Understanding Requirements Checker rules and results
This article is designed to provide you with a clear understanding of the tests conducted by the Workspace DNA Requirements Checker, the significance of the items they evaluate, and the appropriate steps to take when a rule does not pass.
This article explains the WorkspaceDNA Requirements Checker rules, what each one evaluates, and what to do when a rule fails. The Requirements Checker runs a set of automated tests against the Gateway, Task Runners, and the surrounding Azure environment to confirm that everything the WorkspaceDNA service depends on — network connectivity, identity and permissions, storage, Key Vault, and local machine configuration — is in place before and during deployment.
Most rules are advisory: if one fails, deployment can continue but the underlying issue should still be resolved. A small number of rules are marked as blockers — deployment cannot proceed until these are resolved.
Jump to a Specific Rule
- Rule: Check connectivity to the WorkspaceDNA Web Host
- Rule: Check connectivity to the WorkspaceDNA Web Host storage account
- Rule: Check connectivity to the WorkspaceDNA Patch Catalog storage account
- Rule: Check connectivity to the Azure Instance Metadata Service
- Rule: Check connectivity to the Azure platform channel on port 80
- Rule: Check connectivity to the Azure platform channel on port 32526
- Rule: Ensure a managed identity token can be retrieved for the Azure Resource Manager
- Rule: Ensure name resolution works through the Azure platform DNS server
- Rule: Check connectivity to the Azure Resource Manager
- Rule: Check connectivity to Microsoft Graph (only required when Intune or Cloud PC features are used)
- Rule: Check connectivity to Azure KMS for Windows activation (only required for marketplace images)
- Rule: Check connectivity to the alternate KMS endpoint (only required for marketplace images)
- Rule: Ensure the video streaming server is accessible
- Rule: Check connectivity to the video streaming API
- Rule: Ensure the Gateway is listening for inbound video on port 443
- Rule: Ensure that system assigned identity is enabled
- Rule: Ensure the Gateway VM's identity can manage Task Runners in the WorkspaceDNA Resource Group
- Rule: Ensure the Gateway VM's identity can manage Task Runner public IP addresses (only required when Task Runners use public IPs)
- Rule: Ensure the Gateway VM's identity can grant and revoke the roles a Task Runner needs
- Rule: Ensure the Gateway VM's identity can read and attach Task Runners to the Gateway's Vnet
- Rule: Check that the Gateway's subnet has enough unused IP Addresses
- Rule: Ensure the Gateway is accessible on port 5000
- Rule: Ensure the Gateway is accessible on port 5001
- Rule: Check if any local firewall profiles are enabled
- Rule: Ensure an Inbound firewall rule for port 5000 exists (if firewall is enabled)
- Rule: Ensure an Inbound firewall rule for port 5001 exists (if firewall is enabled)
- Rule: Ensure the Gateway control channel can be upgraded to a WebSocket
- Rule: Ensure the Gateway agent control channel can be upgraded to a WebSocket
- Rule: Check that the auto login user is an administrator
- Rule: Check connectivity to the signing timestamp service (only required for auto-remediation signing)
- Rule: Ensure that a legal notice is not configured
- Rule: Ensure that a screen saver is not enabled
- Rule: Ensure that the automatic lock screen is not enabled
- Rule: Ensure a storage account was created
- Rule: Check connectivity to the WorkspaceDNA storage account over HTTPS
- Rule: Check the application package file share is reachable over SMB 3
- Rule: Ensure the storage account can be read from and written to using the machine's identity
- Rule: Ensure the Gateway VM's identity can read and write blob data on the storage account
- Rule: Ensure the Gateway VM's identity can read the storage account's configuration and keys
- Rule: Ensure the application package share is mapped as the T drive
- Rule: Ensure a key vault was created
- Rule: Check connectivity to the WorkspaceDNA Key Vault
- Rule: Ensure the Gateway VM's identity can manage access policies on the key vault
Rule: Check connectivity to the WorkspaceDNA Web Host
The Gateway communicates securely with the Web Host (rimo3cloud.com) over HTTPS on Port 443, therefore access to this URL and Port should not be blocked by endpoint security products, or network appliances/resources such as Network Security Groups, Firewalls or Proxy Servers.
Blocker: This check is critical for the Change360 service. Deployment cannot proceed if it fails. After resolving issues, run "Run Checks" in the Deployment Wizard to validate changes.
What to do if this rule is failing:
- Verify perimeter security devices (firewalls, proxies) aren't blocking traffic to https://rimo3cloud.com on port 443
- If SSL Inspection is employed, ensure required certificates are installed on the Gateway
- Failure pattern: "An attempt was made to access a socket in a way forbidden by its access permissions" typically indicates local firewall or endpoint protection blocking access
Rule: Check connectivity to the WorkspaceDNA Web Host storage account
The Web Host storage account serves two purposes: it provides access to download various platform components and acts as a quarantine area for uploaded packages, therefore access to the storage account on Port 443 (HTTPS) should not be blocked by endpoint security products, or network appliances/resources such as Network Security Groups, Firewalls or Proxy Servers.
What to do if this rule is failing:
- Review perimeter security appliances for blocked traffic to https://r3cjupitersa.blob.core.windows.net on port 443
- For SSL Inspection environments, verify required certificates are installed on the Gateway
- Failure pattern: "An attempt was made to access a socket in a way forbidden by its access permissions" suggests local firewall or endpoint protection interference
Rule: Check connectivity to the WorkspaceDNA Patch360 repository
The Patch360 repository contains the package sources for all applications listed in Patch360, therefore access to the repository, an Azure storage account, on Port 443 (HTTPS) should not be blocked by endpoint security products, or network appliances/resources such as Network Security Groups, Firewalls or Proxy Servers.
