Scripting the generation & creation of Microsoft Identity Manager Sets/Workflows/Sync & Management Policy Rules with the Lithnet Resource Management PowerShell Module

Introduction

Yes, that title is quite a mouthful. And this post is going to be quite long. But worth the read if you are having to create a number of rules in Microsoft/Forefront Identity Manager, or even more so the same rule in multiple environments (eg. Dev, Staging, Production).

My colleague David Minnelli introduced using the Lithnet RMA PowerShell Module and the Import-RMConfig cmdlet recently for bulk creation of MIM Sets and MPR’s. David has a lot of the background on Import-RMConfig and getting started with it. Give that a read for a more detailed background.

In this post I detail using Import-RMConfig to create a Set, Workflow, Synchronization Rule and Management Policy Rule to populate a Development AD Domain with Users from a Production AD Domain. This process is designed to run on a combined MIM Service/Sync Server. If your roles a separated (as they likely will be in a Production environment) you will need to run these scripts on the MIM Sync Server (so it can query the Management Agents, and you will need to add in a line to connect to the MIM Service (eg. Set-ResourceManagementClient ) at the beginning of the script.

In my environment I have two Active Directory Management Agents, each connected to an AD Forest as shown below.

On each of the AD MA’s I have a Constant Flow Attribute (named Source) configured to flow in a value representing the source AD Forest. I’m doing this in my environment as I have more than one production forest (hence the need for automation). You could simply use the Domain attribute for the Set criteria. That attribute is used in the Set later on. Mentioning it up front so it make sense.

Overview

The Import-RMConfig cmdlet uses XML and XOML files that contain the configuration required to create the Set, Workflow, Sync Rule and MPR in the FIM/MIM Service. The order that I approach the creation is, Sync Rule, Workflow, Set and finally the MPR.

Each of these objects as indicated above leverage an XML and/or XOML input file. I’ve simplified base templates and included them in the scripts.

The Sync Rule Script includes a prompt to choose a folder (you can create one through the GUI presented) to store the XML/XOML files to allow the Import-RMConfig to use them. Once generated you can simply reference the files with Import-RMConfig to replicate the creation in another environment.

Creating the Synchronization Rule

For creation of the Sync Rule we need to define which Management Agent will be the target for our Sync Rule. In my script I’ve automated that too (as I have a number to do). I’m querying the MIM Sync Server for all its Active Directory MA’s and then providing a dialog to allow you to choose the target MA for the Sync Rule. That dialog simply looks like the one below.

Creating the Sync Rule will finally ask you to give the Rule a name. This name will then be used as the base Display Name for the Set, MPR and Workflow (and a truncated version as the Rule ID’s).

The script below in the $SyncRuleXML section defines the rules of the Sync Rule. Mine is an Outbound Sync Rule, with a base set of attributes and transforming the users UPN and DN (for the differing Development AD namespace). Update lines 42 and 45 for the users UPN and DN your namespace.

Creating the Workflow

The Workflow script is pretty self-explanatory. A simple Action based workflow and is below.

Creating the Set

The Set is the group of objects that will be synchronized to the target management agent. As my Sync Rule is only for Users my Set is also only contains users. As stated in the Overview I have an attribute that defines the authoritative source for the objects. I’m also using that in my Set criteria.

Creating the Management Policy Rule

The MPR ties everything together. Here’s that part of the script.

Tying them all together

Here is the end to end automation, and the raw script that you could use as the basis for automating similar rule generation. The Sync Rule could easily be updated for Contacts or Groups. Remember the attributes and object classes are case sensitive’.

  • Through the Browse for Folder dialog I created a new folder named ProvisionDevAD

  • I provided a Display Name for the rules

  • I chose the target Management Agent

  • The SyncRule, Workflow, Set and MPR are created. The whole thing takes seconds.

  • Script Complete

Let’s take a look at the completed objects through the MIM Portal.

Sync Rule

The Sync Rule is present as we named it. Including the !__ prefix so it appears at the top of the list.

