Hooray and welcome to 2023! May all of your dreams come true this year, and if not, may we at least be sheltered from P1 outages and other disasters, either natural or man-made…!
🤞🙏🤞🙏🤞🙏🤞🙏🤞🙏🤞🙏🤞🙏🤞🙏🤞🙏
Recently I needed to see if RDP was open on hundreds of machines. No trouble I thought, and wrote a wee script that looped through a list of names or IPs using Test-NetConnection with the -Port switch.
This worked no problem, BUT… Test-NetConnection can take 15+ seconds per machine (especially when waiting to ‘time out’ if the port is closed). Not good enough, we want it to be faster!
Unfortunately, Test-NetConnection does not currently support any switches to control how long it waits for a response, although I’m sure they will add it at some point.
So, a Googling we go then! And again, I am reminded of my average intelligence when it comes to PowerShell. But that’s okay – learning all the time is a good thing, right? 😜
I found this little beowtae of a function called Test-PortScan which uses the TcpClient object to do the job faster. The original is found here. Thanks to the author ‘BSonPosh’. 👌
As I created this script I also thought “I’d like any functions to be at the bottom of the script, so the main editable code is at the top”. Turns out this is quite easy to do by storing the main code in a variable, then calling it after the functions. Full script below; you’ll want to edit the variables at the top – Name of the service or test, the port number, the timeout value (I went with half a second) and change the import and logfile locations if required. Cheers. Simon! 🤟🙂🤟
# uses test-portscan function to quickly check whether a port is open on multiple IP4 endpoints
$mainblock = {
Clear-Host
# set variables
$logfile = 'c:\temp\portcheck_log.txt'
$servicename = 'HTTP'
$portnumber = '80'
$timeoutmilliseconds = '500'
# import single column of endpoints from a file (no header required)
$endpoints = Get-Content C:\Temp\endpoints.txt -Force
# or use an array
# $endpoints = @(
# "192.168.1.254"
# "1.1.1.1"
# "8.8.8.8"
# "www.microsoft.com"
# "www.bleepingcomputer.com"
# )
# create log file
$null = New-Item -ItemType File -Path $logfile -Force
# perform the checks
foreach ( $endpoint in $endpoints ) {
Write-Host "Testing $testname access to $endpoint..."
if ((Test-PortScan -Devices $endpoint -StartPort $portnumber -EndPort $portnumber -Timeout $timeoutmilliseconds).Open -ne 'False') {
Write-Host -ForegroundColor Gray "$testname is not open on $endpoint."
}
else {
Write-Host -ForegroundColor Green "$testname is open on $endpoint! - adding to $logfile"
$output = "$testname is open on $endpoint."
$output | Out-File $logfile -Append -Force
}
}
# end of mainblock
}
# port checking function
function Test-PortScan
{
[CmdletBinding()]
param
(
[Parameter(Mandatory = $true,
ValueFromPipeline = $true,
ValueFromPipelineByPropertyName = $true,
Position = 0,
HelpMessage = 'You must supply at least a device name or IP')]
[Alias('Name')]
[string]$Devices,
[Parameter(Position = 1)]
[int]$StartPort = 1,
[Parameter(Position = 2)]
[int]$EndPort = $StartPort,
[Parameter(Position = 3)]
[int]$Timeout = 100
)
BEGIN
{
$PSObject = @()
$Output = @()
}
PROCESS
{
foreach ($Device in $Devices)
{
Write-Verbose "Not to the port loop yet"
$Port = $StartPort
do
{
#($Port = $StartPort; $Port -gt $EndPort; $Port++)
Write-Verbose "Made it to the port loop"
$requestCallback = $state = $null
$client = New-Object System.Net.Sockets.TcpClient
$beginConnect = $client.BeginConnect($Device, $Port, $requestCallback, $state)
Start-Sleep -Milliseconds $Timeout
if ($client.Connected) { $Open = $true }
else { $Open = $false }
$client.Close()
[pscustomobject]@{
'Computer' = $Device
'Port' = $Port
'Open' = $Open
}
$Port += 1
}
until ($Port -gt $EndPort)
}
}
END
{
}
}
# run mainblock
& $mainblock
# end
Using forms in PowerShell allows you to run interactive scripts that provide information or prompt for input. Ever need to force everyone to reboot? I know, I know… why on earth would you spend time on that Simon? Well (a) I’m absolutely bonkers… and (b) it’s fun! –ahem, refer to (a).