What to do if this rule is failing:
- Examine perimeter security appliances for blocked traffic to https://r3cjupiterpatchsa.blob.core.windows.net on port 443
- When SSL Inspection is active, ensure required certificates exist on the Gateway
- Failure pattern: "An attempt was made to access a socket in a way forbidden by its access permissions" typically shows local firewall or endpoint protection restrictions
Rule: Check connectivity to the Azure Instance Metadata Service
Anything that uses the machine's managed identity obtains its token from the Azure Instance Metadata Service on this link-local address, so if it is unreachable the Gateway and Task Runners cannot authenticate to Azure at all. A Network Security Group cannot block a link-local address, but a host firewall, an endpoint protection product or Group Policy can. This traffic must also bypass any proxy that is configured on the machine.
What to do if this rule is failing:
- Check that no host firewall, endpoint protection product, or Group Policy setting is blocking outbound access to 169.254.169.254 on port 80 — since this is a link-local address, a Network Security Group cannot be the cause
- Confirm the request is bypassing any proxy configured on the machine, as this traffic must not be routed through a proxy
- If this rule fails, expect the managed-identity token rule and DNS rule below to fail too, since both depend on this endpoint
Rule: Check connectivity to the Azure platform channel on port 80
This address is the Azure platform's communication channel to the virtual machine, used for platform health reporting and for the guest agent. It must be reachable on port 80 and must bypass any proxy configured on the machine.
What to do if this rule is failing:
- Check that no host firewall, endpoint protection product, or Group Policy setting is blocking outbound access to 168.63.129.16 on port 80
- Confirm this traffic is bypassing any proxy configured on the machine
- If this fails alongside the DNS rule below (which also targets 168.63.129.16), treat it as the same underlying block rather than two separate issues
Rule: Check connectivity to the Azure platform channel on port 32526
Azure VM extension handling uses port 32526 on the platform channel. The Task Runner setup package is delivered as a VM extension, so if this port is blocked a Task Runner can be provisioned but will never finish configuring itself. This traffic must bypass any proxy configured on the machine.
What to do if this rule is failing:
- Check that no host firewall, endpoint protection product, or Group Policy setting is blocking outbound access to 168.63.129.16 on port 32526
- Confirm this traffic is bypassing any proxy configured on the machine
- If Task Runners are being created but never finish configuring, check this rule first, since it is the most likely cause of that specific symptom
Rule: Ensure a managed identity token can be retrieved for the Azure Resource Manager
Reaching the Instance Metadata Service only proves the address answers. This rule asks it for an access token, which is what the Gateway and Task Runners actually do whenever they call Azure, and reports the client id of the identity the token was issued to. If it fails while the reachability check passes then the endpoint is available but the machine has no usable managed identity, which is a different problem: check that the system assigned identity is enabled on the virtual machine.
What to do if this rule is failing:
- If the Azure Instance Metadata Service reachability rule above passes but this rule fails, the issue is the identity itself, not the network path — go to "Ensure that system assigned identity is enabled" below and confirm the system assigned identity is switched On
- Check whether a user assigned identity is also present on the VM, as this can interfere with system assigned identity detection
- Note the client id reported by this rule when raising a support request, as it identifies which identity the token was issued to
Rule: Ensure name resolution works through the Azure platform DNS server
The Azure platform provides name resolution on 168.63.129.16, and the guest agent depends on it. The machine's own resolver can be working perfectly while that server is unreachable, so this rule asks it directly rather than going through whatever resolver the machine is configured to use. A failure here alongside a passing reachability check on the same address means the port is open but the server is not answering queries.
What to do if this rule is failing:
- If "Check connectivity to the Azure platform channel on port 80" above passes but this rule fails, the port is open but DNS queries themselves are not being answered — check for a host firewall, endpoint protection product, or Group Policy setting that specifically filters DNS traffic to 168.63.129.16, rather than one blocking the address outright
- Confirm the machine is not forcing all DNS queries through a different, non-Azure resolver that is unable to reach 168.63.129.16
Rule: Check connectivity to the Azure Resource Manager
The Gateway calls the Azure Resource Manager to create and remove Task Runner virtual machines, so if this endpoint is unreachable no Task Runner can be provisioned. Where Azure Firewall or Network Security Group rules are used, the AzureResourceManager service tag is preferable to allow-listing the name.
What to do if this rule is failing:
- Verify perimeter security devices aren't blocking traffic to https://management.azure.com on port 443
- Where Azure Firewall or Network Security Group rules are used, allow the AzureResourceManager service tag rather than the individual hostname
- If this rule fails, expect Task Runner creation and removal to fail with it, since the Gateway cannot reach the Resource Manager to perform either action
Rule: Check connectivity to Microsoft Graph (only required when Intune or Cloud PC features are used)
Microsoft Graph is used by two separate features, and either one on its own is reason to need this endpoint. Applications pushed to Intune are exported through Graph from the Task Runner, and Cloud PC snapshot and restore call it from the Gateway - so this requirement applies to both machines, not only to the Task Runner, and a customer who uses Cloud PC without Intune needs it just as much as one who uses Intune without Cloud PC.
The two features call different versions of the API, the beta version for Intune and version 1.0 for Cloud PC, but both on the same host name, so a single allow-list entry covers them. Both authenticate with an application registration in the customer's own tenant rather than with the machine's managed identity, which means the access token is obtained from the Microsoft Entra ID token endpoint and not from the Instance Metadata Service - that endpoint has to be reachable as well, and when it is blocked while this rule passes, every Intune and Cloud PC operation fails while acquiring its token and never reaches Graph at all.