Outbound Sync Rule based on a MPR, Set and Workflow

The Resources will be created and if deleted de-provisioned.

And our base attribute flows.

Set

Our Set was created.

Our naming aligns with what we input

And a Criteria based Set. As per the Overview I have an attribute populated by a Constant flow rule that I based my set on. You’ll want to update for you environment.

Workflow

The Action Workflow was created

All looks great

And it applies our Sync Rule

MPR

And finally our MPR. Created as a Transition In MPR with Action Workflow

Set Transition and naming all aligned

The Transition Set configured for the Set that was created

And the Workflow configured with the Workflow that was just created

Summary

When you have a lot of Sync Rules to create, or you know you will need to re-create them numerous times, potentially in different environments automation is key. This just scratches the surface on what can be achieved, and made so much easier using the Lithnet PowerShell Modules.

Here’s is the full script. Note: You’ll need to make a couple of minor changes as indicated earlier, but you should be able to create a Provisioning Rule end to end pretty quick to validate the process. Then customize accordingly for your environment and requirements. Enjoy.

Automating Source IP Address updates on an Azure Network Security Group RDP Access Rule

Recently I’ve migrated a bunch of Virtual Box Virtual Machines to Azure as detailed here. These VM’s are in Resource Groups with a Network Security Group associated that restricts access to them for RDP based on a source TCPIP address. All good practice. However from a usability perspective, when I want to use these VM’s, I’m not always in the same location, and rarely on a connection with a static IP address.

This post details a simple little script that;

  • Has a couple of variables associated with a Resource Group, Network Security Group, Virtual Machine Name and an RDP Configuration File associated with the VM
  • Gets the public IP Address of the machine I’m running the script from
  • Prompts for Authentication to Azure, and retrieves the NSG associated with the Resource Group
  • Compares the Source IP Address in the ‘RDP’ Inbound Rule to my current IP Address. If they aren’t a match it updates the Source IP Address to be my current public IP Address
  • Starts the Virtual Machine configured at the start of the script
  • Launches Remote Desktop using the RDP Configuration file

The Script

Here’s the raw script. Update lines 2-8 for your environment and away you go. Simple but useful as is often the way.

Diagnosing FIM/MIM ‘kerberos-no-logon-server’ error on an Active Directory Management Agent

Overview

I have a complex customer environment where Microsoft Identity Manager is managing identities across three Active Directory Forests. The Forests all serve different purposes and are contained in different network zones. Accordingly there are firewalls between the zone where the MIM Sync Server is located and two of the other AD Forests as shown in the graphic below.

As part of the project the maintainers of the network infrastructure had implemented rules to allow the MIM Sync server to connect to the other two AD Forests. I had successfully been able to create the Active Directory Management Agents for each of the Forests and perform synchronization imports.

The Error ‘kerberos-no-logon-server’

Everything was going well right up to the point I went to export changes to the two AD Forests that were separated by firewalls. I received the ‘kerberos-no-logon-server’ error as shown below from the run profile output.

I started investigating the error as I hadn’t encountered this one before. There were a few posts on the possibilities mainly dealing with properties of the AD MA’s configuration. But I did stumble on a mention of kerberos being used when provisioning users to Active Directory and setting the initial password. That aligned with what I was doing. I had provided the networking engineers with my firewall port requirements. Those are (no PCNS required for this implementation) ;

  • 389 TCP – LDAP
  • 636 TCP – LDAPS
  • 88 TCP – Kerberos
  • 464 TCP/UDP – Kerberos
  • 53 TCP – DNS
  • 3268 TCP/UDP – Global Catalog
  • 3269 TCP/UDP – Global Catalog
  • 135 TCP – RPC

My old school immediate thought was to Telnet to each of the ports to see if the firewall was allowing me through. But with a couple of forests to test against and UDP ports as well, it wasn’t going to be that easy. I found a nice little Test-Port function that did both TCP and UDP. I already had an older script for testing TCP ports via PowerShell. So I combined them.