The main problem with using forms with PowerShell is that the script needs to run in the logged on users context, which prevents us performing any admin tasks. Depending on the requirements you may also need to deploy some PowerShell modules out to the workstations… uuurg that is already sounding OTT!
Well, for that I’ll share another post soon where I took a ‘queue’ approach and split the code to have a workstation-side script that runs in user context, and a server-side script that runs under a service account (assigned whichever admin permissions are required for its tasks). The workstation script gathers required data using forms etc., then writes a csv file into a shared folder on the server. The server script is checking the folder, sees a new file and processes it. Once a successful result is confirmed, the file is deleted. Meanwhile the workstation script has been in a ‘Please wait’ state, checking for the file deletion so it can continue knowing a server-side task is complete. “Coolio Glesias” I hear you say!
Anyway, that’s a longer post for another time! 🙂🙂🙂
This script below will prompt the user to reboot their workstation as soon as possible, with a mandatory reboot countdown visible. The form cannot be closed (unless the user can kill the PowerShell process – block that with a GPO that disables Task Manager and command line access). If the countdown reaches zero, a reboot is forced.
You can use GPO to deploy the script to a folder on each workstation, and also to configure a Scheduled Task for running it. Or in my case I am importing it into an MSP platform so it can be run automatically or on demand when a reboot is pending on client computers.
Aside from what it does (which is nothing fantastic!),it’s good example of using forms in PowerShell. I’ll post about my cross-domain password sync script soon which adds different types of message box as functions. Until then, Hei Konei Ra and have a Merry Xmas!
🧑🎄🍻🧑🎄🍻🧑🎄
# displays a non-dismissable warning with countdown in top left of screen
# force the local machine to reboot after $timeinseconds e.g. 28800 seconds = 8 hours.
# update variables as required
# default 8 hour setting
$timeinseconds = 28800
# icon for top-left of forms
$formicon = "$env:SystemRoot\UUS\amd64\WindowsUpdateWarning.ico"
# load the stuff we need for the forms
Add-Type -AssemblyName System.Windows.Forms
Add-Type -AssemblyName System.Drawing
# start the countdown
$Form = New-Object system.Windows.Forms.Form
$Form.Text = "SBEnterprises - Reboot Required"
$Form.Location = New-Object System.Drawing.Point(15,10)
$Form.AutoSize = $true # .Size = New-Object System.Drawing.Size(350,250)
$Form.AutoSizeMode = 'GrowAndShrink'
$Form.ShowInTaskbar = $false
$Form.Icon = New-Object System.Drawing.Icon "$formicon"
$Form.ShowIcon = $true
$Form.Font = New-Object System.Drawing.Font("Segoe UI",10)
$Label = New-Object System.Windows.Forms.Label
$Label.Location = New-Object System.Drawing.Point(5,10)
$Label.AutoSize = $true # .Size = New-Object System.Drawing.Size(350,250)
$Label.TextAlign = 'MiddleCenter'
$Label.UseWaitCursor = $false
$Form.Controls.Add($Label)
While ($timeinseconds -gt 0) {
$timespan = New-TimeSpan -Seconds $timeinseconds
$hour = $timespan.Hours
$min = $timespan.Minutes
$sec = $timespan.Seconds
$Form.Show()
$Label.Text = "WARNING!`nYour workstation requires a reboot to finish installing updates...`nPlease reboot as soon as possible.`n`nA mandatory reboot will occur in $hour hours, $min minutes, $sec seconds.`n`n"
Start-Sleep -Seconds 1
$timeinseconds = $timeinseconds - 1
}
$Form.Close()
Restart-Computer -Force
# end
Due to the popularity of the initial script (over 5000 views and 3rd in the list on a google search – cheers!) Azure AD – Export Groups and Members to CSV, and thanks to David for asking, this script goes next level and will export the groups and the members with properties ObjectID, Display Name, UserPrincipalName and Email Address. It caters for the main member ‘types’, User, Device, Group, and Contact. If another type of object is a member the output will say ‘Unknown object’. This could be a service principal or other object which you can investigate using the ObjectId. If a group has no members, ‘No members’ is output as the member display name.
