and you can see your new template in the list.
After you have configured the manually managed cloud robot pool, you can add machines to it.
- Go to Tenant > Machines.
- At the right of the pool row, click More Actions and select Add or Manage VMs:
The Manage Cloud Robot - VM page opens.
- In the top right, click Add Virtual Machine.
The Add Virtual Machine page opens.
If, at the bottom, the Cost is displayed in red font, this means there are insufficient robot units (RUs) available in the tenant and you cannot create a new machine.
If you are an organization administrator in Automation Cloud, you can click Add more robot units to the tenant balance to go to Admin > Tenants and allocate the needed RUs to this tenant.
Otherwise reach out to your organization administrator to request additional RUs.
- (Optional) Fill in the Alias field with a name to help you distinguish the machine. If you have enough RUs, the Add more VMs option becomes available, allowing you to add multiple machines in the pool and provision them.
4.1. (Optional) Click Add more VMs to add as many VMs as you need.
4.2. If you reconsider the number of VMs, you can remove the ones you do not need by clicking the bin icon in the alias name field.
The RU cost is immediately updated to reflect the number of VMs. If the cost exceeds your RU balance, the Add more VMs option becomes unavailable.
- Fill in the username and password fields with the administrator credentials used to connect to the VM(s). Every VM in the pool shares the same credentials.
- Click Provision. A pop-up box asks you to confirm the number of machines to provision.
Back on the Manage Virtual Machines page, a success message is displayed at the top of the page, and the machines appear in the list with the Provisioning status.
It may take a few minutes to complete provisioning. All machine options are unavailable until the machine is provisioned.
After provisioning is complete, the machine status changes to Running.
While the machine is running, you can connect to it through Windows Remote Desktop and customize its configuration.
- Go to Tenant > Machines.
- At the right of the pool row, click and select Manage Virtual Machines.
- On the Manage Virtual Machines page, check that the Machine Status column shows Running for the machine.
If the machine is stopped, at the right of the row, click and select Start Virtual Machine. Wait a few minutes until the status changes to Running.
- (Optional) If the Accept Jobs column shows Enabled for the machine, click and select Disable Accept Jobs. Wait a few moments for the status to change to Disabled.
We recommend doing this to prevent our auto-scaling engine from shutting down the machine while you are connected to it. This can happen if the machine becomes idle, even if your are connected to it.
- Check that the Remote Desktop column shows Enabled for the machine.
If disabled, at the right of the row, click and select Enable Remote Desktop. Wait until the Remote Desktop value changes to Enabled.
- In the Remote Desktop column, click the info icon to see the IP address for the machine and copy it.
In this step, you can edit the IP address used to access the remote desktop. By inputting a different value than existing one, you can access your remote desktop only from the newly saved IP address.
- Connect to the machine using Windows Remote Desktop as follows:
- use the machine's IP address
- use the admin credentials you defined for the machine; if you do not remember the credentials, you can reset them.
- Now you can customize the machine as needed, by installing or updating software, or by changing its configuration.
The machines we provide have UiPath Robot software and the supported web browsers preinstalled.
Do not modify UiPath software on the machine because these changes may cause problems.
- (Optional) When you finish the configuration, log out of the machine and then disable remote desktop for the machine from Orchestrator.
Remote desktop is automatically disabled on the machine after 24 hours.
- If you disabled Accept Jobs for the machine earlier, on the Manage Virtual Machines page, click and select Enable Accept Jobs to re-enable it.
If you want to use this customized machine for a machine template, you can create a customized image. You can then use the customized image when creating new cloud robot pools, instead of a standard machine image.
To be able to run automations with your new cloud robots, some setup is required in Orchestrator:
A newly-provisioned machine is not set to accept jobs by default. After any configuration you want to perform is complete and the machine is ready for use, you must manually set it to accept jobs.
To allow a machine to be used for running jobs:
- Go to Tenant > Machines.
- At the right of the pool row, click and select Manage Virtual Machines.
- On the Manage Virtual Machines page, at the right of the machine row, click and select Enable Accept Jobs.
A success message is displayed along the top and the Accept Jobs column shows Enabling.
After a few moments, when the status changes to Enabled, the machine can be used to run jobs in the folders where the pool is assigned.
After a machine is set to accept jobs, our auto-scaling engine takes control and has the ability to shut down and start machines automatically.
Add the Cloud Robots - VM machine template to folders which contain unattended automations that you want to run using cloud robots.
Now you and your users can start to run unattended automations using cloud robots.
To run a job using cloud robots: Start a job as you normally would, but for the Runtime license select Cloud Robot - VM.
- Some fields which do not apply when the runtime is Cloud Robot - VM are not available or cannot be edited. Configure job execution using the available options.
- In the case of cloud robots, selecting the Keep Account/Machine allocation on job resumption checkbox means the job resumes on a machine that was created from the same template, not necessarily the exact same machine.
The job will run on one of the available machines in the machine template that is allocated to the folder.
If you have a machine that you have customized according to your automation needs and you want to reuse it, you can create an image of the customized machine. This creates a disk capture that is available to use in a new Cloud Robots - VM template as a customized machine.