Identifying the cause

As suspected connectivity to the forest where my MIM Sync Server was located was all good. The other two, not so much. GC connectivity wouldn’t give me the Kerberos error, but not having Kerberos Port 464 certainly would.

In the backwards and forwards with the networking team I had to test connectivity many times so I added a running output as well as a summary output. The running output highlighting ports that weren’t accessible.

Here’s the raw script if you’re in a similar situation. Get the Test-Port Function from the URL in line 1 to test UDP Port connectivity. Add additional ports to the arrays if required (eg. for PCNS), and update the forest names in lines 21-23.

Summary

I’m sure this is going to become more relevant in a Cloud/Hybrid world where MIM Servers will be in Azure, Active Directory Forests will be in different networks and separated by firewalls and Network Security Groups.

An alternate method for dealing with Orphaned MetaVerse Objects

Update 21 April ’17. The LithnetMIISAutomation PS Module now has a -Force switch for Delete-CSObject

As often happens in development environments, data changes, configurations change and at some point you end up with a whole bunch of objects that are in no-mans land. This happened to me today. I had thousands of objects that we basically empty but had previously triggered to be exported to the MIM Service prior to them actually being deleted from the source management agent.

An example of one of the objects. A group with a Pending Export to the MIM Service.

A closer look at the object and there is no attribute data present as the source object had been removed.

And only a single connector, to the MIM Service which it will never reach as it doesn’t contain the mandatory attributes.

Normally to clean up such a mess you’d probably be looking at deleting the Connector Space for the MIM Service and then refreshing it from the MIM Service and these objects would be gone. However, this development environment is rather large, and that wasn’t something I had time or was prepared for at this time. So here’s how I worked around the issue.

Deleting spurious objects from the Connector Space

There’s two approaches;

  1. Select each of the errors, select the MIM Service Connector and select delete. That would work but I had thousands.
  2. Automate the process described in point 1. That’s the approach I took

Using the ever versatile Lithnet MIM Sync Powershell Module I retrieved the last run details for my MIM Service MA. I grabbed all the errors, inspected the errors for the ones that were failing creation to the MIM Service and then deleted the CSObject for that orphan.

Here’s where it got more than a little clink clink cowboy-ish. The Delete-CSObject cmdlet requires confirmation to delete the CSObject. There is not a switch to force the delete, or accept confirmation globally*. I wasn’t going to click Yes or press Enter 5000 times either.

So I wrote a small script that loops and checks for the Confirm disconnection dialog and sends the enter key to window.

Here’s the two little scripts.

This first script retrieves the last run details and loops through the errors.

This second script which I ran in a second separate PowerShell Runspace loops around and presses enter at the right time.

*I’ve submitted an enhancement request to Ryan to add a confirm parameter to Delete-CSObject

Adapting to the changes in the AzureAD Preview PowerShell Module ADAL Helper Library

I’m a big proponent of using PowerShell for integration and automation of Azure Active Directory Services using the Azure AD GraphAPI. You may have seen many of my posts leverage the evolving Azure AD Preview PowerShell Module helper libraries. Lines in my scripts that use this look like the one below. In this case using preview version 2.0.0.52.

# the default path to where the ADAL GraphAPI PS Module puts the Libs
Add-Type -Path 'C:\Program Files\WindowsPowerShell\Modules\AzureADPreview\2.0.0.52\Microsoft.IdentityModel.Clients.ActiveDirectory.dll'

The benefit of using this library is the simplification of Authentication to AzureAD, from which we can then receive a token and interact with the GraphAPI via PowerShell using Invoke-RestMethod.

Earlier this week it was bought to my attention that implementation of some of my scripts were failing when using the latest v2 releases of the AzureAD PowerShell Module (v2.0.0.98).  Looking into it the last version I had working is v2.0.0.52. v2.0.0.55 doesn’t work with my scripts either.  So anything after v2.0.0.52 the following will not work

What’s Changed?