The script uses the AzAD cmdlets as well as the AzureAD cmdlets, so make sure you have installed and imported them.
To install them:
Install-Module Az -SkipPublisherCheck -Force -AllowClobber -Confirm:$false
Just a quick one on ABPs – there are many posts about this topic now, but there are few that mention how to resolve issues such as recipients not appearing in the list when they should be. In this case you need to ‘tickle’ (yes that’s the official Microsoft term) the objects to get them to play ball.
ABPs seemed complex when I first looked at them, and my first introduction was with a tenant that had 20,000 user objects! We don’t get many opportunities to work with environments of this size in New Zealand, so it was a great job to get involved with. #dontmessitupmaaate!
ABPs are most commonly used in large environments though, or where separation is needed. Examples of this are:
Multiple schools under one tenant – you don’t want students from one school seeing students from other schools in the Global Address List.
Multiple companies under one tenant. You don’t want Fabrikam users seeing Contoso users.
In both of these scenarios, you may want management or executive level staff to see all recipients in their GAL.
Here is the code to create each ABP… I’m using my fictional company SB Enterprises which exists in a large multi-company tenant. Customattribute1 is used across the tenant to identify the objects related to a particular company with a three-letter-acronym. (this could be the Company attribute, or any other attribute as long as we can use it for filtering in the commands). For Meeting Room, Equipment, or Contact objects, those are created with the company TLA at the front e.g. SBE_MeetingRoom1.
Firstly let’s connect to Exchange Online powershell:
Looking good? Now let’s create the Global Address List and the Offline Global Address List:
# the global address list has it's own filter for all SBE objects
New-GlobalAddressList -Name "SBE_GlobalAddressList" -RecipientFilter "(CustomAttribute1 -eq 'SBE') -or (Name -like 'SBE_*')"
# the offline address list includes the address lists we created
New-OfflineAddressBook -Name "SBE_OfflineAddressList" -AddressLists "SBE_AddressList","SBE_Groups","SBE_Shared Mailboxes","SBE_Rooms","SBE_Contacts"
Great! Now we can create the Address Book Policy using all of the above:
Done! Well almost – we have to assign it to the users… I recommend assigning to a pilot group first to get some feedback in case of any issues. When ready, use this command to assign the ABP to all applicable users:
Now, there are several things that can make it seem like things are not working, but I can tell you 99% of the time you just have to wait. It’s the old ‘cloud time phenomenon’ where things may take from 1 to 48 hours to take effect 🤣🤣🤣.
The most common issue I have come across is someone in the pilot group pointing out that someone or something is missing from the GAL. This is due to the object not being processed when the Address List was created. It should be a member of the filter; and it is upon object creation or update that membership of address list filters is determined.
This is where ‘tickling’ comes in. You need to change something i.e. any attribute of the offending object, then change it back again.
You can do this in the portal for a single object (e.g. change the last name one letter, save, then change it back again)… but seeing as I know this problem exists, I now do this as part of the initial setup so I don’t have to deal with it later.
Let’s run this to change an attribute – I’ve checked all objects Customattribute5 is blank, so I can use it for this purpose (you don’t have to use tickle obviously, any value will do):
# get the users we want to 'tickle'
$users = Get-User -ResultSize unlimited -Filter "Customattribute1 -eq 'SBE'"
# tickle them by modifying a value (make sure the value was null for all objects beforehand)
foreach ($user in $users) {
$id = $user.DistinguishedName
Set-Mailbox $id -CustomAttribute5 "tickled"
}
# then return the value to null
foreach ($user in $users) {
$id = $user.DistinguishedName
Set-Mailbox $id -CustomAttribute5 $null
}
And voila, the object now appears on the list (taking into account the ‘cloud time phenomenon’ mentioned above).