This rule can be disregarded only by customers using neither feature. Bypassing TLS inspection is recommended for this endpoint because the large chunked uploads used for application content commonly fail under inspection.
What to do if this rule is failing:
- If the Intune and Cloud PC features are not in use, this rule can be disregarded
- If either Intune or Cloud PC features are in use, verify perimeter security devices aren't blocking traffic to https://graph.microsoft.com on port 443
- Bypass TLS/SSL inspection for this endpoint, since large chunked application uploads commonly fail when an inspecting proxy is in the path
Rule: Check connectivity to Azure KMS for Windows activation (only required for marketplace images)
Virtual machines created from Azure marketplace Windows images activate against the Azure Key Management Service on Port 1688. This rule is only relevant when marketplace images are used. The traffic is not HTTP and cannot be proxied, and there is no service tag for it, so it must be allowed by name or address. If it is blocked the Task Runner will report Windows as not activated.
What to do if this rule is failing:
- If the deployment is not using Azure marketplace Windows images, this rule can be disregarded
- If marketplace images are in use, allow outbound traffic to azkms.core.windows.net on port 1688 by name or address directly — this traffic is not HTTP, cannot be routed through a proxy, and has no service tag available
- A Task Runner reporting Windows as not activated is the expected symptom when this endpoint is blocked
Rule: Check connectivity to the alternate KMS endpoint (only required for marketplace images)
This is the second of the two Key Management Service endpoints used by Azure marketplace Windows images for activation on Port 1688, checked separately so that it is clear which of the two is unreachable. As with the other endpoint the traffic is not HTTP, cannot be proxied and has no service tag.
What to do if this rule is failing:
- If the deployment is not using Azure marketplace Windows images, this rule can be disregarded
- If marketplace images are in use, allow outbound traffic to kms.core.windows.net on port 1688 by name or address directly, in the same way as the primary KMS endpoint above
- Check this rule alongside "Check connectivity to Azure KMS for Windows activation" — since both are checked separately, compare the two results to see whether one or both endpoints are blocked
Rule: Ensure the video streaming server is accessible
Video is securely streamed to the Video Server (wowzastag.rimo3activ.com) over RTMPS on Port 443 therefore access to this URL and Port should not be blocked by endpoint security products, or network appliances/resources such as Network Security Groups, Firewalls or Proxy Servers.
What to do if this rule is failing:
- Check perimeter security appliances for blocked traffic to https://wowzastag.rimo3activ.com on port 443
- If SSL Inspection applies, verify required certificates are installed on the Gateway
- Failure pattern: "An attempt was made to access a socket in a way forbidden by its access permissions" indicates local firewall or endpoint protection blocking
Rule: Check connectivity to the video streaming API
The Video Server API endpoints (e.g. start recording, stop recording) are accessed via wowzastag.rimo3activ.com over HTTP on Port 8087 therefore access to this URL and Port should not be blocked by endpoint security products, or network appliances/resources such as Network Security Groups, Firewalls or Proxy Servers.
What to do if this rule is failing:
- Review perimeter security appliances for blocked traffic to http://wowzastag.rimo3activ.com on port 8087
- When SSL Inspection is implemented, confirm required certificates are on the Gateway
- Failure pattern: "An attempt was made to access a socket in a way forbidden by its access permissions" suggests local firewall or endpoint protection interference
- Test from the Gateway VM by browsing to http://wowzastag.rimo3activ.com:8087 and verifying you receive a login prompt, click cancel and you should then receive a 401 authentication error message rather than a blocked connection
Rule: Ensure the Gateway is listening for inbound video on port 443
Screen recordings are relayed from Task Runners to the Gateway over RTMPS on port 443, so the Gateway itself has to be listening on that port to accept the inbound stream. This is a different check from the firewall rules: a firewall rule can permit the traffic while nothing is there to receive it, and the symptom in both cases is simply no screen recording. This rule is only relevant when screen recording is enabled.
What to do if this rule is failing:
- If screen recording is not enabled, this rule can be disregarded
- Confirm the Gateway service responsible for receiving the video relay is running and actually listening on port 443, rather than only checking that firewall rules allow the port
- If firewall and network checks for port 443 all pass but screen recordings are still not being captured, this rule — not a firewall rule — is the most likely cause, since it is checking whether anything is listening, not whether traffic is permitted
Rule: Ensure that system assigned identity is enabled
A system assigned identity is used to grant the GW VM rights to perform actions in the WorkspaceDNA resource group. If the system assigned identity fails to get enabled it can be manually enabled in the Azure Portal.
Blocker: This check is essential for the Change360 service. Deployment cannot proceed if it fails. After making corrections, run "Run Checks" in the Deployment Wizard to confirm updates.
What to do if this rule is failing:
- Access the Gateway VM in Azure and verify system assigned identity under Security → Identity is set to On
- Check whether the Gateway VM has a user assigned identity under Security → Identity → User assigned, as this can interfere with system assigned identity detection
Rule: Ensure the Gateway VM's identity can manage Task Runners in the WorkspaceDNA Resource Group
The Gateway creates and removes Task Runner virtual machines in the WorkspaceDNA Resource Group, which needs the permissions listed below. Provisioning a Task Runner is more than the virtual machine itself: it also creates and removes the machine's disk and its network interface, and delivers the setup package as a virtual machine extension, so a role granting only the virtual machine actions leaves provisioning failing partway through.
