550 5.4.316 Message expired, connection refused (Socket error code 10061)

In a recent engagement of Exchange migration to O365 the client started experiencing random inbound delivery failures. The error in the NDR was that in the title:

Research pointed to a couple of articles that all suggest checking the firewall:

• O365 sources blacklisted/quarantined by an over-zealous Fortigate IPS rule - https://pariswells.com/blog/research/office-365-failed-550-5-4-316-message-expired-connection-refusedsocket-error-code-10061
• Microsoft's own experience and recommendation - https://docs.microsoft.com/en-us/office365/securitycompliance/mail-flow-intelligence-in-office-365

It turned out to be not a firewall, but a case of asymmetric routing. Close enough. A new device has been introduced into the customer's environment to set up a VPN with a sister company at around the same time when the first delivery error reports started to came in. Setting up the VPN affected routing, resulting in egress/ingress SMTP traffic to/from the same source took very different paths.

Once routing has been corrected, email started to flow normally again.

Till next time.

New Office Build Causes Credentials Prompt Loop

Another day, another challenge.

We've just provisioned a brand new laptop with a brand new Office (latest build) for a user. Yet we could not configure a mail profile: it kept prompting for credentials, no matter what we did. We tried DOMAIN\samAccountName as well as UPN to no avail.

The user is in a hosted Exchange environment with a UPN of domain.local on Exchange 2013.

Interestingly other users worked well.

To cut the long story short, it looks like the newest Office build requires that the domain component of the user's UPN matches the domain component of the primary SMTP address. The user components need not match.

Here is what failed and what worked in my test:

• DOMAIN\samAccountName FAILED
• user@domain.local FAILED
• UPN fully matches email address WORKED
• domain components of UPN and email address match, user components are different WORKED

Office builds that displayed this behavior are 1809 and 1810. Older builds worked with any of the above combinations. In my troubleshooting I used the latest Office 365 suite, 16.0.11001.20064 at the time of this writing, which failed/worked as described above.

Is it a bug? Is it Microsoft silently pushing users to adhere to standards imposed by Office 365 EXO? Can't tell. The fact is that it looks like we are being forced to adopt consistency. Not necessarily a bad thing, although without adequate documentation on Microsoft's part it is going to cause some major headaches in a number of organizations.

Lessons learned: start sticking to de facto standards and best practice: match users' UPN to their primary SMTP address.

Have a nice weekend.

Microsoft Builds Licensing Engine into the Exchange HCW

Hi There,

I am critical when Microsoft bricks my servers with dodgy updates, but I also give them kudos when due. Happy to say that it is kudos time.

In its July 20, 2018 update of the HCW, Microsoft added a welcome feature. As you may already be aware, in a hybrid environment where there are no local mailboxes anymore and the on-prem Exchange server is there purely for administrative purposes, you're eligible for a free Hybrid license. If you didn't know then click here to find out more. It used to be a separate process, a separate tool, thus extra time and administration.

Not anymore. If you run the HCW on a new, unlicensesd server, the first thing you'll notice is a big red message telling you that you are running an Unlicensed Product, and the Next button is grayed out. Also, the Version information states that the server is a "Standard Evaluation Edition".

Don't freak out. If you look closer, you now get a link to license this server now:

Click the link and you'll be prompted to log on to your O365 tenant admin account:

The wizard then goes on to validate the environment, then obtain and install your free, brand spanking new Hybrid edition license:

Once the license is in, you are given access to the Next button, and the rest of the process is pretty much the same as before. As an added feature, you are also given a copy product key link which reveals not only the product key in clear text, but the entire PowerShell command that was used to install it also - it's there just as an FYI, the wizard did it all for you.

If you restart the HCW, you'll notice that Standard Evaluation version has been updated to Coexistence Edition:

Pretty cool, huh?

Till later,

Zoltan

.NET Framework 4.7.2 Breaks AAD Connect and Exchange

Hi There,

Time for a new post.

Microsoft made .Net Framework 4.7.2 available on Windows Update on 10 July 2018, just about a week ago. As an "Important / Recommended" update, it gets under the radar at many organizations where all "Important" updates are installed as default practice. .NET updates used to come as "Optional". This time, however, Microsoft deemed this update "Important" for whatever odd reason that escapes me.