First up the PowerShell Module has been renamed. It is no longer AzureADPreview, it is just AzureAD. So the path it gets installed into (depending on the version you have) is now;

'C:\Program Files\WindowsPowerShell\Modules\AzureAD\2.0.0.98\Microsoft.IdentityModel.Clients.ActiveDirectory.dll'

Looking into the updated PowerShell Module there has been a change to the Microsoft.IdentityModel.Clients.ActiveDirectory.dll library.

A number of the methods in the library have changed. I believe this is part of Microsoft transitioning the endpoint to use GraphAPI. With that understanding I approached using PowerShell to integrate with the GraphAPI more akin to the way I do when not using the helper library.

User PowerShell and the ADAL Helper Library to connect to AzureAD via the GraphAPI

Here is the updated script to connect (and retrieve a batch of users). You will need to update lines 4, 17 & 18 for your Tenant name and the username and password (non-MFA enabled) you will be connecting with.

Getting started configuring the latest Microsoft Identity Manager IBM Notes Management Agent with Domino v9.x

Lotus Notes. My old nemesis as both a user and as an administrator is back to haunt me again.

There’s a reasonable amount written by others on the trials and tribulations of getting the FIM/MIM Notes MA configured and working. However they are all referencing older versions of the MA and older versions of Domino. (If you are looking details on the previous versions checkout Michael’s great post here). The info on permissions is still valid, so make sure you’re on top of that.

The latest Notes MA (March 2017) is available from here and supports v9.x of Domino and v9.x of the Notes Client.

Here’s a couple of the differences straight up;

  • The v9 Lotus Notes Client does not require the v8 client requirements of setting up the MA using x86 architecture. Use “Process” as MA Architecture as per the documentation
  • I managed to get the v9.0.0 and v9.0.1 FP4 clients to work. The latest is v9.0.1 FP8 available from here
  • There are two technical references from Microsoft for the Notes MA. This is the old version. After some hunting I located the documentation for the latest release here. Yes the new version of the MA and the associated documentation only become available mid-March 2017

Here’s a couple of quick tips for getting up and running with the latest MIM Notes MA;

  • Install the v9.x Lotus Notes Client on the MIM Sync Server
    • Select Single User Install
  • Open the names.nsf and schema.nsf databases using the Notes ID you will be supplying to the Notes MA configuration

Enable Logging

Like anything, knowing what is going on, helps diagnose issues. This is how I enabled event log logging for the latest IBM Notes Management Agent

  • Edit the miiserver.exe.config file in the BIN folder. More than likely that will be in this path C:\Program Files\Microsoft Forefront Identity Manager\2010\Synchronization Service\Bin
  • Add the following lines as shown in the graphic below

<source name="ConnectorsLog" switchValue="Information">
   <listeners>
     <add name="ConnectorsLogListener" 
     type="System.Diagnostics.EventLogTraceListener, System, Version=2.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089"
     initializeData="ForefrontIdentityManager.ManagementAgent"
     traceOutputOptions="LogicalOperationStack, DateTime, Timestamp, Callstack" />
  </listeners>
</source>

Restart the Forefront Identity Manager Synchronization Service in services and when you run the Notes MA you will then see output in the Event Log under Forefront Identity Manager Management Agent.

Change the following line substituting Error, WarningInformation or Verbose for what you want logged.

 <source name="ConnectorsLog" switchValue="Information">

If you want it to log to the Application Log then change the initializeData value to whatever you want. I used IBM Domino MA.

 <source name="ConnectorsLog" switchValue="Information">
    <listeners>
    <add name="ConnectorsLogListener" 
    type="System.Diagnostics.EventLogTraceListener, System, Version=2.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089"
    initializeData="IBM Domino MA"
    traceOutputOptions="LogicalOperationStack, DateTime, Timestamp, Callstack" />
 </listeners>
    

… restart the Forefront Identity Manager Synchronization Service in services and when you run the Notes MA the logged data will land in the Application Log.

If you’ve used the Notes MA before, this should get you up and running with the latest.

Joining Identities between Active Directory and Azure Active Directory using Microsoft Identity Manager

Introduction

One of the foundations of Identity Management is the ability to join an identity between disparate connected systems. As we extend our management of identities into cloud services this adds a few twists.