This checks what the machine's identity can actually do rather than which role it holds, because a role name is only a proxy for its permissions - any role granting these actions at this scope or above satisfies the rule, and the message names precisely which permission is missing when one is. Virtual Machine Contributor grants all of them, as does Owner.
The required actions are:
- Microsoft.Compute/virtualMachines/write
- Microsoft.Compute/virtualMachines/delete
- Microsoft.Compute/virtualMachines/extensions/write
- Microsoft.Compute/disks/write
- Microsoft.Compute/disks/delete
- Microsoft.Network/networkInterfaces/write
- Microsoft.Network/networkInterfaces/delete
- Microsoft.Network/networkInterfaces/join/action
Blocker: This check is critical for the Change360 service. Deployment cannot proceed if it fails. After resolving issues, run "Run Checks" in the Deployment Wizard to validate changes.
What to do if this rule is failing:
- Confirm the Gateway's system assigned identity was successfully created (see related rule above)
- Manually grant the Gateway system assigned identity the Owner role, or a custom role with the above permissions, on the Change360 resource group using Azure Portal procedures.
Rule: Ensure the Gateway VM's identity can manage Task Runner public IP addresses (only required when FUAT Task Runners use public IPs)
When a Facilitated UAT Task Runner configuration asks for a public IP address the Gateway creates one alongside the machine and removes it again afterwards, which needs the permissions listed below in the WorkspaceDNA Resource Group. This is checked separately from the other Task Runner permissions and only warns, because customers who don't use Facilitated UAT will never need it. It is worth checking even so: Virtual Machine Contributor, which covers everything else a Task Runner needs, grants only read and join on public IP addresses and not write or delete, so a deployment that turns Facilitated UAT on later may start seeing issues with nothing else having changed. Network Contributor grants all three actions, as does Owner.
The required actions are:
- Microsoft.Network/publicIPAddresses/write
- Microsoft.Network/publicIPAddresses/delete
- Microsoft.Network/publicIPAddresses/join/action
This rule depends on "Ensure that system assigned identity is enabled" above having passed.
What to do if this rule is failing:
- If Facilitated UAT Task Runners do not use public IP addresses, this rule can be disregarded — it only returns a warning
- If Facilitated UAT Task Runners do use public IP addresses, confirm the Gateway's system assigned identity was successfully created (see "Ensure that system assigned identity is enabled" above)
- Manually assign the Gateway system assigned identity the Network Contributor role, Owner role, or custom role with the above permissions on the WorkspaceDNA resource group — Virtual Machine Contributor alone is not enough, since it grants read and join on public IP addresses but not write or delete
Rule: Ensure the Gateway VM's identity can grant and revoke the roles a Task Runner needs
Each Task Runner gets its own system assigned identity, and the Gateway grants that identity the roles it needs to reach the storage account and the key vault — Storage Blob Data Contributor, Storage Account Contributor, Reader, Key Vault Reader and Key Vault Secrets User — then revokes them when the Task Runner is removed. Creating and removing a role assignment needs the permissions listed below in the WorkspaceDNA Resource Group .
This is the permission most easily lost when a deployment is hardened: the MPO grants the Gateway Owner, which covers it silently, and a least privilege role set assembled from the resource specific roles does not — none of Virtual Machine Contributor, Network Contributor, Storage Account Contributor or Key Vault Contributor grants it. The symptom is a Task Runner that provisions and starts cleanly and then cannot read the storage account or the key vault, which looks like a storage or vault problem rather than a Gateway one. Role Based Access Control Administrator grants these actions, as do Owner and User Access Administrator.
Note, what a pass here does and does not mean: this reads the actions the identity is granted, and Azure does not report the ABAC conditions attached to them, so an assignment that is restricted to a specific list of roles — the recommended way to grant this — passes this rule whether or not its condition actually permits the five roles above. If this rule passes and Task Runners still come up without access, check the condition text on the Gateway's Role Based Access Control Administrator assignment.
The required actions are:
- Microsoft.Authorization/roleAssignments/write
- Microsoft.Authorization/roleAssignments/delete
This rule depends on "Ensure that system assigned identity is enabled" above having passed.
Blocker: This check is critical for the Change360 service. Deployment cannot proceed if it fails. After resolving issues, run "Run Checks" in the Deployment Wizard to validate changes.
What to do if this rule is failing:
- Confirm the Gateway's system assigned identity was successfully created (see "Ensure that system assigned identity is enabled" above)
- Manually assign the Gateway system assigned identity the Role Based Access Control Administrator role (or Owner, or User Access Administrator) on the WorkspaceDNA resource group — none of Virtual Machine Contributor, Network Contributor, Storage Account Contributor or Key Vault Contributor grant this permission, so a least-privilege role set assembled only from those will fail this rule even though the Gateway appears to have everything else it needs
- If Role Based Access Control Administrator has been scoped with a condition restricting which roles it can assign (the recommended way to grant it), be aware this rule passes whether or not the condition actually permits the five roles the Gateway needs — Storage Blob Data Contributor, Storage Account Contributor, Reader, Key Vault Reader and Key Vault Secrets User — since Azure does not report the condition text to this check
- If this rule passes but Task Runners provision and start cleanly, then cannot read the storage account or the key vault, check the condition text on the Gateway's Role Based Access Control Administrator assignment rather than treating it as a storage or vault problem
Rule: Ensure the Gateway VM's identity can read and attach Task Runners to the Gateway's Vnet
Task Runner network interfaces are attached to the subnet of the virtual network the Gateway sits in, which needs the join action listed below. The network and its subnets must also be readable, because the Gateway reads the subnet's configuration and its remaining addresses before provisioning a Task Runner into it — read access is asked for separately so that a network which can be seen but not joined, and one which cannot be seen at all, are told apart rather than both being reported as an unevaluable scope. The virtual network's own resource group is discovered rather than assumed, so this is correct even when the network lives in a different resource group from the Gateway. Network Contributor grants all of these actions, as do Owner and Contributor on the network; Reader grants the two read actions alone.