Although Microsoft "strongly recommends" the installation of this update, reports have emerged that it doesn't play nicely with AAD Connect. and Exchange. Specifically, CPU utilization of the Microsoft.Identity.Health.AadSync.MonitoringAgent.Startup.exe process goes through the roof, grinding the server to a halt:

Secondly, Microsoft has not (yet) updated the Exchange server prerequisites to reflect support for .NET Framework 4.7.2 - see https://docs.microsoft.com/en-us/exchange/plan-and-deploy/system-requirements:

Sure enough, the update bricks the Exchange OWA and ECP portals too. After you log on, you get a pristine, white browser window, devoid from anything:

I thought OK, let's rebuild some virtual directories. Well, for that I need EMS - as long as it works. It fell flat too:

In fact, looking at the IIS logs, it becomes clear that pretty much everything has gone south.

As recovery steps, first I removed .Net 4.7.2 as some sources indicate on the Internet. Unfortunately that didn't fix the AAD Connect high CPU problem - it returned after an hour or so. And it certainly didn't fix the Exchange problem.

As far as Exchange is concerned, I tried the following:

• Removed .Net 4.7.2
• Removed and reinstalled .Net 4.7.1
• Installed Exchange 2013 CU21 - the server was a tad outdated, on CU13

No joy. The screenshots above were taken after the recovery attempt.

My recommendation to you, dear reader, is to block the installation of .Net 4.7.2 for the time being. It is NOT an "important" update, no matter how much Microsoft would like you to believe.

The update can be blocked with a Registry setting, as documented at KB4342394.

I am in for rebuilding the Exchange server bricked by Microsoft's (not so) "important" .Net update. Thank you Mr. Microsoft, yet another .Net blunder to add to the list.

Happy patching!

Add-Endum

Microsoft has come to its senses and re-published .NET Framework 4.7.2 where it belongs, under "Optional" updates.

Changing the Garbage Collection Mode on MSExchangeMapiFrontEndPool

Hi There,

I stumbled on a script at https://gallery.technet.microsoft.com/office/Exchange-2013-Performance-23bcca58, referenced by https://blogs.technet.microsoft.com/rmilne/2015/04/06/exchange-2013-performance-health-check-script/.

Ran the script on an Exchange 2013 system, and among others, it threw an error for the MSExchangeMapiFrontEndAppPool using Workstation garbage collection mode instead of Server mode:

Not being a .Net developer, looked up the issue, and all sources that I could find indicated that the garbage collection should be configured for Server mode for improved performance. Just as the script said.

I was (and still am) gobsmacked as to why the Exchange team chose this mode while everything else is optimised to the extreme.

The script is great, kudos for Marc Nivens, except it doesn't provide any reference as to how to go about changing the garbage collection mode, which would be very helpful for folk like me who aren't developers.

So here is how to do it:

First, we need to find where garbage collection is configured. For that we need to identify the configuration file for the .Net application. There are a number of ways, two of which are shown below:

1. The DOS way:

1	%WINDIR%\System32\Inetsrv\appcmd list apppool "MSExchangeMapiFrontEndAppPool" /text:"CLRConfigFile"

2. The PowerShell way (in case you want to script it):

1 2 3

[System.Reflection.Assembly]::LoadWithPartialName("Microsoft.Web.Administration") | Out-Null $serverManager = new-object Microsoft.Web.Administration.ServerManager (($serverManager.ApplicationPools | ?{$_.Name -eq "MSExchangeMapiFrontEndAppPool"}).Attributes | ?{$_.Name -eq "CLRConfigFile"}).Value

Essentially the file is %ExchangeInstallPath%\bin\MSExchangeMapiFrontEndAppPool_CLRConfig.config

Open the file in an elevated Notepad and change the <gcServer enabled="false" /> line to <gcServer enabled="true" />. Here is the file with the default value:

After the change (and a server restart) I got this status:

A word of caution: When a new CU is installed then it overwrites the config file, thus reverting back the garbage collector to Workstation mode. Don't forget to update the file again after the CU is installed.

Reference: https://msdn.microsoft.com/en-us/library/cc165011(v=office.11).aspx

Happy garbage collecting!

Assign or Remove O365 Licenses

Hi Folks,

Here is a script for managing O365 licenses. Get it from the TechNet Gallery.

If you work with O365 then sooner or later you''ll be asked to assign some licences - or remove them to that matter - to more than just one user. Using the portal will take forever.

A PowerShell script is the natural option, but it may be daunting to get the hang of the licensing object structures and to get your head around building a DisabledOptions list of services of what NOT should be enabled as opposed to just say "this is enabled, that is not".

I've put together a script which will do it all: it will enable or disable a particular service within a plan.

The script implements an interactive, user-driven flow of actions, which resembles the way NTDSUTIL or DISKPART work.

Prerequisites: You'll need the Microsoft Azure Active Directory Module for Windows PowerShell.

