Jump to content


Adminoob

Established Members
  • Posts

    26
  • Joined

  • Last visited

Posts posted by Adminoob

  1. I have an SCCM 2012 R2 distribution point that was accidentally reinstalled (the DP role removed, then re-added - no content was deleted) the packages and content library are all still on the DP, is there any way to get the reinstalled DP to recognize and associate with the now old content? it seems like it should just be a matter of recreating the content mapping, but I have no Idea how to do that. Please help!

  2. In SCCM 2007 I could apply client patches during the config manager client install task of an OSD like thus: PATCH="%_SMSTSMDataPath%\OSD\AH2000A6\hotfix\KB2994331\Client\x64\Configmgr2012ac-r2-kb2994331-x64.msp"

     

    Can it be done the same way in 2012? the hotfixes no longer reside in the client install folder, so the .msp's would have to be copied there manually and the client package updated. I've found at least one post where this was being done, but I cannot find any official MS KB saying this is supported or not: http://www.deploymentresearch.com/Research/tabid/62/EntryId/189/Installing-CU3-for-ConfigMgr-2012-R2.aspx

  3. when I create a package and then distribute it to a DP it first copies a temp file to the root of D:\SMS_DP$, Then it transfers those files to D:\SCCMContentLib, and creates the necessary files within the 3 subfolders, finally, and this only happens if the box is checked to copy to the package to a share, a folder with the extracted files is copied to D:\SMSPKGD$\(PACKAGE_ID). It also creates a .tar file in smssig.

     

    When I distribute an application pretty much the same thing happens minus the extracted file in SMSPKGD since applications cannot be run from the DP, which I find annoying and odd in itself, but that is another matter.

  4. Hello!


    Ok, no one, not even Microsoft engineers can seem to provide a clear and definitive answer on this topic, everyone's understanding of the .PCK file with relation to SCCM 2012 is clear as mud.


    I understand what .PCK files are...in the SCCM 2007 world, and how they work, all that jazz, what I don't understand is how/when/why they would be used by SCCM 2012...it seems that sccm 2012 would only need these .pck files during a reassignment of a shared DP in a migration scenario...even that surmise is sketchy since I've watched very closely the DP content conversion process and at no time did the server access the .PCK files to populate the single instance store.


    I am going to do some more testing on my own, my ultimate goal is to permanent delete the contents (not the folder itself) of the SMSPKG folder, essentially all the .PCK and .DL$ files to save space on my server.


    If any soul out there knows..not guesses...the purpose of the .PCK file in 2012, let me know please, don't just copy/paste other people's posts as an answer, explain in clear terms each step of the process that 2012 would utilize a .PCK, or if the answer really is that 2012 does not use .PCK files at all, let me know.


    I have a single primary site with 16 secondary sites in my SCCM 2007 environment, I'm migrating to a single primary with 16 distribution points...so even if .PCK's are used to populate secondary sites, I should not need them since I won't have any secondary sites once all of my DP's are migrated...yeah?


    Thanks!


  5. Hello,

    If my current SCCM 2007 R2 is installed on a 32bit Windows Server 2003, and I plan on migrating to SCCM 2012 on Server 2008 R2 x64, will there be any problems with the migration process going from a 32bit to 64bit environment?

    The reason I ask is that I read in a forum post somewhere that an admin in a similar situation did the following steps: performed a full backup of SCCM 2007 running on W2K3 32bit, reinstalled SCCM 2007 on a different server running W2K8 x64 and restored from that backup he made essentially creating a mirrored install but on a x64 OS instead, they then installed SCCM 2012 on the new production server (also W2K8 x64) and finally went into the standard migration steps.

    This seemed rather convoluted to me, and I found no other posts that claimed there was any need to first mirror your SCCM 2007 from a 32 to a 64bit environment before attempting a full migration to SCCM 2012.

    Thanks in advance,

    T

  6. Hi,

    The easiest way would be to remove the capture step from the TS and capture the image using the Capture Media instead, then you can install the software and make the changes you need..