The required actions are:
- Microsoft.Network/virtualNetworks/read
- Microsoft.Network/virtualNetworks/subnets/read
- Microsoft.Network/virtualNetworks/subnets/join/action
What to do if this rule is failing:
- Verify the Gateway's system assigned identity was successfully created (see related rule above)
- Manually assign the Gateway system assigned identity the Network Contributor role, or a custom role granting the above permissions, on the virtual network (Vnet) connected to the Gateway using Azure Portal procedures.
Rule: Check that the Gateway's subnet has enough unused IP Addresses
Task Runners are provisioned and deprovisioned on demand and by default will consume 13 IP Addresses during times of high activity when all Task Runners are provisioned concurrently. If enough used IP addresses are not available this can result in Task Runner provisioning steps failing. If sufficient IP addresses are not available on the subnet the Gateway should be moved, or redeployed, to a different subnet with enough unused IP Addresses.
This rule is designed to validate that the minimum number of IP addresses required for the default Task Runner configuration are available when Change360 resources are initially deployed. If additional Task Runners are added after deployment, including Cloud PC Task Runners on the same subnet or Persistent Task Runners, these will consume additional IP addresses that are not considered by this rule when it is run on an ad hoc basis after deployment. As a result, Task Runner provisioning may fail due to insufficient available IP addresses, even if this rule reports a passing status.
What to do if this rule is failing:
- Navigate to the Vnet used by the Gateway in Azure Portal and select Subnets under Settings
- Examine the Available IPs for the Gateway's subnet; if fewer than 13 addresses exist, consider moving the Gateway to a subnet with minimum 13 available IPs or to a dedicated subnet
Rule: Ensure the Gateway is accessible on port 5000
Task Runners communicate with the Gateway on port 5000 therefore access to the Gateway on this port should not be blocked by any local firewalls, endpoint protection products, network appliances or other security resources. If this rule is failing deployment can continue but the issue must be resolved before Task Runners can be successfully provisioned.
What to do if this rule is failing:
- Determine if a local firewall is enabled on the Gateway VM and, if so, verify firewall rules allow port 5000 inbound
- Check whether the Gateway's network card associates with a Network Security Group and ensure port 5000 inbound is not being explicitly or inherently denied by any rules.
Rule: Ensure the Gateway is accessible on port 5001
Centralized logging information is sent from Task Runners to the Gateway on port 5001 therefore access to the Gateway on this port should not be blocked by any local firewalls, endpoint protection products, network appliances or other security resources. If this rule is failing deployment can continue but the issue should be resolved before attempting to provision Task Runners to ensure logs are available for troubleshooting if needed.
What to do if this rule is failing:
- Verify whether a local firewall exists on the Gateway VM; if present, confirm firewall rules permit port 5001 inbound
- Determine if the Gateway's network card connects to a Network Security Group and verify port 5001 inbound is not being explicitly or inherently denied by any rules.
Rule: Check if any local firewall profiles are enabled
If any local firewall profiles are enabled then appropriate rules will need to be in place to allow communication between the Gateway and Task Runners. If any firewall profiles are enabled then additional checks will be performed to ensure that rules exist to allow communication on the required ports. If no firewall profiles are enabled this rule will return a warning and the additional port checks will be skipped.
What to do if this rule is failing:
- If a local firewall is detected, no immediate action is required, but review the following related rules to confirm they pass and that the firewall doesn't block inbound traffic on ports 5000 or 5001:
Rule: Ensure an Inbound firewall rule for port 5000 exists (if firewall is enabled)
Task Runners communicate with the Gateway on port 5000 therefore if a firewall is enabled a rule should be present to allow inbound traffic on port 5000.
What to do if this rule is failing:
- Add a local firewall rule permitting inbound TCP traffic to the Gateway virtual machine on port 5000
- Optionally restrict the source to the subnet where the Gateway resides (where Task Runners also deploy); note that limiting access may prevent Domain Join Wizard access from other subnets
Rule: Ensure an Inbound firewall rule for port 5001 exists (if firewall is enabled)
Centralized logging information is passed from the Task Runner to the Gateway over port 5001 therefore if a firewall is enabled a rule should be present to allow inbound traffic on port 5001.
What to do if this rule is failing:
- Add a local firewall rule enabling inbound TCP traffic to the Gateway virtual machine on port 5001
- Optionally limit the source to the subnet containing the Gateway, where Task Runners also deploy; no other WorkspaceDNA services communicate with the Gateway on port 5001
Rule: Ensure the Gateway control channel can be upgraded to a WebSocket
The Gateway holds a persistent WebSocket connection to the Web Host for its control channel. Appliances that terminate or rewrite HTTP frequently fail the HTTP Upgrade handshake while leaving ordinary requests working, so this check can fail even when the HTTPS check on the same host and port passes — and when it does, the symptom in production is a Task Runner that appears to go offline mid-sequence for no reason. TLS inspection must be bypassed for this endpoint, and any firewall, load balancer or proxy idle timeout should be at least 4 minutes because the hub sends a keep-alive every 60 seconds.
