Chat now with support
Chat with Support

On Demand Migration Current - Active Directory User Guide

Offline Domain Join (ODJ)

Normally, right after a Device is Cutover to a new domain VPN users can’t log in to their workstation because Windows must be able to contact the target domain to authenticate against a domain controller for that very first login. Typically, a remote user not on the VPN would need to log in to their machine first and then establish a VPN connection.

Active Directory answers this problem by building on Microsoft’s Offline Domain Join (ODJ) process to allow a workstation to join a domain without contacting a Domain Controller. This solution is achieved by first creating an ODJ file for each workstation and then taking advantage of Windows’ ability to cache credentials. If users have logged in to the target domain previously, Windows can log them in again even if they can no longer reach a domain controller by using cached credentials.

Active Directory’s ODJ process has users pre-login to the new domain before the computer needs to be Cutover so the target credentials can be cached and used for the first post-Cutover login without the need to contact a domain controller first. Then when the administrator is ready to Cutover workstations using the ODJ Action, Active Directory allows the workstation to join the new domain without having the user connect to the corporate VPN and manually join their workstation to the new domain.

The computers that the ODJ process is being run on must have network connectivity to BOTH the source and target environments at the same time sometime pre-cutover in order to have the Cache Credentials function work properly.

Additionally, computers will be unable to save the cached credentials unless the source environment trusts the target environment. The AD Offline Domain Join Credential Cache Quick Start Guide provides guidance on configuring an AD trust to enable Offline Domain Join functionality.

1. CREATING ODJ FILES FOR EACH WORKSTATION  

The first step in the ODJ process is for an administrator to use Microsoft’s DJOIN utility to create a provisioning file. Only the provisioning part of the DJOIN process is needed. Complete information on DJOIN can be found here.

The Provision, Domain, Machine, and Savefile parameters are required at a minimum. There is the option to control where the target machine will be created using the MachineOU parameter as in the sample shown here.

DJOIN.EXE /Provision /Domain BTADLAB.com /Machine Sales220 /Savefile “\\server\odj-share\Sales220.txt” /MACHINEOU OU=SalesComputers,OU=Sales,DC=BTADLAB,DC=COM

The file must be saved in the ODJ folder in the Repositories path that was configured in the UI.

Generating these files can be completed for all in scope workstations early in the migration process.

WARNING: Be sure to name each text file with the exact matching machine name.

 

2. CONFIGURING THE CREDENTIAL CACHE PROFILE  

The next step is to configure the existing Default Credential Cache profile with the IP address of a Target domain controller, or to create a new profile for this setting.

From the Profiles page select Credential Cache Profiles. Click on Add to create a new profile, or Edit to modify an existing profile. If you choose to use the Default profile, you must edit it to include a Target DC IP address.

  • The Target Domain Controller Ping Interval setting determines how long the script will sleep before pinging the DC again.
  • The Timeout Before Job Failure setting determines the Credential Cache app timeout value that will be used for the job once downloaded to the agent managed machine.
  • The Timeout For User Credential setting determines how long the user is presented with a dialog box to enter their target domain credentials.

 

3. CACHE CREDENTIALS JOBS  

Now that a profile has been configured with a target DC IP address, we can assign a Cache Credentials job to each in scope workstation.

In the Devices list, select one or more Computers. Select Cache Credentials from the Actions dropdown menu and click the Apply Action button. The Credential Cache Options box appears.

Select a Credential Cache Profile.

A date and time for the Cache Credentials job can be chosen to run the job at a later time. This date/time combination represents the earliest time that this job could run.

If a date/time is not chosen, this job will run on the workstation the next time the agent checks for jobs.

The Devices list will reflect a status of "Queued". When the job is collected by the agent the status will change to "In Progress", and then finally it will transition to a status of "Completed" or "Failed".

On the workstation side, when the Cache Credentials Job is received, the user will be prompted to enter their target credentials. Below is an example of what the user will see when the Cache Credential job runs:

Figure 1: Enter Cutover Credentials

 

4. REACL  

Following Cache Credentials the next recommended step is the ReACL process. The ReACL process can be run repeatedly as needed before ODJ, but it is suggested to be run at least once right after the Cache Credentials process is run.

In the Devices list, select one or more Computers. Select ReACL from the Actions dropdown menu and click the Apply Action button.

The Job Options box appears. A specific date/time combination can be chosen for when to run the job, or just click Apply Action to have this job received by the workstations during their next check for jobs.