How it works:

1. Create a text file with all the UPNs for all the users who will have a license assigned or removed. No header line, just a list of UPNs:

In this example we have 3x valid users, 1x non-existent user and 1x incorrectly formatted user - will see its significance later in this post.

2. Run the script:

Manage-O365Licenses.ps1 -InputFile "Your_Text_File_With_UPNs.txt" -Location CountryCode

where the country code is the ISO 3166 Alpha-2 two-letter code of the country that the users will be assigned to. For the full list of country codes see https://www.iso.org/obp/ui/#search/code/.

For example, if your users' UPNs are listed in Users.txt and they are all in Australia, you would use the following command ...:

Manage-O365Licenses.ps1 -InputFile Users.txt -Location AU

... or simply launch the script with no parameters at all and it will prompt for the file and location:

Manage-O365Licenses.ps1

NOTE: If your users are distributed across different countries then you'll have to run the script separately for each country. It doesn't allow for mixing locations in one run.

If you happen to be already connected to O365 (you ran Connect-MSOLService previously and authenticated successfully) then the script will move on to the next step. Otherwise it will prompt for credentials, then it will connect to O365.

3. Select what you want to do: Disable or Enable licenses:

Select 0 or 1 and press Enter, or X to exit.

Your selection will be confirmed in green (see next screenshot).

4. Select the plan. Same process: enter a number, or X to exit:

Some plans are exposed in PowerShell, however they are hidden in he GUI. I couldn't identify a reason. If it isn't in the GUI then probably it shouldn't be touched. These show up as "--- not available for selection in the Portal ---".

Also, the list only contains the plans that I have come across, and it is probably incomplete, or Microsoft will come up with new ones. If you come across such a plan then its cryptic name will be displayed with "(no friendly name found)" appended, similar to the following:

For the record, the screenshot is a simulation of what you may encounter. The SPE_3 plan is the equivalent of "Microsoft 365 E3", and I removed the definition for the sake of illustration. It is back in the version that's available for download in the TechNet Gallery.

5. Select the service. Same process: enter a number, or X to exit. For illustration, I typed a couple of invalid selections to demo how it is handled:

6. Then a last chance of change of heart: review your selections and select Y or N:

The script runs and displays the statistics:

What do the stats mean? Let's open the license.log file:

There is an entry for each user:

• SUCCESS: The operation has completed successfully for the user.
• NOCHANGE: The license is already at the desired state, no change needed.
• FAIL: The user does not exist in the organization (e.g. fakeuser@yourorg.com.au) or it failed format validation (e.g. gibberish is not in the correct format).
• NOAVAILABLELICENSE: Ran out of licenses (not illustrated here).

WARNING: The script overwrites the log file! If you want to preserve the logs between different runs then move or rename the file before you run the script again.

The scripts will also catch some errors. In case an error is caught, a debug.log file is created. It is only created if a cmdlet generates one of those red errors.

Just like the license.log file, debug.log is wiped every time the script runs. However it is only created if a cmdlet generates an error. It looks like this ...:

... and its associated entry in the license.log file:

Some error handling features:

• Validation of input file. Exits if it doesn't exist.
• Validation of country code. Exits if it is invalid.
• Graceful exit if credential input is cancelled.
• Graceful exit if wrong credentials are provided.
• Graceful handling of invalid selection input (out of range or invalid characters).
• Option to exit at any point.
• UPN format validation.
• Existing user validation.
• Licence count validation.
• Catches some error conditions with error details in debug.log. No error, no log.
• Detailed record of changes and other stats in license.log (not an error but good to have).

Licence Agreement

The scripts are licenced under the NMF (Not My Fault) agreement.
Edit, change and use the scripts at your own risk.
Give credit where due.

Download: Manage-O365Licenses.ps1

Enjoy :-)

Exchange 2016 CU7 Bug Causes HCW to Fail

Hi There,

If you plan to deploy a hybrid Exchange server using Exchange 2016 CU7, then don't. It's broken. Use a pre-CU7 server instead.

A bug slipped into Exchange 2016 CU7 which prevents the HCW from completeing. The HCW fails to get past the domain ownership validation:

No matter how hard you try, you can't get past this screen.

Looking at the HCW log at C:\Users\<username>\AppData\Roaming\Microsoft\Exchange Hybrid Configuration, the following errors are logged:

*ERROR* 10300 [Client=UX, Thread=1] System.ArgumentOutOfRangeException: The UTC time represented when the offset is applied must be between year 0 and 10,000.
                                Parameter name: offset
                                   at System.DateTimeOffset.ValidateDate(DateTime dateTime, TimeSpan offset)
                                   at System.DateTimeOffset..ctor(DateTime dateTime)
                                   at Microsoft.Online.CSE.Hybrid.App.AppData.BuildSessionProperties(NotificationType type)

...

