The new option in the console for Client diagnostics - Collect Log files, need to be able to trigger option from the client running a task sequence. The reason for this is when a task sequence logic encounters a error and you want to auto capture the logs, this client action will be key to automating the log captures. Also for Windows 10 Servicing where the machine runs on the Internet and capturing the logs works great over the CMG, but again just want to automate the log capture than rather manually capturing the logs.

We need the ability to deploy application groups with PowerShell.

Similar to Set-CMApplicationDeployment - but for groups.

When building collections (particularly query collections) it would be handy to be able to preview the devices/users in the collection as written before you commit. Today we have to edit the collection, update membership, refresh then confirm that the membership is correct. If you have a button inside of "Edit Query Statement" it would save a lot of hassle.

The provided Application Model Operation System Conditions for Windows 10 versions are limited in scope. Currently the Operator options for Windows 10 are two choices.

Windows 10 (Select This to get the two below)
All Windows 10 (ARM64)
All Windows 10 (64-bit)
All Windows 10 (32-bit)

We're using "Windows 10 Enterprise for Virtual Desktops" and we're not able to select this operating system from here.

Please add "Windows 10 Enterprise for Virtual Desktops" under windows 10 node or please provide a way to edit this list.

Thank you.

The essential need to run executable content from package before drive partitioning step (or any other case of unavailability of writable partiotion). "Data Access" package feature seems not working in OSD TS.

Machines with an EEC client certificate can connect to the DP to download the content and install the agent but the client never registers with the site.

once i suspected the ECC certs I was able to find thread on technet which confirms the same issue I was seeing
https://social.technet.microsoft.com/Forums/en-US/cc9ec0ff-5998-4225-9ce1-2c7b5fe5677d/sccm-and-ecc-certificates-not-supported?forum=ConfigMgrDeployment

When you need to troubleshoot desktop analytics you can use DesktopAnalyticsLogsCollector.ps1 but then you have collect the log files or have remote access to the clients.

Can you integrate the Desktop Analytics Logs Collector into the Client Diagnostics work being taken as part of TP 1912).

Add a vendor parameter to Get-CMSoftwareUpdate

This will help identify updates from specific vendors that have been provided through Third-party Software Update Catalogues.

In our environment SCCM Server installs client on discovered resource but they failed to install due to machine specific issue, could you prepare a tool which we can give it to floor support technician to run on machine which have SCCM Client issue or Installation failure, the tool should check for all prerequisites for client install also firewall issues like (WMI and File and Print sharing not allowed ) it should the issue with which it becomes easy to fix the underlying problem and get the sccm client successfully deployed

The new Model field on drivers is useful, but needs to be extended to other content types, like Packages & Applications, where Model specific items need to be installed (e.g. specific HP SoftPaqs, or video drivers that need to run from .EXE etc).

It would be nice to have ability to enable Tamper Protection in defender via SCCM antimalware policy

Currently, Get-CMPackage only returns “ordinary packages”, not software update packages or task sequence packages or other types. If you run the command in verbose mode, you can see the WQL query filters PackageType=0 and activity <>3.

Documentation for this cmdlet states “Get-CMPackage returns all packages” and this is wrong. It returns all packages where type = 0. Can we please create PackageType as a parameter so all packages can be managed with one cmdlet?

Take the OU path from 'Network Settings" and turn it into a variable e.g. _SMSMachineOUpath
I use dynamic variables to determine the OU based on location and device so my Domain OU path has OU=%devicetype%,OU=Location,OU=workstations,DC=Contoso,DC.com
so it would be nice to have the actual value in a variable that I can reference later say in a script to move existing object to the above created OU. I can see the full path in the SMSTS.log, but not in a variable. Also removing the LDAP:\ from the start could be handy in some situations I guess.

Actually, it is possible to create a collection with a cmdlet. But we cannot create it in a specific location. We need moving it after creating the collection. Why not just add a parameter to the New-CMDeviceCollection cmdlet?

I would like a way to monitor the ConfigMgr PXE Responder Service when an error occurs. For instance, when starting an osdeploy and the bootimage fail to load with error 0xC0000001, the smspxe.log simply stops its output, not generating any error. With the WDS-service, this error was easily detected in the eventlog.

Now that optimization of the WIM and removal of superseded updates is available as an option for offline servicing in the console, it would be great if this option could be added as a switch to the New-CMOperatingSystemImageUpdateSchedule cmdlet. I use this cmdlet to automate scheduling of updates in an ADG against all my WIMs so it would be great if I could shrink them down at the same time without needing to do this in the console.

The task sequence has a 32MB size limit for all policy. But for some reason they are actually limited to 16MB. But the Product Group is aware of the problem. I know they are working on fixing the 16MB limit.

This will help us troubleshoot and determine when the task sequence has reached this size.