What to do if this rule is failing:
- If "Check connectivity to the WorkspaceDNA Web Host" above passes but this rule fails, look for a proxy, firewall, or load balancer that terminates or rewrites HTTP traffic to rimo3cloud.com, since that can break the WebSocket Upgrade handshake while leaving plain HTTPS requests working
- Bypass TLS inspection for rimo3cloud.com on port 443
- Ensure any firewall, load balancer, or proxy idle timeout on this path is at least 4 minutes — a shorter timeout will drop the connection between keep-alives, which are sent every 60 seconds
- If Task Runners appear to go offline mid-sequence for no obvious reason, check this rule first
Rule: Ensure the Gateway agent control channel can be upgraded to a WebSocket
Task Runners hold a persistent WebSocket to the Gateway on port 5000 for their control channel. An appliance that terminates or rewrites HTTP inside the virtual network can fail the HTTP Upgrade while leaving ordinary requests on the same port working, so this check can fail even when the plain connectivity check on port 5000 passes — and when it does, the symptom is a Task Runner that appears to go offline mid-sequence for no reason. The probe deliberately ignores any machine proxy, because the Gateway is reached directly on its private address. Any idle timeout on this path should be at least 4 minutes, since the hub sends a keep-alive every 60 seconds.
What to do if this rule is failing:
- If "Ensure the Gateway is accessible on port 5000" above passes but this rule fails, look for a local firewall, endpoint protection product, or network appliance inside the virtual network that terminates or rewrites HTTP traffic, since that can break the WebSocket Upgrade handshake while leaving plain requests on the same port working
- Ensure any idle timeout on this path (local firewall, NSG, or network appliance) is at least 4 minutes, since keep-alives are sent every 60 seconds
- If Task Runners appear to go offline mid-sequence for no obvious reason, check this rule as well as the equivalent Gateway control channel rule above
Rule: Check that the auto login user is an administrator
The WorkspaceDNA service leverages the auto login capability of Windows to automate various activities, including the automated install and uninstall of applications which requires administrative rights. The auto login user therefore needs to be granted local admin rights on devices created by the service to ensure it can carry out the necessary automated activities.
What to do if this rule is failing:
- The provisioning process automatically adds the auto-login user to the local admins group; however, domain-join policies may remove unrecognized users from this group, in which case those policies must be blocked for the WorkspaceDNA organizational unit.
- Check policies that add the auto login user into the local admin group as these policies don't always result in the desired behaviour. Instead the policy should be blocked and the Task Runner provisioning script should be allowed to add the auto login user to the local admin group.
Rule: Check connectivity to the signing timestamp service (only required for MSIX signing)
Packages signed as part of MSIX package creation are timestamped against this service over Port 80. This rule is only relevant when Modernization to MSIX is in scope and can be disregarded otherwise.
What to do if this rule is failing:
- If Modernization to MSIX is not in scope, this rule can be disregarded
- If Modernization to MSIX is in scope, verify perimeter security devices aren't blocking traffic to timestamp.digicert.com on port 80
Rule: Ensure that a legal notice is not configured
The WorkspaceDNA service leverages the auto login capability of Windows to automate various activities, a legal notice will prevent auto login from completing successfully and therefore interrupt various automated activities. Any legal notice for interactive or RDP Login should be disabled to prevent service interruption.
What to do if this rule is failing:
- Check whether the following registry keys are configured on the Gateway:
- HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System\LegalNoticeCaption
- HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System\LegalNoticeText
- Check whether these registry keys are configured on the Gateway:
- HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Terminal Server\WinStations\RDP-Tcp\LogonMessageTitle
- HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Terminal Server\WinStations\RDP-Tcp\LogonMessage
- These keys may be established by Group Policy or custom images; if values reappear after deletion, identify what enforces the configuration and block it for Change360 resources
Rule: Ensure that a screen saver is not enabled
A screen saver could potentially interrupt the video capture of smoke tests preventing useful information from being captured in screenshots and the video recording. It is recommended that any screensavers be disabled on the devices used by the WorkspaceDNA service.
This check is being run against the Gateway on the assumption that a similar policy may also be applied to Task Runners. However if policies are filtered based on OS type or version it is possible that this check may produce both false negatives and false positives
What to do if this rule is failing:
- Connect to the Gateway and verify whether a screensaver is configured; disable it if active
- If the screensaver cannot be disabled, it may be enforced through one or more Group Policy Objects (GPOs). In this case, any applicable GPOs should be blocked from applying to the WorkspaceDNA organisational unit (OU) that contains the Gateway and Task Runner computer accounts. If separate GPOs are applied to Task Runners, these must also be identified and blocked as required.
Rule: Ensure that the automatic lock screen is not enabled
The lock screen could potentially interrupt the video capture of smoke tests preventing useful information from being captured in screenshots and the video recording. It is recommended that the automatic lock screen be disabled on the devices used by the WorkspaceDNA service.
This check is being run against the Gateway on the assumption that a similar policy may also be applied to Task Runners. However if policies are filtered based on OS type or version it is possible that this check may produce both false negatives and false positives
What to do if this rule is failing:
- Access the Gateway and verify the screensaver configuration and whether "On resume, display log-on screen" is selected; deselect if necessary
- Check "If you've been away, when should Windows require you to sign in again?" under Settings → Accounts → Sign-in options: Additional settings and change to Never if needed
- If above settings are configured but cannot be edited, it may be enforced through one or more Group Policy Objects (GPOs). In this case, any applicable GPOs should be blocked from applying to the WorkspaceDNA organisational unit (OU) that contains the Gateway and Task Runner computer accounts. If separate GPOs are applied to Task Runners, these must also be identified and blocked as required.