To create an image of a customized machine:
- Go to Tenant > Machines.
- At the right of the pool row, click and select Add or manage VMs.
- On the Manage Virtual Machines page, check that the Machine Status column shows Stopped for the machine.
If the machine is running, at the right of the row, click and select Stop Virtual Machine. Wait until the status changes to Stopped.
- Check that the Accept Jobs column shows Disabled for the machine.
If enabled, at the right of the row, click and select Disable Accept Jobs. Wait until the status changes to Disabled.
- At the right of the machine row, click and select Capture Image.
You can create up to 20 custom images. If you cannot create a new one, consider deleting custom images to allow for creating new ones.
The Capture machine as base image dialog opens:
- Fill in the Image Name field and, optionally, also fill in the Description field.
- Optionally fill in the Implicit Username field with the user name you want VMs to use by default.
If you leave the Implicit Username field blank, the username uirobot is used.
- Click Create.
- Restart the machine ( > Start Machine) and then re-enable accept jobs for the machine if you want it to be available for running jobs.
After the custom image is created, you can use it as the base image for a Cloud Robot - VM machine template by selecting Customized for the Machine Image while creating the new template.
You can have up to 20 custom images of cloud robot machines.
If you can no longer create new ones, consider deleting some of your unused images as described below.
To view and manage your customized machine images for Cloud Robot - VM:
- Go to Tenant > Settings and select the Cloud Robot Images tab:
All your existing images are listed on the page.
- At the right of the image row, click and select:
- Edit Image to change the name or description of the image.
- Delete Image if you want to remove the image so that it is no longer available for use.
If you forgot the administrator credentials for one of your machines, you can set a new username and password.
- Go to Tenant > Machines.
- At the right of the pool row, click More Actions and select Add or manage VMs.
- On the Manage Virtual Machines page, at the right of the machine row, click More Actions and select Reset Admin Account.
- On the Reset Admin Account page, define a new username and password for the administrator account of the machine.
The old credentials will no longer work after the reset.
- Click Update.
You can monitor your cloud robots to check for errors on the following pages:
- You can see pending jobs that are waiting for an available machine on the Jobs page.
- You can see the created virtual machines in Orchestrator, on the Monitoring page of the modern folder. When a machine becomes available, it is already connected to Orchestrator, so it runs the next pending job.
If virtual machines are not being created, check the configuration of your cloud robot pool in Orchestrator.
- On the Alerts page in Orchestrator, you can set the State filter to All to see details about the Cloud Robots component and see as new robots are created or any errors. To see alerts, you need to have the Alerts - View and Machines - View permissions from a role that is directly assigned to your account (not inherited from groups), and to be added to the folders where cloud robots are used.
- You can view logs for a particular VM, as described below.
Logs are valuable reports used by administrators for diagnosing and troubleshooting machines. They provide detailed reports on important events happening on a machine such as machine events, job events, or Azure issues.
Logs are kept for 30 days. The oldest log entry is the first one deleted at the end of the retention period.
To view logs for your machine:
- Go to Tenant > Machines.
- At the right of the pool row, click More Actions and select Add or manage - VM.
- On the Manage Cloud Robot - VM page, at the right of the machine row, click More Actions and select View Logs.
- On the Machine Event Logs page, you can see logs generated for that machine, most recent logs displayed first.
- For details about one of the logs, click Show details at the right of the log row. The Log details window shows a detailed report about that specific event.
All standard machines that we provide are Microsoft Azure virtual machines of the type Standard_E2s_v4, which have sufficient computing power for basic automations.
In addition, all standard VMs come with:
- the UiPath Robot software preinstalled (details below)
- the supported web browsers you might need for running automations.
If you need additional software or to set up the VM in a certain way, you can further customize it.
When choosing the machine size, here are some things to consider:
- How large are your datasets?
- What kind of applications are you planning to run and what are their requirements?
- How many applications do you plan to leverage?
- Will you leverage ML Skills and AI Packages?
- What kind of jobs will the machine run - small routine tasks, heavy-duty, one-time setup, debugging?
Based on the answers, you may be able to use a small machine, or may require a more powerful machine.
The machines sizes that you can choose from have the following technical specifications:
|Size||vCPU||Memory (GiB)||Max data disks||Max uncached disk throughput (IOPS/MBps)||Max burst uncached disk throughput (IOPS/MBps)||Max NICs||Expected network bandwidth (Mbps)|
We take care to update your VMs to use the latest version of UiPath Robot within approximately 2 weeks of a new version becoming available.
The update happens:
- when a machine needing an update first starts
- before a machine that was running jobs shuts down
- for machines that are in constant use and are not able to receive the update for 2 weeks since the update was available, we schedule a short maintenance window on the machine to apply the update.
All virtual machines are created and hosted in our Microsoft Azure subscription.
Will the machine images that I created be automatically updated with the latest Windows version and updates?
No. If you want to install the latest Windows version and updates, you can:
- connect to the machine using RDP to manually update Windows.
Then, if you want, you can also create an image of the customized machine and use it in other machine templates.