A key concept is to use an anchor that is persistent. Something that doesn’t change through a users life-cycle. A user’s Security IDentifier (SID) in Active Directory is perfect. It doesn’t change when a user or group may get renamed.  What gets interesting is how the SID is represented when returned using different methods. That is what I quickly cover in this post.

Overview

The defacto method for connecting your OnPremise Active Directory to Azure Active Directory is to use Azure Active Directory Connect. AADC will synchronise users and groups SID’s to the corresponding object in AAD into the onPremisesSecurityIdentifier attribute. 

When the onPremisesSecurityIdentifier  attribute is retrieved via the GraphAPI the format looks like this: S-1-5-21-3878594291-2115959936-132693609-65242

Using an Active Directory Management Agent for FIM/MIM and synchronizing the objectSID from the OnPremise AD represents the value in the Metaverse in binary format which when viewed as text looks like this:  15000005210002431664623112825230126105190232781820

Translating SID formats so we can join identities

What we need to do is translate the string representation of the SID returned from the GraphAPI and AzureAD so that we have it in a binary format. Then we can then use those attributes in join rules to match users/groups between AzureAD and our OnPremise Active Directory.

 

In my environments I’m using the out of the box FIM/MIM Active Directory Management Agent. For Azure AD/Office 365 I’m using the Granfeldt PowerShell Management Agent to integrate with Azure AD via the GraphAPI.

overview1

On my AzureAD PowerShell Management Agent I have an attribute named AADonPremiseSID configured with the format as Binary in my PSMA Schema.ps1 as shown below.

$obj | Add-Member -Type NoteProperty -Name "AADonPremiseSID|Binary" -Value 0x10

On my Azure AD PSMA I have the following lines in my Import.ps1 which essentially takes the value retrieved from the GraphAPI S-1-5-21-3878594291-2115959936-132693609-65242 and converts it to a binary array that in text looks something like 15000005210002431664623112825230126105190232721825400 and stores it in the AADonPremiseSID binary attribute in the connector space.

# Create SID .NET object using SID string from AAD S-1-500-........ 
$sid = New-Object system.Security.Principal.SecurityIdentifier $user.onPremisesSecurityIdentifier
 
#Create a byte array for the length of the users SID
$BinarySid = new-object byte[]($sid.BinaryLength)

#Copy the binary SID into the byte array, starting at index 0
$sid.GetBinaryForm($BinarySid, 0)

#Add the SID to the user in the connector space
$obj.Add("AADonPremiseSID",$BinarySid)

This then lets me join my users (and groups using the same method) between AD and AAD.  Essentially a line to put it into Security Identifier format, two lines to convert it to a binary array and a line to store it in the connector space. Simple when you don’t over think it.

I’m posting this because I know I’m going to need to do this often. Hope it helps someone else too.

Migrating VirtualBox VDI Virtual Machines to Azure

Overview

Over the years I’ve transitioned through a number of laptops and for whatever reason they never fully get put out to pasture. Two specific laptops are used semi-regularly for functions associated with a few virtual machines they hold. Over the last 10 years or so, I’ve been a big proponent of VirtualBox. It’s footprint and functionality aligned with my needs. The downside these days is needing to sometimes carry two laptops just to use an application or two contained inside a Virtual Machine on VirtualBox.

It’s 2017 and time to get with the times. Dedicate an evening of working through the process of migrating those VM’s.

DISCLAIMER and CONSIDERATIONS Keep in mind that if you are migrating legacy operating systems, you’ll need a method to remote into them once they are in Azure. Check the configuration of them before  you convert and migrate them. Do they have firewalls? Is the network interface on the VM configured for dynamic or static addressing? Do the VM have remote access configured, VNC, RDP, SSH. As they are also likely to be less secure my process below includes a Network Security Group as part of the Azure Resource Group with no rules specified. You’ll need to add some inbound rules for the method you’ll be using to remote into your Virtual Machine. And I STRONGLY suggest locking those rules down to a single host or home subnet.