As always, thanks for coming dudes & dudettes! ✌🍻✌
I’m starting off-topic (it does say Ramblings at the top of the screen after all) but stay with me!
😃👀🍻👀😃
Hiding your DNS might seem like a good idea; I have nothing against as long as it is free, but to me it is like changing RDP port 3389 to 4444 (not that you have RDP directly exposed to the internet of course – you have an RD Gateway!). Sure, you’ve made it harder to identify an available target, but if someone is really after you they’ll have methods to get around this pretty quickly using port scans or other techniques. You’ve also made it harder for people trying to help to work out what is going on.
The best thing you can do is hurry up on that cloud journey so you can offload concerns about external IP addresses exposing your on-premises entry points. And as long as you are keeping those entry points secure and up to date, there shouldn’t be any major concern here anyway.
If you are running RDS or Citrix services, generally these do not work well with SSL inspection or pre-authentication so are configured to pass-through directly to the entry point i.e. Cloudflare is providing minimal benefit here.
Instead of being fancy pants and paying for these services to provide minimal benefit, let’s look at a great set up for minimal cost that let’s you easily add, remove or change DNS entries right in the portal you use everyday!
First, make sure your entry points are secure:
Ideally, make sure you have a Web Application Firewall in front of web facing services. This comes at a cost though, which some of us like to avoid if possible 😜
Whether you do or not, review the ciphers available and remove them from least secure upwards until you work out the minimal and most secure configuration. For Windows IIS, remove any ciphers you don’t need as described here: Secure your Web Server
Do you have an RD Gateway / Web server? The same applies with IIS, and make sure you don’t allow ‘Domain Users’ in the access policies. Refine this to a group that contains only the users that need access. I’m sure you can get by without ‘Administrator’ being available to brute force hack from outside.
Geo-blocking – use it! Most firewalls have some geo capability these days. If your users are in New Zealand and Australia, restricting access to those regions only at the firewall provides a huge security benefit.
Right what was I actually posting about? Ah yes – Azure DNS is easy to set up and costs literally a dollar and cents per month. You don’t transfer your records to Azure though as they are not (and may never be) a registrar. But no bother, once you’ve set up your records in Azure, you simply change the ‘nameserver’ configuration with your existing provider. I use Free Parking in New Zealand, a great low(ish) cost no-frills provider. About $45/year for a domain name and of course they have DNS management, but I’d rather do it in Azure so I after I configured the zone, I copied the four nameserver entries on the right:
…then changed my nameservers from Free Parking to Azure – there will be somewhere you can do this in your providers portal, or just log a request for them to do it:
Done! Once the update is complete you are now serving and managing your DNS entries from the familiar Azure portal.
Here is my Azure DNS zone cost… as you can see – cheap as cheeps mate! Yes I’ll pay a whopping NZD$1.58/month or $18.96/year. That’s only 5.42857 Steak ‘n’ Cheese from Mrs. Miggins pie shop! 🥧
The console is intuitive and I love the fact I can manage my DNS easily and securely from within my Azure tenant. I can also get some metrics about the DNS usage that I couldn’t get before:
Hi! hope you are well… I used this script yesterday and thought “I must post about that!” So here it is… 😃
This can be a good way to deal with an employee leaving, or if you end up with a dual-mailbox scenario, although Microsoft have made that much less likely since adding the Hybrid Exchange option to AD Connect a while back…
Right, enough kafuffle – onward! 👀
Suppose a user leaves – your off-boarding procedure may include something like:
Exporting the mailbox to a PST file which is stored somewhere
Converting the mailbox to a Shared Mailbox and giving a another user access to it
But what if there is an incoming user that would benefit from having the existing data in their newly created mailbox? It might be a training coordinator or HR administrator’s mailbox that has a heap of relevant information that the incoming user can search through for information. This is much easier to use when it is in your main mailbox, rather than an attached shared mailbox (or worse a PST file!).