*ERROR* 10277 [Client=UX, Activity=Domain Ownership, Session=OnPremises, Cmdlet=Set-FederatedOrganizationIdentifier, Thread=14] FINISH Time=2598.6ms Results=PowerShell failed to invoke 'Set-FederatedOrganizationIdentifier': Object reference not set to an instance of an object. An unexpected error has occurred and a Watson dump is being generated: Object reference not set to an instance of an object.

...

*ERROR* 10224 [Client=UX, Page=DomainProof, Thread=14] Microsoft.Online.CSE.Hybrid.PowerShell.PowerShellInvokeException: PowerShell failed to invoke 'Set-FederatedOrganizationIdentifier': Object reference not set to an instance of an object. An unexpected error has occurred and a Watson dump is being generated: Object reference not set to an instance of an object. ---> System.Management.Automation.RemoteException: Object reference not set to an instance of an object.
                                   at System.Management.Automation.Runspaces.AsyncResult.EndInvoke()
                                   at System.Management.Automation.PowerShell.EndInvoke(IAsyncResult asyncResult)
                                   at Microsoft.Online.CSE.Hybrid.Provider.PowerShell.PowerShellProvider.PowerShellInstance.Invoke(String cmdlet, IReadOnlyDictionary`2 parameters, Int32 millisecondsTimeout)
                                   --- End of inner exception stack trace ---
                                   at Microsoft.Online.CSE.Hybrid.PowerShell.RemotePowershellSession.RunCommandInternal(Cmdlet cmdlet, SessionParameters parameters, Int32 millisecondsTimeout, PowerShellRetrySettings retrySettings, Boolean skipCmdletLogging)
                                   at Microsoft.Online.CSE.Hybrid.Session.PowerShellOnPremisesSession.SetFederatedOrganizationIdentifier(SmtpDomain accountNamespace, String delegationTrustLink, SmtpDomain defaultDomain)
                                   at Microsoft.Online.CSE.Hybrid.App.ViewModel.Pages.DomainProof.DomainInfo.AddFederatedDomain(IOnPremisesSession session, AppData appData)
                                   at System.Collections.Generic.List`1.ForEach(Action`1 action)
                                   at Microsoft.Online.CSE.Hybrid.App.ViewModel.Pages.DomainProof.VerifyActivity(IOnPremisesSession session, EnvironmentBase environment)

Issues with the HCW have also been reported by admins on Rhoderick Milen's blog at https://blogs.technet.microsoft.com/rmilne/2017/09/19/exchange-2016-cu7-released/. 

Microsoft is aware of it, and a fix will be included with CU8. Some suggest using a pre-CU7 server to run the HCW on - I haven't tried, not (yet) an option in the affected environment.

Update:

• Tried running the HCW on a CU6 server that I installed temporarily. Since the CU7 server was still selected as the hybrid server, the wizard FAILED.
• Tried the workaround suggested by AloneInTheDarK. While some reported success, it didn't work for me - see https://social.technet.microsoft.com/Forums/en-US/3ac7cfe6-6c97-4f68-84b9-d498aafd4ea1/exchange-2016-cu6?forum=Exch2016Adm. Microsoft advised that it is not a recommended workaround, even if it would have worked.
• Downgraded the CU7 server to CU6. SUCCESSFUL. It involved:
• Installed a temporary CU6 server to host all the mailboxes specific to Exchange 2016 while the CU7 server was downgraded (Arbitration, AuditLog etc. - see http://ezoltan.blogspot.com/2016/06/delete-that-stubborn-exchange-2016.html and see https://technet.microsoft.com/en-us/library/mt441791(v=exchg.150).aspx). This is needed so that the CU7 server can be uninstalled.
• Uninstalled Exchange 2016 CU7 and deleted the binaries from the installation folder.
• Installed Exchange 2016 CU6 on the same machine.
• Configured the downgraded server (virtual directories, certificate etc.), and moved all system mailboxes back to it..
• Uninstalled the temp Exchange 2016 server.
• Ran the HCW wizard - SUCCESSFUL

It looks like Microsoft has a hotfix. At this time I don't know whether it will ever be published as CU8 is due in the near future, but if you are in dire need then open a case and request the hotfix. I didn't get to test because it came in one day late, after I went through the downgrade ordeal.

Take care,

Zoltan