- Check whether these registry keys exist and delete them if present, it may be necessary to also remove these key from custom images if present:
- HKCU\Control Panel\Desktop\ScreenSaverIsSecure=1
- HKLM\Software\Microsoft\Windows\CurrentVersion\Policies\System\
InactivityTimeoutSecs
- If deleted registry keys reappear, determine what enforces them and block that configuration for Change360 resources, including Task Runners.
Rule: Ensure a storage account was created
A storage account is required to save uploaded package source files, Modernized applications created within the WorkspaceDNA platform, store log files and cache downloadable package files. If a storage account could not be created by the MPO you may need to pre-create a storage account and rerun the MPO selecting the existing storage account.
What to do if this rule is failing:
- Verify whether Azure policies prevent storage account creation; temporarily disable policies for the Change360 resource group if necessary to allow deployment completion, then reconfigure the storage account to meet policy requirements
- Reference detailed procedures on restarting the Rimo3 deployment process
Rule: Check connectivity to the WorkspaceDNA storage account over HTTPS
The storage account holds uploaded package sources, Modernized applications, log files and cached downloads, and is accessed over HTTPS on Port 443. Where Azure Firewall or Network Security Group rules are used, the Storage service tag is preferable to allow-listing the name. Note that inspecting proxies frequently impose request body size or timeout limits that large package transfers exceed, so bypassing TLS inspection for this endpoint is recommended.
What to do if this rule is failing:
- Verify perimeter security devices aren't blocking traffic to <StorageAccountName>.blob.core.windows.net on port 443
- Where Azure Firewall or Network Security Group rules are used, allow the Storage service tag rather than the individual hostname
- Bypass TLS/SSL inspection for this endpoint, since large package transfers commonly exceed the request body size or timeout limits an inspecting proxy imposes
Rule: Check the application package file share is reachable over SMB 3
Application packages are read from the file share on the storage account using SMB 3 over Port 445. This traffic is not HTTP and cannot traverse a web proxy under any configuration, so it needs a direct outbound path. Outbound Port 445 is commonly blocked both by internet service providers and by corporate policy, and this rule failing with the port unreachable is the most likely cause of packages not being found and the service continuing in degraded mode. If the rule reports that a dialect below SMB 3 was negotiated then the port is open and SMB works, but the SMB 3 features the file share requires are unavailable. Shared key access on the storage account and outgoing NTLM must also be permitted.
What to do if this rule is failing:
- Confirm outbound port 445 is not blocked — this port cannot be routed through a proxy under any configuration, and is commonly blocked by internet service providers or corporate policy; this is the most likely cause if packages are not being found and the service is running in degraded mode
- If the rule reports that a dialect below SMB 3 was negotiated, the port is open but the SMB 3 features required by the file share are unavailable — check that shared key access is enabled on the storage account and that outgoing NTLM is permitted
- If this rule fails, expect "Ensure the application package share is mapped as the T drive" below to fail too, since the share cannot be mapped if it cannot be reached
Rule: Ensure the storage account can be read from and written to using the machine's identity
This is the only check that actually attempts to read and write data on the storage account, rather than checking that access is granted. It performs the same operations the platform itself performs: it lists the containers on the account, creates a container of its own named wdna-reqcheck-<timestamp>, writes a small blob into it, reads that blob back, and then removes both.
The Storage Blob Data Contributor check confirms the role is assigned; this rule confirms that assignment actually works, which is not guaranteed - a role assigned moments ago can take several minutes to propagate, and the account's firewall or virtual network rules can block the request even when the role is correct.
A failure reporting AuthorizationPermissionMismatch means that role has not taken effect yet. A failure reporting AuthorizationFailure means the account's network rules are rejecting this machine instead, which is a different fix entirely. If the rule passes but reports that its container could not be removed then read and write both work but the identity cannot delete, and the named container should be removed by hand.
What to do if this rule is failing:
- A failure reporting AuthorizationPermissionMismatch means the Storage Blob Data Contributor role is missing from the machine's system assigned identity, or was assigned too recently to have propagated — assign the role (see "Ensure the Gateway VM's identity can read and write blob data on the storage account" below) and allow time for it to take effect
- A failure reporting AuthorizationFailure means the storage account's firewall or virtual network rules are rejecting this machine — check "Ensure the storage account firewall permits the Rimo3 platform source addresses" below, and confirm the machine's own address is permitted if it connects from outside the expected range
- If the rule passes but reports that its test container (named rimo3-reqcheck-<timestamp>) could not be removed, read and write access both work but the identity cannot delete — remove the named container manually
- Remember that holding Owner or Contributor at the management level does not grant blob data access, since those roles carry no data actions
Rule: Ensure the Gateway VM's identity can read and write blob data on the storage account
Checks that the Gateway VM's identity is assigned a role granting the blob data actions above - read, write and delete - on the storage account. This is a check of what is assigned, not of whether it actually works: it looks at the role assignment in Azure, not at whether a blob call succeeds. Azure keeps data-plane permissions separate from management-plane permissions, so a role granting every management action - Owner included - grants no data access at all; Storage Blob Data Contributor (or Storage Blob Data Owner) is required specifically. Because this only reflects what has been assigned, a role added moments ago can pass here before it has fully propagated - the read/write probe against the storage account is what actually proves read and write work end to end.