    Regards,

    Jörgen

     

    Thanks Jorgen (a bit late, sorry!)

     

    I did manage to simply add an unattend.xml to the B&C TS, and this worked fine, but I found out the hard way that you MUST remove that file prior to running the capture process, otherwise it just continues to boot into audit mode (the sysprep process should remove that file, but it doesn't , not sure why that is.)

     

    T

  7. ...To apply the patch to the existing workstations I use the command in an advertisement CCMSETUP.EXE /noservice mp:sccm-dr.hc.ad SMSSITECODE=DRA FSP=sccm-dr.hc.ad SMSSLP=sccm-dr.hc.ad PATCH="sccm2007ac-sp2-kb977384-x86-enu.msp"

     

    Thanks.

     

    Sorry for the dumb question: Do those existing workstations have the CCM client installed already? And how did you figure out that you must disable those specific services in order to get the patch to install?

  8. Last week I finished pushing out the KB977384 hotfix ahead of SCCM 2007 R3 upgrade. Today I finished installing the R3 on the server. I am trying to configure the OEM Prestage but need some help. I have my task sequence which works for Bare metal builds and wanted to either modify the existing TS to include the OEM Prestage or create another TS. If someone has examples or a link I could follow.

     

    Thanks.

    Hi there, I'm in the same boat, did you ever resolve this issue? I tried following Microsoft's method here: http://support.microsoft.com/kb/977384 but on my test deployment machine, both the ConfigMgr client and the hotfix failed to install, the error on the deployment machine referenced a problem copying the hotfix to a temp location. I'm going to attach a portion of my client.msi.txt file from the destination pc that failed. Also, here is the exact commands I'm using in my TS (it is the Microsoft suggested commands, plus the standard setting for cache)

    PATCH="%_SMSTSMDataPath%\OSD\CEC0030\i386\hotfix\KB977384\SCCM2007AC-SP2-KB977384-x86-enu.msp" SMSCACHESIZE=8000

     

    Thanks,

     

    T

  9. This is from another forum:

    Run this script instead of step 3 (it's a vbs script). It will set the BackupFile variable to weatherer you type in the inputbox.

     

    ----------start script -------------

     

    set env = CreateObject("Microsoft.SMS.TSEnvironment")

    Set ProgressUI = CreateObject("Microsoft.SMS.TsProgressUI")

     

    ProgressUI.CloseProgressDialog

    BackupFile = InputBox("Name of wim file")

     

    env("BackupFile") = BackupFile

     

    ----------end script -------------

     

    I would like to mention, that the beauty of adding images to the same WIM, is the fact that it will reuse files already in the wim. this means you could have 10 XP images in the same WIM, and the size of the wim will only grow with the changes you make. I kind of explain this a little bit here http://blog.coretech.dk/mip/capture-none-syspreped-image-using-sccm-running-imagex-from-a-ts/'>http://blog.coretech.dk/mip/capture-none-syspreped-image-using-sccm-running-imagex-from-a-ts/

     

     

    Michael Petersen http://blog.coretech.dk/mip/

  10. Hello, I have added a TS directly under "Setup windows and ConfigMgr" as a command line; it did not work, and perhaps it is the way I configured it, I simply put in this for the command line: wsname.exe /n:$SERIALNUM

    and left the rest defaults except for 'Run in' which was the UNC to the directory where wsname.exe was kept (it was a hidden share, should it be a standard share?) And I did not run the command as any particular user, should I run it as the administrator account?

     

    I'm trying a different route now, I created a package for the wsname.exe command and am using that instead of a 'Run command line' TS. Well see how that works out, or doesn't, as the case may be.

     

    Any suggestions or examples from people who have successfully used wsname.exe is very appreciated.

     

    Thanks!

     

    Ah the packaged wsname.exe worked like a charm!

  11. I have several computers that have XP Pro 32bit on them, but the machine architecture is actually x64; I have successfully captured wim images from these machines using either x64 or x86 boot images, should I ALWAYS use x64 boot images to capture on X64 arch. (even when the OS I'm capturing is 32bit?)