The VM Conversion Process

This blog post covers the migration of a Windows Virtual Machine in VDI format from VirtualBox on SUSE Linux to Azure.

  • With the VM Started un-install the VBox Guest Additions from the virtual machine

RemoveExtensions2

  • Shutdown the VM
  • In VirtualBox Manager select the VM and Settings
    1. Select Storage. If the VBoxGuestAdditions CD/DVD is attached then remove it.
    2. Take note of the VM’s disk(s) location (WinXPv2.vdi in my case) and naming. Mine just had a single hard disk. You’ll need the path for the conversion utility.

RemoveVBoxAdditions

RemoveAdditionsDVD

  • Virtual Box includes a utility named vboxmanage. We can use that to convert the VM virtual hard disk from VDI to VHD format. Simply run vboxmanage clonehd –format VHD –variant Fixed
    • You will need to make sure you have enough space on your laptop hard disk for the VHD which will be about the same size as your VDI Hard Disk
      • If you don’t on Linux you’ll get a slightly cryptic message like
        • Could not create the clone medium (VERR_EOF)
        • VBOX_E_FILE_ERROR (0x80bb0004)
      • the –variant Fixed switch is not shown in the virtual disk conversion screenshot (three images further down the page). One of my other VM’s was Dynamic. Size needs to be Fixed for the VHD to be associated with a VM in Azure
      • Below shows determining an existing disk that is Dynamic and needs to be converted to Fixed

DynamicDisk

  • Below shows determining an existing disk that is Fixed and doesn’t need to be converted

FixedDisk

  • Converting the VDI virtual disk to VHD

Convert60Percent

Preparing our Azure Environment for our new Virtual Machine

  • Whilst the conversion was taking place I logged into the Azure Portal and created a new Resource Group for my VM to go into. I also created a new Storage Account in that Resource Group to put the VM’s VHD into. Basically I’m keeping these specific individual VM’s that serve a very specific purpose in their own little compartment.

RGandSG

  • Using the fantastic Azure Storage Explorer which works on Linux, Mac and Windows I created a Blob Container in my newly created Storage Account named vhds.

CreateBlobContainer

Upload the Converted Virtual Hard Disk

  • By now my virtual disk had converted. Using the Azure Storage Explorer I uploaded my converted virtual disk. NOTE Make sure you have the ‘upload vhd/vhdx files as Page Blobs’ selected. 

UploadVHD

For a couple of other VM’s I wrote a little PowerShell script to upload the VHD’s to blob storage.

Create the Azure VM

The following script follows on from the Resource Group, Storage Account and the Virtual Machine Virtual Disk we created and uploaded to Azure and creates the VM to attached to the virtual disk.

All the variables are up front, we create the Network Security Group, Subnet and Virtual Network. Then the Public IP and Network Interface. Finally we define the details for the VM with the networking and the uploaded VHD before creating the VM.

And we’re done. VM created and started.

VMCreated
Happy days and good bye to a number of old laptops.

Standalone installation of the MIM Self Service Password Reset Portals ends prematurely

Today I was performing a standalone installation of the MIM Self Service Password Reset Portals (Enrollment and Reset). These Portals rely on IIS and not the normal prerequisites associated with the MIM Service Portal (SharePoint etc).  As such using PowerShell I’d only installed the Web Server Role with the usual dependencies.

On starting the MIM Service and Portal installation I got the dreaded Microsoft Identity Manager Service and Portal Setup Wizard ended prematurely dialog. So straight into debug mode running with an installation log as per the command below.

msiexec /i “f:\Service and Portal\Service and Portal.msi” /l*v c:\temp\install.log

InstallTerminatedPrematurely

Into the install.log file created from the installation process I see the following.

Looking into the install log I notice the following error.