This script will connect to Exchange Online and prompt you with a list of soft-deleted mailboxes (these are the ones that have been deleted but remain for 30 days before being permanently deleted). Select the recently deleted mailbox and click OK. Now you are prompted with a list of all active mailboxes. Select the mailbox to merge the data into and click OK. Now the soft-deleted mailbox will be merged into the active mailbox. Superb!
Make sure you have run ‘Install-Module ExchangeOnlineManagement‘ so you can run the script successfully.
Here’s the code:
# set variables to null
$source = $null
$target = $null
#Connect to EOL...
Try {
Connect-ExchangeOnline -ShowBanner:$false
}
Catch {
Write-Warning "Failed to connect to Exchange Online!"
Read-Host "Press a key to exit..."
Exit
}
# prompt to select source mailbox
$source = Get-Mailbox -SoftDeletedMailbox -ResultSize unlimited | Select DisplayName,ExchangeGuid,PrimarySmtpAddress,ArchiveStatus,DistinguishedName | Out-GridView -Title "Select source soft-deleted mailbox and click OK..." -PassThru
if ($source -eq $null) {
Write-Warning "No source selected. Press any key to exit..."
Read-Host
Exit
}
# prompt to select target mailbox
$target = Get-Mailbox -ResultSize unlimited | Select Name,PrimarySmtpAddress,DistinguishedName | Out-GridView -Title "Select target mailbox and click OK..." -PassThru
if ($target -eq $null) {
Write-Warning "No target selected. Press any key to exit..."
Read-Host
Exit
}
# start merging source into target
Try {
New-MailboxRestoreRequest -SourceMailbox $source.DistinguishedName -TargetMailbox $target.PrimarySmtpAddress -AllowLegacyDNMismatch
}
Catch {
$errormsg = $Error[0].Exception.Message
Write-Warning "An error occured. $errormsg. Press any key to exit..."
Read-Host
Exit
}
As usual please let me know if you have any trouble or have a scenario that warrants a modified solution and I’ll try to help.
I thought I would share some PowerShell I used during a recent cross-tenant migration. Firstly, the Microsoft documentation is really good and got the journey off to a good start:
PLEASE NOTE: Following is some fairly raw PowerShell. You can’t just press Go! You’ll need to understand and update the bits required and run accordingly! Hopefully everything that requires updating is in bold-italics. If not please punish me via comment! =)
That said… let’s continue 😃:
Firstly, let’s create a mail-enabled security group in the source tenant Exchange Admin console, ‘zzMigUsers@sourcedomain.com’ should do the trick! ..then add the users (mailboxes) you will migrate (the migration endpoint is scoped to a group, anyone not in the group will fail to migrate).
Now load PowerShell ISE and paste all of the below code bits in, then save it for later. First, we need our commands to be able to switch quickly between tenants (disconnect before connecting each time):
Now we can connect to the target tenant and create the Org Relationship and Migration Endpoint – you’ll need the ‘sourcedomain‘, app ID and secret to paste in here where the bold italics are (follow the MS article to set up the App Registration and Enterprise App, easy as!): Target tenant app setup.