  12. Hello!

     

    I have a capture only task sequence that works beautifully, and is very simple, but there is one thing I would like to change. Here are the components of my SCCM capture-only TS:

     

    1. Use Toolkit Package (MDT Integration)

    2. Set Backup Location (TS variable: ComputerBackupLocation )

    3. Set Backup Filename (TS variable: BackupFile )

    4. Create Image (MDT create image script: cscript "%scriptroot%\ZTIBackup.wsf" )

     

    I don't think it matters, but this is capturing an XP image.

     

    I would like to eliminate #3 and prompt for an image name (not a computer name, the .wim name)

     

    How can this be done? Via another TS variable perhaps (I'm reading through the MS documentation on that right now, but so far, nothing pops out that looks like it will prompt for the .wim name during the OSD)

     

    Thanks!

  13. This is a loaded question :)

     

    There are a lot of ways to do it. The way I use is to put a step in the task sequence to run an app called wsname.exe to rename the computer BEFORE it joins the domain. Once MDT and COnfigMGR 2012 work together then you can use MDT variables like OSDComputerName to query WMI for a SN# or use a task sequence variable.

     

    Like I said there are a ton of ways to do it. Your best bet is to rename the computer before joining the domain because it is less of a headache because once the machine is joined to the domain you have to worry about permissions, the task sequence failing because the object exists, and things like that.

     

    Hello, I have added a TS directly under "Setup windows and ConfigMgr" as a command line; it did not work, and perhaps it is the way I configured it, I simply put in this for the command line: wsname.exe /n:$SERIALNUM

    and left the rest defaults except for 'Run in' which was the UNC to the directory where wsname.exe was kept (it was a hidden share, should it be a standard share?) And I did not run the command as any particular user, should I run it as the administrator account?

     

    I'm trying a different route now, I created a package for the wsname.exe command and am using that instead of a 'Run command line' TS. Well see how that works out, or doesn't, as the case may be.

     

    Any suggestions or examples from people who have successfully used wsname.exe is very appreciated.

     

    Thanks!

  14. Alright, I had this working fine in MDT, but I am trying a similar approach in SCCM to no avail; I added a package to SCCM with the necessary Powershell script (tested and working) to rename the destination PC with the SMBios computer name (Lenovo computers use the system s/n, which we use to identify them in Active Directory/SCCM). Right before the rename TS I have a command line running the PS comand to set the execution policy to unrestricted. These two TS' I put right after the configure Windows TS, it didn't rename the pc. I moved it to other locations, and cannot get it to work regardless. I figured the trouble might be with UAC, but then every application package I have created thus far works fine during the software install phase, but I tested my script within Windows and found that it worked as expected.

     

    A couple points/questions:

     

    1. I do NOT want to prompt for a computer name during any stage of the OSD, this must be a fully automated process.

     

    2. Is VBScript a better option?

     

    3. Should I perhaps try creating an MDT TS in SCCM and then mimic the process that used to work when all I was running was MDT?

     

    4. Is there any better method to accomplish this?

     

    Thanks!

  15. Success!

     

    Alas, I have found a simple solution to my problem! At least a partial solution; as I stated originally, I really just wanted to mirror the same imaging methods I was used to in Novell Zenworks; setup a windows box with all the trimmings, snapshot it, upload it, sysprep it, snapshot again for final deployment via pxe/zenworks.

     

    I can achieve this functionality, and the solutions were so simple as to be embarrassing to admit.

     

    To capture any OS in a non-sysprepped state via pxe OSD task sequence:

     

    create a new custom ts

     

    first ts: 'Use Toolkit Pakage" - first integrate MDT 2010 into SCCM, then create a package for the MDT Deployment Toolkit

     

    second ts: "Set Backup Location" - set a task sequence variable with 'ComputerBackupLocation' and a value that points to your backup share location

    third ts: "Set Backup File Name" - yet another ts variable called "BackupFile" with a value that will be your image name 'myimage.wim'

    fourth ts: "Create Image" - here's the magic ts, it is a 'Run Command Line' ts with the command line of cscript "%scriptroot%\ZTIBackup.wsf

     

    Advertise this, make it available to pxe boot image and voila! It simply works! Tested it on a Windows 7 x64 and XP machine, no issues.