- configure a maintenance window for the template when you can push updates to all machines.
Only machines that belong to the machine template where the maintenance window was configured and for which Accept Jobs is Enabled enter the maintenance window.
Machines that are not running are automatically started for the maintenance window.
The required robot units are consumed once per month, when the machine first starts. After that, it can run indefinitely for the whole month without consuming additional robot units.
We automatically disable the most recently-created machines if they consume more robot units than you have available for your tenant.
You must add sufficient robot units for that machine to the tenant. These will be consumed and your disabled machine will be automatically re-enabled within approximately 30 minutes.
If you do not allocate sufficient robot units to re-enable all of your disabled machines, only some machines are re-enabled. We start with the oldest machine and move up to the newest.
After being re-enabled, a machine is in a stopped state. To use it to run jobs, you must start the machine.
Yes. If you have disabled machines and add robot units to your tenant, they are automatically consumed to re-enable your disabled machines.
If, for example, you want to allocate some more robot units to a tenant for Automation Cloud robots - serverless, you need to make sure that you don't have disabled machines that would consume the robot units.
To prevent this, you must delete the disabled cloud robots - VM machines before you allocate the robot units to the tenant.
When defining the automatic machine template, make sure you have enough robot units (RUs)on your machines. The machine pool cannot consume robot units partially.
If you do not have enough RUs, the following consumption restrictions apply:
- If your robot units do not match to the maximum number of VMs defined in the pool, we delete all machines in the pool and stop creating new ones until you allocate enough robot units to support the maximum number of machines.
Instead of adding more robot units, we recommend reducing the maximum number of VMs in the pool.
- A job running on a machine that does not have enough RUs generates the following alert "No VMs in <Pool_name> due to insufficient robot units."
- As soon as enough robot units become available, they are automatically consumed.
- If you have multiple pools in an overconsumption state, we allocate any available robot units to the last created pool subsets.
Example: You have five pools, each having a maximum of three machines, totaling 15 machines. Your RUs can support two machines, meaning all five pools are now in an overconsumption state, so you cannot use them.
- You add the required RUs to support five more machines. Now, you can use a total of seven machines.
- Two pools become available and they consume RUs for six machines (two pools of three machines each).
- Three pools remain in the overconsumption state, and the available RUs serve one machine. Therefore, the two pools are created last.
- Once automatic pools are created, the allocated RUs are consumed.
- When a pool is deleted, the corresponding RUs are released and can be consumed by other pools.
To avoid managing different accounts on each VM, and to use a single set of credentials for authentication, you can join your VMs to a domain. This way you can integrate your machines with an existing identity infrastructure, and your cloud robots can run jobs under a domain user.
There are three mainly used identity solutions, summarized in the following table, along with the corresponding domain connection approach and the current availability in UiPath:
|Identity solution||Availability in UiPath||Connection to the domain|
|Active Directory Domain Services (AD DS)||via site-to-site VPN Gateway|
|Azure Active Directory Domain Services (Azure AD DS)||via site-to-site VPN Gateway|
|Azure Active Directory|
This solution is currently under development.
Use this solution to manage identity and access in on-premises environments.
AD DS allows your cloud robots to authenticate and access on-premises network resources via on-premises domain controllers. The site-to-site connection to the domain controller is established through the VPN Gateway service.
You continue to maintain all of the associated infrastructure and directory components.
To integrate your on-premises Active Directory with Azure, connect your on-premises environment to an Azure AD domain in the cloud via site-to-site VPN Gateway.
Read the Azure documentation for more details.
Use this solution to manage domain services such as domain join, group policy, LDAP, or Kerberos/NTLM authentication, the same as you would use for an on-premises Active Directory.
Azure AD DS allows your Automation Cloud robots to authenticate and access cloud network resources via Azure AD DS managed domains. Additionally, you can join and manage VMs in Azure. The site-to-site connection to the Azure AD DS managed domains is established through the VPN Gateway service.
You do not need to deploy or maintain the AD DS infrastructure for components such as VMs, Windows servers, or domain controllers.
To join a Automation Cloud Robot - VM to an Azure AD DS managed domain:
The following procedure is the same for connecting a computer to an on-premises AD DS domain.
- Sign in to your VM.
- Access Server Manager.
- On the left-side panel of the Server Manager, select Local Server, then select WORKGROUP.
- In the System Properties > Computer Name tab, click Change. The Computer Name/Domain Changes window opens.
- In the Domain field, specify the name of your managed domain. For example,
- To join the domain, provide the credentials of a user that is part of the managed domain.
Note 1: The user account must exist in your managed domain or in your Azure AD tenant. During domain joining, accounts from external directories associated with your Azure AD tenant cannot authenticate correctly.
Note 2: Account credentials can be specified either in the UPN format, which is the recommended format, or in an SAMAccountName format. For example, user "VMadmin" in UPN format is "email@example.com", and in SAMAccountName format is "AADDSCONTOSO\vmadmin".
- Click OK. Wait a few seconds until the VM is joined to the managed domain. If successful, a welcome message box is displayed.
- Restart the VM to complete the domain-join process.
Updated 7 days ago