# connect to target tenant here
# Enable customization if tenant is dehydrated
$dehy = Get-OrganizationConfig | fl isdehydrated
if ($dehy -eq $true) {Enable-OrganizationCustomization}
# Create Migration Endpoint in target tenant
$AppId = "paste the app id here"
$Credential = New-Object -TypeName System.Management.Automation.PSCredential -ArgumentList $AppId, (ConvertTo-SecureString -String "paste the app secret here" -AsPlainText -Force)
New-MigrationEndpoint -RemoteServer outlook.office.com -RemoteTenant "sourcedomain.onmicrosoft.com" -Credentials $Credential -ExchangeRemoteMove:$true -Name "SourceDomainMigEndpoint" -ApplicationId $AppId
# Create Org Relationship in target tenant
$sourceTenantId="paste source tenant id here"
$orgrels=Get-OrganizationRelationship
$existingOrgRel = $orgrels | ?{$_.DomainNames -like $sourceTenantId}
If ($null -ne $existingOrgRel)
{
Set-OrganizationRelationship $existingOrgRel.Name -Enabled:$true -MailboxMoveEnabled:$true -MailboxMoveCapability Inbound
}
If ($null -eq $existingOrgRel)
{
New-OrganizationRelationship "SourceDomainOrgRel" -Enabled:$true -MailboxMoveEnabled:$true -MailboxMoveCapability Inbound -DomainNames $sourceTenantId
}
Let’s use our disconnect / connect commands above to disconnect from the target tenant and connect to the source tenant to set up the other side:
# connect to source tenant here
# Configure OrgRel in source tenant
$targetTenantId="insert the target tenant ID here"
$appId="insert the app id here"
$scope="zzMigUsers@sourcedomain.com"
$orgrels=Get-OrganizationRelationship
$existingOrgRel = $orgrels | ?{$_.DomainNames -like $targetTenantId}
If ($null -ne $existingOrgRel)
{
Set-OrganizationRelationship $existingOrgRel.Name -Enabled:$true -MailboxMoveEnabled:$true -MailboxMoveCapability RemoteOutbound -OAuthApplicationId $appId -MailboxMovePublishedScopes $scope
}
If ($null -eq $existingOrgRel)
{
New-OrganizationRelationship "targetdomainOrgRel" -Enabled:$true -MailboxMoveEnabled:$true -MailboxMoveCapability RemoteOutbound -DomainNames $targetTenantId -OAuthApplicationId $appId -MailboxMovePublishedScopes $scope
}
Hopefully you got to this point without issue – if not let me know in the comments section and I’ll try to help! Next, we should be getting a successful test of what we have set up. Run this command since you are still connected to the source tenant – you should see the expected values returned:
Now, disconnect and connect to the target tenant again, and run this:
# Confirm OrgRel and MigEndpoint in Target tenant
Get-MigrationEndpoint
Get-OrganizationRelationship | fl name, DomainNames, MailboxMoveEnabled, MailboxMoveCapability
Test-MigrationServerAvailability -Endpoint "SourceDomainMigEndpoint"
Awesome, green lights!? The tenants are configured. Now disconnect / connect to the source tenant, we’ll need to get the following details out into a CSV file so we can create the MailUser objects in the target tenant (yes I know we could have done that earlier =)). Limit this if required based on the group we created earlier, scoped to your migration users.
Disconnect / connect to the target tenant again. We want to test one user first, so I used these commands (replace bold italic with values for one user from the CSV output):
# create MailUsers in target tenant - ONE AT A TIME HERE OR BELOW FOR BATCHES
$originalalias = 'sales'
$newalias = 'sourcedomain.sales'
$userdisplayname = 'sourcedomain Sales Department'
$exchguid = '80ddefd4-26cb-7621-c497-g6c044b04dd9'
$archguid = '70edegc5-36f8-8639-b287-f6g035408ec2'
$x500address = 'x500:/o=ExchangeLabs/ou=Exchange Administrative Group (FYDIBOHF23SPDLT)/cn=Recipients/cn=ce04185c5ff841f885d660e46e6c8bc5-sourcedomain SAL'
$usersourceaddress = $originalalias + "@sourcedomain.onmicrosoft.com"
$usertargetaddress = $newalias + "@targetdomain.com"
$usertargettenantaddress = $newalias + "@targetdomain.onmicrosoft.com"
New-MailUser -Name $newalias -DisplayName $userdisplayname -ExternalEmailAddress $usersourceaddress -MicrosoftOnlineServicesID $usertargetaddress -Password (ConvertTo-SecureString -String 'Hellosourcedomain2099' -AsPlainText -Force)
Set-MailUser $usertargetaddress -ExchangeGuid $exchguid -ArchiveGuid $archguid -PrimarySmtpAddress $usertargetaddress