InstallLogError

In a more readable form

SFXCA: Ensure that the proper version of the .NET Framework is installed, or that there is a matching supportedRuntime element in CustomAction.config. If you are binding to .NET 4 or greater add useLegacyV2RuntimeActivationPolicy=true to the element.
CustomAction GetIISVersionFromRegistry returned actual error code 1603 (note this may not be 100% accurate if translation happened inside sandbox)

I have .NET Framework 4.5 installed. Could it be MIM 2016 SP1 still needed .NET Framework 3.5 ?

DotNETVersions

Taking a punt I installed .NET 3.5.

DOTNET35Install

Restarted the Install. Success.

Success

There you go. Even when you’re only looking to have the SSPR Portals installed you need to have .NET Framework 3.5 installed. And FYI I needed to have the Windows Server 2012 R2 media handy to get the sources for .NET Framework 3.5.

A workaround for the Microsoft Identity Manager limitation of not allowing simultaneous Management Agents running Synchronisation Profiles

Why ?

For those of you that may have missed it, in early 2016 Microsoft released a hotfix for Microsoft Identity Manager that included a change that removed the ability for multiple management agents on a Microsoft Identity Manager Synchronization Server to simultaneously run synchronization run profiles. I detailed the error you get in this blog post.

At the time it didn’t hurt me too much as I didn’t require any other fixes that were incorporated into that hotfix (and the subsequent hotfix). That all changed with MIM 2016 SP1 which includes functionality that I do require. The trade-off however is I am now coming up against the inability to perform simultaneous delta/full synchronization run profiles.

Having had this functionality for all the previous versions of the product that I’ve worked with for the last 15+ years, you don’t realise how much you need it until its gone. I understand Microsoft introduced this constraint to protect the integrity of the system, but my argument is that it should be up to the implementor. Where I use this functionality a LOT is with Management Agents processing different object types from different connected systems (eg. Users, Groups, Contacts, Photos, other entity types). For speed of operation I want my management agents running synchronization run profiles simultaneously.

Overview of my workaround

First up this is a workaround not a solution to actually running multiple sync run profiles at once. I really hope Microsoft remove this limitation in the next hotfix. What I’m doing is trying to have my run sequences run as timely as possible.

What I’m doing is:

  • splitting my run profiles into individual steps
    • eg. instead of doing a Delta Import and a Delta Sync in a single multi-step run profile I’m doing a Delta Import as one run profile and the sync as a separate one. Same for Full Import Full Synchronization
  • running my imports in parallel. Essentially the Delta or Full imports are all running simultaneously
  • utilising the Lithnet MIIS Automation Powershell Module for the Start-ManagementAgent and Wait-ManagementAgent cmdlets to run the synchronisation run profiles

Example Script

This example script incorporates the three principles from above. The environment I built this example in has multiple Active Directory Forests. The example queries the Sync Server for all the Active Directory MA’s then performs the Imports and Sync’s. This shows the concept, but for your environment potentially without multiple Active Directories you will probably want to change the list of MA’s you use as input to the execution script.

Prerequisites

You’ll need;

The following script takes each of the Active Directory Management Agents I have (via a dynamic query in Line 9) and performs a simultaneous Delta Import on them. If your Run Profiles for Delta Imports aren’t named DI then update Line 22 accordingly.

-RunProfileName DI

With imports processed we want to kick into the synchronisation run profiles. Ideally though we want to kick these off as soon as the first import has finished.  The logic is;

  • We loop through each of the MA’s looking for the first MA that has finished its Import (lines 24-37)
  • We process the Sync on the first completed MA that completed its import and wait for it to complete (lines 40-44)
  • Enumerate through the remaining MA’s, verifying they’ve finished their Import and run the Sync Run Profile (lines 48-62)

If your Run Profiles for Delta Sync’s aren’t named DS then update Line 40 and 53 accordingly.

-RunProfileName DS

Summary

While we don’t have our old luxury of being able to choose for ourselves when we want to execute synchronisation run profiles simultaneously (please allow us to Microsoft), we can come up with band-aid workarounds to still get our synchronisation run profiles to execute as quickly as possible.

Let me know what solutions you’ve come up with to workaround this constraint.