     

    Now to reverse the process, even simpler:

     

    2. To apply a non-sysprepped wim via pxe OSD task sequence:

     

    first ts: "Partition Disk 0" - change the format to quick, otherwise left everything default here

    second ts: "Apply Operating System Image" - point this to your previously captured non-sysprepped image

    get a cup of coffee

     

    Again, tested on W7 and XP, works fine. I understand that this method is only valid if you are capturing and reapplying to the same model due to HAL and Mass Storage driver issues, but this will get me started and I can work on creating my 'thin-image' and custom application deployment ts's at a later date.

     

    Thanks all,

     

    Sincerely,

     

    T

  16. Well, this is just my own two cents from my perspective... We try to keep the base image fairly light, and install additional apps on-the-fly as we deploy the OS to the actual end user system. And drivers are also installed as you deploy the image to the end user system - that's part of the beauty of SCCM - no more stuffing drivers into an image.

     

    Now, with that said, you can certainly install desired apps, customize the shoes off the system and do a ton of other things (to give you some scale, my OS customization VBScript is over 1000 lines of code). All without having to interact manually with the thing. What you do is create a build and capture task sequence that lays down the base OEM operating system. Then you have this TS install apps, drivers, scripts, what-have-you (this is where you do all those things you used to do manually when you were in the world of Ghost). Once that's done, the process automatically captures the image and you're up and rolling.

     

    It'll take you awhile to set up a full blow build and capture TS to do everything you want, but it'll be well worth it as you move forward and have to include all the updates and changes for your apps. You can check out this page if you want more steps on the TS: http://myitforum.com/cs2/blogs/rbennett806/pages/using-sccm-to-capture-and-deploy-windows-7.aspx

     

    And if you desire to customize something, but don't know how to script it, if Google can't help you, place a post (try to keep to one thing at a time if you can - it makes it easier for other to help answer with the desired script code).

     

    Thank you for your reply. That is good advice of course, and if I were more than a team of one and didn't have a deadline quickly approaching to revamp all of my imaging methods, I would certainly approach it from that angle without a doubt.

     

    However, I have apps like Adobe educational suite (12-15 gb) in one image, plus at least 30-40 additional applications of varying sizes and types, depending on which of the 7 different image classes we are talking about. Even if I were to script all of these via vbs and ts, it would take the build and capture process HOURS to complete, I suppose that is true of creating the base build as I do currently... I may just give up and use Imagex/MDT to do this.

     

    In the Novell Zenworks days of my recent past, I would:

    1. Install/customize Windows

    2. Snapshot an image of that gold OS to the server via Zenworks

    3. Sysprep and seal that machine

    4. Upload that image to the server as well

    5. Deploy as necessary to hundreds of computers via Zenworks over PXE.

     

    What took me weeks to accomplish the old way, will now, if I follow the thin-client paradigm, take me months, that is, until I have recreated these monolithic images of ours using thousands of lines of code. Care to share an example of your vbs code so I can get a running start?

  17. Hello,

     

    I have a Windows 7 x64 machine that is updated/customized, but NOT domain joined; your method here is fine for a domain joined machine, but as you pointed out, it isn't a good idea to sysprep and re-deploy a previously domain-joined machine because of the various changes that occur. How can I modify your method here to sysprep/capture my W7 image?

     

    It simply isn't feasible to do a build and capture ts, while that is fine if I want a vanilla OS with no apps or updates, the fact is, our organization has WAY too many apps and customizations to include as either custom ts software installs or to be somehow delivered via GP/AD. It is much faster to build a gold image (thick image) with everything on it, then prep it and capture an image for later deployment. I am NOT going to make custom ts software installs for things like Adobe CS4 and a dozen other HUGE apps, the imaging time would take hours for the base image and all of the software updates to come down.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.