Set-MailUser $usertargetaddress -EmailAddresses @{Add="$x500address","$usertargettenantaddress"}
Set-MailUser $usertargetaddress -EmailAddresses @{Remove="smtp:$usersourceaddress"}
NOTE: Recently I found this command was not working and I had to use a one-liner copy of the batch1-users.csv mentioned below. After the test user is successfully migrated, reference your original file with all the mailboxes in it. Just in case you get this issue in your scenario 😃
Okay! When this is working and we are confident all is well, copy the sourceusers.csv to a new file named batch1-detail.csv. Open the file and add three columns before DisplayName as below. ‘originalalias’ (the bit before the UPN), ‘newalias’ (set as needed for the target tenant) and ‘newupn’ (new UPN in the target tenant):
Assuming the file is called batch1-detail.csv, create another CSV called batch1-users.csv. This is the input for the migration batch command (very simple and worked well for me), just copy the target UPNs from batch1-detail.csv into the EmailAddress column and populate the other columns as below:
Now you can use this code to create the MailUser objects with the batch1-detail.csv file:
NOTE: At this point you are migrating the mailboxes because of the “-Autostart” switch. Users in the batch will have ‘Syncing’ status. When ready to complete they will be ‘Synced’.
You can run this command to check the status of a batch:
# Get status of a migration batch:
Get-MigrationBatch $batchname | fl
Or this to check status of a single user:
# Get details for a single mailbox
Get-MigrationUser 'insertupnhere' | Get-MigrationUserStatistics | fl
Or this to get the status of all migration users, updated every 5 minutes:
# output progress of all users every 5 mins
$x = 0
Do{
$x = $x + 1
Get-MigrationUser | Get-MigrationUserStatistics
Start-Sleep -Seconds 300
}
Until ( $x -eq 480 )
When the mailboxes in a batch have ‘Synced’ status, go ahead and complete the batch using this command:
# Complete a Synced batch
Complete-MigrationBatch $batchname -Confirm:$false
This can take a bit of time so have a beer or cup of tea then come back – done! Once completed the mailuser you created in the target tenant will become a mailbox, and the mailbox in the source tenant will become a mailuser with a forwarding address set to the users target onmicrosoft.com address 👍
Some other commands I used are:
# Restart a failed batch
## Start-MigrationBatch $batchname
# remove a completed batch
## Remove-MigrationBatch $batchname -Confirm:$false
# Remove failed users from batches due to failure
## Get-MigrationUser | ? { $_.status -eq 'failed' } | Remove-MigrationUser -Confirm:$false
This migration went really smoothly… once the user was migrated, the forwarding was set correctly in the source tenant, and permissions were intact for access to Shared Mailboxes etc. Now we are planning to tidy up and move additional email aliases, the accepted domains and MX records across to the new tenant… I’m impressed with how relatively easy this was!
Until next time… ‘Hei kone ra’ from New Zealand Aotearoa! 🍻😜
Greetings! 👀 After a comment on my initial post asking for user email addresses in the output, I ended up getting a bit confused for 4 hours while trying to achieve the goal (it was a Friday night so several beers were involved) 🍻 !!
When I started seeing the dreaded pages of red errors in my results I soon realised I was not thinking that objects other than users can be members of a group. Of course! So I need to cater for computers, nested groups and users with no email address.
The result is below and from initial testing it seems to work well. Key points:
As with the original script, the CSV will output AD groups and members.
Where a group has no members, the group name is output with ‘No Members’ in the members column (and also now in the EmailAddress column).
The CSV has an ‘EmailAddress’ column added:
Where the member is a user and has an email address, the address is displayed.
Where the member is a user and does not have an address, ‘No Email Address’ is displayed.
Where the member is a computer, ‘Computer Object’ is displayed.
Where the member is a group, ‘Nested Group’ is displayed.