The required data actions are:
- Microsoft.Storage/storageAccounts/blobServices/containers/blobs/read
- Microsoft.Storage/storageAccounts/blobServices/containers/blobs/write
- Microsoft.Storage/storageAccounts/blobServices/containers/blobs/delete
What to do if this rule is failing:
- Assign the Storage Blob Data Contributor role to the Gateway VM's system assigned identity, scoped to the storage account.
- Remember that Owner or Contributor at the management level does not grant these actions — Azure keeps management and data permissions separate, so the data role must be assigned explicitly
- Allow time for a newly assigned role to propagate before re-running the check
Rule: Ensure the Gateway VM's identity can read the storage account's configuration and keys
Reading the storage account's configuration and its access keys is a management plane operation on, separate from the blob data access checked alongside it. The keys are needed because the application package file share is reached with a shared key. Storage Account Contributor grants these actions, and so do Owner and Contributor at the resource group or above, since they cover every management action.
The required actions are:
- Microsoft.Storage/storageAccounts/read
- Microsoft.Storage/storageAccounts/listKeys/action
What to do if this rule is failing:
- Assign the Storage Account Contributor role, or a custom role with the above permissions, to the Gateway VM's system assigned identity, scoped to the storage account — Owner or Contributor at the resource group or above also grants these actions
- This permission is required because the application package file share is reached using a shared key, which must be read from the storage account's configuration
Rule: Ensure the application package share is mapped as the T drive
Application packages are read from the file share through the T drive. This is one level above the SMB 3 reachability check on the same share: the share can be perfectly reachable while the mapping is absent or disconnected, and both look the same to the platform — packages simply cannot be read. A disconnected mapping is reported as such rather than as a missing one.
What to do if this rule is failing:
- If the SMB 3 reachability check above passes but this rule fails, the share is reachable but is not mapped as the T drive, or the existing mapping has been disconnected — check which of the two the rule reports.
- The T drive is mapped automatically by the Gateway Agent on startup, if the T drive is not present look in the Gateway logs, C:\Logs\gateway-log-{Date}.txt, for "map drive" stderr for details of any errors associated with the drive failing to map.
- If the T drive is disconnected, first check the Gateway log files for any errors related to mapping the T drive.
- Restart the Gateway and immediately run the Health Check again to confirm that the rule reports a successful result.
- You can then run the Health Check at regular intervals until the rule fails again.
- This will help determine how long it takes for the drive to become disconnected, which can provide valuable information when troubleshooting the root cause of the disconnection.
Rule: Ensure a key vault was created
A key vault is required to securely store properties and credentials used by the WorkspaceDNA service. If a key vault could not be created by the MPO then a key vault can be manually created in the WorkspaceDNA resource group, ensuring the Gateway system assigned identity has full rights to secrets. Restart the Gateway to discover the key vault.
What to do if this rule is failing:
- Manual key vault creation in the Change360 resource group is required if the deployment fails to create one
- Ensure the Gateway system assigned identity possesses full permissions for secrets management
Rule: Check connectivity to the WorkspaceDNA Key Vault
Properties and credentials used by the WorkspaceDNA service are stored in the Key Vault. Task Runners read it with their own identity, so it must be reachable from the Task Runner subnet and not only from the Gateway — a Key Vault firewall or private endpoint scoped to the Gateway alone will cause provisioning to spend its full retry window waiting before completing without an agent. Where Azure Firewall or Network Security Group rules are used, the AzureKeyVault service tag is preferable to allow-listing the name.
What to do if this rule is failing:
- Verify perimeter security devices aren't blocking traffic to <KeyVaultName>.vault.azure.net on port 443
- Where Azure Firewall or Network Security Group rules are used, allow the AzureKeyVault service tag rather than the individual hostname
- Confirm the Key Vault is reachable from the Task Runner subnet, not only from the Gateway — a Key Vault firewall or private endpoint scoped to the Gateway alone will let this rule pass while Task Runner provisioning still times out waiting for the agent to come online
Rule: Ensure the Gateway VM's identity can manage access policies on the key vault
The key vault is deployed by the MPO in access policy mode rather than with RBAC authorization enabled, so Azure ignores key vault data plane role assignments on it entirely. The Gateway grants each Task Runner access to the vault by writing an access policy for it, and writing an access policy is a management plane operation on the vault itself — which is why the actions listed below are management actions and no data actions are asked for here. This is the distinction to keep in mind when the rule fails: Key Vault Contributor grants these actions, and Key Vault Administrator does not, because it is a data plane role and confers nothing on a vault that is not in RBAC mode. Owner and Contributor on the vault or above also grant them.
The required actions are:
- Microsoft.KeyVault/vaults/read
- Microsoft.KeyVault/vaults/write
- Microsoft.KeyVault/vaults/accessPolicies/write
This rule depends on "Ensure that system assigned identity is enabled" above having passed.
Blocker: This check is critical for the Change360 service. Deployment cannot proceed if it fails. After resolving issues, run "Run Checks" in the Deployment Wizard to validate changes.
What to do if this rule is failing:
- Confirm the Gateway's system assigned identity was successfully created (see "Ensure that system assigned identity is enabled" above)
- Manually assign the Gateway system assigned identity the Key Vault Contributor role (or Owner/Contributor at the vault or above), scoped to the key vault (##KeyVaultName##)
- Do not assign Key Vault Administrator in its place — it is a data plane role and grants nothing on this vault, because the key vault is deployed in access policy mode rather than with RBAC authorization enabled, so key vault data plane role assignments are ignored entirely
- If this rule fails, expect Task Runners to be unable to be granted access to the vault, since writing an access policy for each Task Runner is a management plane operation that depends on this permission