 

5. OFFLINE DOMAIN JOIN JOB  

The final step is the actual Offline Domain Join job itself. This is similar to the Cutover process used for machines that are directly connected to the network.

WARNING: Do not perform the Cutover process on Offline Domain Join workstations. The Offline Domain Join process takes the place of Cutover for workstations connecting via VPN.

In the Devices list, select one or more Computers. Select Offline Domain Join from the Actions dropdown menu and click the Apply Action button.

The Job Options box appears. A specific date/time combination can be chosen for when to run the job, or just click Apply Action to have this job received by the workstations during their next check for jobs.

WARNING: The Offline Domain Join (Job Scheduling Options dialog box) start date and time must be set AFTER the Cache Credentials job (Cache Credential Options dialog box) start date and time.

The Offline Domain Join process does not support rollback.

Custom Action Example

Scenario: The administrator wants to download the Get-DiskSpace.ps1 file on selected workstations.

Steps:

  1. Copy the Get-DiskSpace.ps1 file into the file location of the configured Custom Downloads Repository.
  2. On the Actions screen, click the New button under the Actions table.
  3. Add a Custom Action named "Get-Diskspace". The Action Display Name is how the option will appear in the Actions menu. Select the "Computer" Action Target so the menu option will appear in the Actions menu on the Computer Actions screen.

    Add a Custom Action screen

  4. Click the New button under the Tasks table.
  5. Select the "Download File" Task Type and enter the File Name. The full path to the file should not be entered.

    Add a Custom Task screen

    File Download screen

  6. Select the task, select "Get-Diskspace" from the Select Action drop-down menu and click the Add To button.

    Ensure the Action is active. The action can be expanded to view the associated Task.

    Expanded Action displaying Task

  7. On the Devices + Servers screen, select the devices and then select the new "Get Disk Space" option from the Actions menu. When the new download job is run successfully against a machine, the Get-DiskSpace.ps1 file will reside locally on disk at c:\Program Files (x86)\Binary Tree\P365ActiveDirectoryAgent\Downloads\.

FAQs

General FAQs

What ports does the agent use to connect?  

The Active Directory Agent uses three outbound ports

  • TCP 443/80
  • UDP 3030

 

I’ve installed the agent and the device isn’t ready, what do I do?  

On startup the Active Directory agent will phone home sometime in the first four hours. This communication offset time is an agent specific random and evenly distributed offset so the initial communications of a large number of devices set up at once will be spaced out and not overload the servers. If your device has the agent installed but you haven’t waited up to four hours yet, wait up to four hours and then re-check the Ready Devices list to see if your device shows up before proceeding to other network troubleshooting operations.

 

How do I adjust the agent polling interval?  

In Active Directory, unlike in Active Directory Pro, the agent polling interval is not end user adjustable. The polling interval is every two minutes over UDP and every four hours over TCP. Should this amount of network traffic be an issue contact Support to investigate reducing the polling interval. Note that reducing the polling interval will cause queued migration actions to take longer to be picked up by the agents.

 

How many migration actions can I queue at once?  

In Active Directory you can queue as many migration actions at once as you would like. There is a hard limit of 600 agents per each two minutes who will be notified by the job availability cache that they have a job waiting for them. Therefore the minimum amount of time which could be required for those queued migration actions to be picked up will be the number of queued migration actions divided by 600 and multiplied by two minutes.

 

Agent last contact time isn’t updating every two minutes is something wrong?  

In Active Directory the Agent last contact time is updated based on the TCP connection with the back end. This will occur if a job is retrieved or every 4 hours as a fall back. It is not expected that the agent last contact time will update every two minutes even if the UDP requests to the job availability cache are functioning correctly.

 

The machine name was changed during the migration project, will it keep working?  

Changing the machine name during a migration project may have unexpected implications. We recommend that you avoid doing so if possible. If not, you will need to delete the device from Power365 and reinstall the agent. Use the Status Resets action to reset the registration status of the device, then set the Directory Sync Workflow to reconcile on next run and run it. This will delete the old device from the database. Finally, reinstall the Active Directory Agent on that device and run the Workflow again to read it in.

 

How do I remove Devices from Active Directory?  

To delete registered Devices from Active Directory, first use the Status Resets action to clear their registration status, then run the related Directory Sync Workflow with the setting ‘Reconcile on next run’ enabled. Any unregistered device records in scope of the Workflow will be removed from the database during a reconcile.

Related Documents

The document was helpful.

Select Rating

I easily found the information I needed.

Select Rating