Voilà mes amis ! Code is below – as usual please comment if it helped or you made it better or it didn’t work for you ✌😃🤞. Thanks for coming, until nek tiya !
Cloud services have improved our lives and made our jobs easier – BUT they have also given hackers a worldwide platform of unlimited power with which to attack us… very sad but very true!
😲 😲 😲 😲 😲 😲 😲 😲
This makes it evenmore critical to secure our external-facing services as much as we can.
Hopefully you have a WAF in front of your web server, but if you are like me and have a small site that does not justify the associated costs of advanced protection, here are some basic steps to take on your Windows Server. Note that ‘Strict High Transport Security’ (step 4) is available from IIS 10 in 2019 Server.
NOTE: When you run the test, remember to check the box if you do not want the result to be displayed on the page…
👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍
OK – to business! There are four initial steps we can perform on a personal or small business web server that is exposed to the interwebs… resulting in an A+ score from an SSL test. Note that this is ideally run from the web server with traffic allowed inbound on port 80 and 443 (you can redirect 80 to 443, but 80 does need to be open for LetsEncrypt to work without manual intervention – AFAIK 😃).
Apply a Lets Encrypt certificate.
a) they are free!
b) they have a great reputation.
c) they are so easy to install it is not even funny!
Go to win-acme.com, click downloads and grab the latest version. Extract to c:\program files\win-acme. Run wacs.exe and follow the prompts… you can manually specify hostname, additional SAN names if required, or generate a wildcard. Use the default in memory validation; this creates a virtual directory (which is in memory and removed afterwards) in IIS for LetsEncrypt to connect to to verify the request. The app then automatically creates and applies a certificate that is valid for 3 months, then sets up a scheduled task to automatically renew the cert before expiry. Boom! 😁😁
Open PowerShell ISE (run as admin), paste and run the code below to confirm TLS 1.0 and 1.1 are disabled and TLS 1.2 is enabled for the system and .NET:
4. Enable HTTP Strict Transport Security… (Windows 2019+)
a) In IIS Manager, open the HTTP Response Headers section.
b) Click Add.
c) In the Name field, add “Strict-Transport-Security“.
d) In the Value field, add “max-age=31536000” (this corresponds to a one year period validity).
d) Click OK.
Oh and don’t forget to redirect port 80 to 443… even though the above step effectively forces the browser to use HTTPS, there is no harm doing it with the URL Rewrite feature (I use this so that I can also block connections to my WordPress admin page).
Here is the rule I use at the web site level to redirect any HTTP request to HTTPS:
EDIT 02/07/22: totes forgot about step 5 – CAA (Certificate Authority Authorization) records which I enabled on this site. I use namecheap.com who offer these records free with domain name registration. Most DNS providers should support CAA records, if not, well… switch providers I say! 😎 Here is a screenshot of my records:
Superbulous! Now run the test, grab a cup of tea and a biscuit and pat yourself on the back for being so awesome!! 😎 😎 😎
If you aren’t licensed for and using Conditional Access policies, please do not disable the security defaults feature just because something isn’t working (e.g. scan to email). Microsoft introduced the defaults for a very good reason – they realised that tenants without Azure AD Premium P1 licensing and correctly configured CA policies were wide open to Phishing and Password Spray attacks, via connections to Exchange Online using basic authentication protocols such as POP, IMAP and SMTP.
Connections using basic authentication do not support and therefore bypass MFA. If you disable this setting you are effectively turning off many security features.
Let’s find a solution to these problems and leave our tenant protected ‘by default’.
Here’s how to allow certain things (e.g. Teams meeting room devices and printers) while leaving our tenant secure:
1. Add any external IPs of company locations to Trusted IPs under MFA settings. In most cases you would do this for all company owned office locations.
Now your users and devices will be able to connect without MFA requirement from trusted offices, and you can set up Scan to Email functions to use the account you created.
Righto – time have a cup of tea and reward yourself for not chopping off a leg to fix an itch! 🤣🤣
recent comms…