This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(300.8, 95%, 39%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EBD%3C/text%3E%3C/svg%3E)
I am setting up an Endpoint Configuration Manager distribution point on my supplier's infrastructure so they can build PC's for us before they are shipped to site. All PC's need to be named automatically based on certain criteria during the task sequence. The idea is to have the entire process as automated and hands-off as possible.
Naming convention rules:
The biggest challenge I think I'm facing here is making sure I don't end up with duplicate names as a number of machines could be getting built simultaneously. I'd also like the hostname to be displayed on-screen during the built so techs can put the correct asset label on the PC. I would really like it if the PC label got printed automatically at the same time that the PC name was assigned but that might be something for the magic fairy dust.
--->Yes,we could set up a read only DC on our DP to reduce the query time.
--->As we mentioned we also could use CSV file to import multiple computers in our SCCM.
Here is a similar situation we may want to see it:
how-can-i-import-computers-into-sccm-2012-using-a-file
Note: this is non-official article just for your reference.
--->There are could be same name when we query at the same time so we need to add one lock variable to limit the action to wait for the previous action.The main logical may like this:
1.We need to prepare one txt file we prepare previously:
The content may like this:
"Locked = False ,Sequence Number = 0002"
2.We open our script to read this txt file every time and plus the computer name,the structure may like this,we could add more details on our own:
We define the status as two types:false(continue) and true(wait for 5 sec).
===start->read from txt file->if locked=false->{ get computer name,query AD, confirm, pass variable to TS}->Sequence Number ++1 and then save to file.
===start->read from txt file->if Locked = True then save file and ->{sleep 5 seconds go to start}
Have a nice day!
If the response is helpful, please click "Accept Answer" and upvote it.
Note: Please follow the steps in our documentation to enable e-mail notifications if you want to receive the related email notification for this thread.
May we know the current status of the question? If there is any other assistance we can provide, please feel free to let us know, we will do our best to help you.
For creating the objects in CM before starting OSD, this script would have some examples for you. It's how I create VMs, but check CM first to make sure the name it available before creating it and starting OSD.
https://docs.recastsoftware.com/ConfigMgr-Docs/Hyper-V-Automate-OSD-Test-Machines.html
Should be able to get some ideas from that. I'm a huge fan of only imaging known computers. If you can get the MAC Address for the machines ahead of time, you can onboard them and pre-provision everything the machines would need.
Thank you for posting in Microsoft Q&A forum.
We should be able to set available environment variables from the Power shell script, and then set the OSD Computer Name TS variable as an environment variable in the subsequent steps, and our desired target may need to add some lines to the rule so that we could verify whether the device name exists, if our device name exists in our AD, and our script also could continue to add the name plus until the computer name is the new name of our AD, just do it before applying the Windows settings.
There are many guides available for us to choose the computer name .It could be serial number or prompt for host name or any format that we would like to have. Here is an article for us to refer to:
https://msendpointmgr.com/2013/06/03/automate-computer-name-during-osd-in-configmgr-2012-with-powershell/
Note: This is non-official Microsoft article just for your reference.
Have a nice day!
If the response is helpful, please click "Accept Answer" and upvote it.
Note: Please follow the steps in our documentation to enable e-mail notifications if you want to receive the related email notification for this thread.
How would I make sure that I don't end up with duplicate names if the reference source is AD? If 6 PC's are being built simultaneously then I could potentially end up with 6 PC's with the same name if AD is the reference source, as the script will take the last name used and increment by one for the next PC. Therefore with all PC's running the same script and following the same rule, they will all query AD at around the same time for the last used name and get back the same result. Can I instead reference a local DB (such as a csv file) for the machine names? What I can then do is have the script take the next available name and remove it from the list, leaving only genuine available names. I can have this list refreshed each morning before a build run is done and the DP only uses this local DB as the reference source. A stand-alone script can run each morning on the DP to generate the new list that all TS's run from this DP will use.
Don't do unknown deployments perhaps.
Script the process that creates the machines in CM. Then you'll ensure you don't create duplicates. Then all machines are known as you image them and grab the names you created for them in CM.
Change your Computer Naming Standard. Have your PreFix, then an identifier for the build site (if needed), then use Serial Number
PPP-SS-######### (3 Prefix | SS for the SiteCode | # = Part of Serial)
Then you'll always have a unique name and it won't matter if you check CM or AD first.
The naming convention has only just been rather hotly debated about until we finally settled on what we have now. Serial number's were discussed but it didn't pass the majority vote. The standard we have now won't be changed under any circumstances so that's what I need to work with, hereby adding to the many challenges I already have on this project.
I like the idea of not doing unknown deployments though, I can create the object in AD first as well which would be made easier if I could have a DC set up on the DP. The local DP only then checks it's local version of AD data which can be set for a delayed sync, say overnight only. All machines built at the supplier's warehouse will be named in a given range unique to that distribution point. It's one way we're tracking the machines.
Apologies for the lack of response, things have been a bit hectic.
At the moment I'm thinking about a separate computer (just a low-end mini desktop would do) that has a CSV file I can update using scheduled tasks each morning at say 5am that contains the next available 100 or so names. The TS, which has been customised for and deployed to only this distribution point, will reference this CSV file for the name. Still working on where in the TS I'm going to do this, whether it's right at the start when it sets the name or just as a change computer name PS script later on but I can do any testing on the DP I have in my office as it's a similar configuration to what I'll have at the warehouse. The label printer is also connected to this box and I'll need to create a trigger that prints the label when the file is queried for the next available name.
Thank you very much for the update and If there is any other assistance we could provide, please feel free to let us know, we will do our best to help you!
Have a good day!