Adam Bise

Yes! Thank you. That works much better!

I want to create a collection using WQL to include computers with PC Bios SMBIOS BIOS Version "less than" 2.18.6. Problem is the version is a string and the less than operator uses sort order for strings so I wind up with machines with BIOS version 2.5.9 being GREATER than 2.18.6 I did find this post https://social.technet.microsoft.com/Forums/en-US/aacdcbda-6d14-4adb-b032-c9034d975403/querying-for-version-number-baseline?forum=configmanagergeneral My question is, would this be the best approach? Is there a simpler regex query I could use to achieve the same result? Thank you

I was having the same issue with an OSD deployment.OnOptionalExecutionRequests failed for program * : 0x87d01106 logged. All references to the TS were showing their content status was good. I found this post and deleted a bunch of update ADR deployments. Suddenly the issue disappeared. Strange thing was, this didn't seem to affect my app deployments, only my OSD TS, and even then only the windows 7 ones, and even further, only when installing from software center, PXE worked fine. I don't know how I would have fixed if not for your post, thank you!

Where can I read more about this issue or about the limitations on how many deployments a collection can have before running into this issue? Thanks!

I think this happens when you click start before running sysprep. Something about clicking start in Windows 10 seems to trigger store app installs. Also you can fix by creating a new local user, logging in as the new local user, delete local admin profile using control panel, log in as local admin (dont click start) delete new user profile and account, run capture media. If you are customizing an image prior to capture (and sysprep) as local administrator etc, try doing so without clicking start, or dump the local administrator profile mentioned above. Seems kind of retarded, perhaps I should have been using Audit Mode, although I read the same issue happens in that scenario as well. Anyway, you will save a little HDD space by running sysprep from a newly created profile.

I tried updating to CU1 and that didn't fix the issue. I was hoping it was because of KB3073015 Since then I have tried deleting the Win10 devices form the database and re-adding them via system discovery, but still no client push. I did get the following message in ccm.log after the machines were re-added: Execute query exec [sp_CP_CheckNewAssignedMachine] N'P01', 1 SMS_CLIENT_CONFIG_MANAGER 10/28/2015 9:48:28 AM 6456 (0x1938) I created a win10 collection to use for installing the client for now.

I had a similar issue and wound up fixing it by writing zeros to the first few MB of the disk and retrying. Something about the existing partitions it didn't like.

OS name and version field says Windows 10 Pro 10.0

Hoping I am missing something, because from what I can tell automatic client push is not working for Windows 10 clients. I have SCCM 2012 R2 SP1 The windows 10 clients are in the same IP address range as windows 7-8 clients, where client push is working automatically. Yet for Windows 10 clients, the job never shows up in ccm.log at all. Unless I trigger a manual install by right clicking and choosing install client, at which point it works great. So it's just the AUTOMATIC part that is broken. For these Windows 10 clients, they do show up in devices as no client, and properties show they are assigned to a site by the boundary group. Any suggestions?

It works that way because most environments have multiple sites and multiple DPs which may not all have access back to a single server \ share. If your boss ever said to provision a new site then you would have shot yourself in the foot because you would have to re-create all of your applications instead of simply rolling out a new DP and distributing content. The only reason I install over the network is for a couple programs that exceed our client cache size and I'm in a single site environment. But I use our DP for everything else because we may need to roll out new sites in the future, at which point I will only need to look at those large programs instead of remaking them all.

As long as "Install for system" is selected in the user experience tab of your DT Perhaps "Whether or not a user is logged in" is also needed, but I'm not sure about that. I always choose that option for any install that doesn't require running in the user context.

I have applications which install over the network and it works just fine. Only a script folder is deployed. Then the script runs things like: "\\server\share$\program\setup.exe" /silent for example. I use a hidden $ share with authenticated users allow read, which includes computers, for both share and folder security.

I had this issue when I deployed a TS referencing variables to a collection that was missing the variable.

We had the same issues on our site after upgrading to 2012 R2 SP1 The command in SQL Mgmt Studio worked: delete from dbo.CI_CurrentRuleDetail where Setting_CI_ID = 'XXX' Where XXX is the Configuration Item ID in the line above the error in objreplmgr.log.. objreplmgr.log: *** delete vCI_ConfigurationItems where CI_ID=XXX SMS_OBJECT_REPLICATION_MANAGER 7/30/2015 8:49:32 AM 3156 (0x0C54) *** [23000][547][Microsoft] [sql Server]The DELETE statement conflicted with the REFERENCE constraint "CI_CurrentRuleDetail_CIID_FK". The conflict occurred in database "CM_P01", table "dbo.CI_CurrentRuleDetail", column 'Setting_CI_ID'.Failed to delete Deployment Type ScopeId_E36C69C0-4828-4E8D-AB4A-A820638A16BF/DeploymentType_16b9837d-6ef5-4bd6-8055-3d633c901d06/3 Thank You!

I tried the above attempt but it didn't work, in fact only thing I could find on google was only for after the user had logged on and using MDT http://blog.danovich.com.au/2013/01/24/popup-message-during-sccm-task-sequence-in-full-os/ I did come up with something that does work but with a few caveats. Create a package with a script message.cmd: reg add "HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System" /v legalnoticecaption /t REG_SZ /d "Title Text" /f reg add "HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System" /v legalnoticetext /t REG_SZ /d "Message Text" /f reg add "HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\RunOnce" /v legalnoticecaption /t REG_SZ /d "reg add \"HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System\" /v legalnoticecaption /t REG_SZ /d \"\" /f" /f reg add "HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\RunOnce" /v legalnoticetext /t REG_SZ /d "reg add \"HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System\" /v legalnoticetext /t REG_SZ /d \"\" /f" /f Add a run command line as the last TS step referencing the package: cmd /c message.cmd Windows 7 machines that still use ctrl-alt-del policy will have to do that to see it, also only a local administrator can process RunOnce so the message will persist until an admin logs on. You may need to reboot again after running it also, not sure.

Assuming this is a software update deployment. Had a similar issue and it wound up being some updates were deployed twice by the ADR. Our ADRs create new software update groups, so these updates were members of more than one group deployed to the same machines and didn't install like you describe.

smsts.log Microsoft Deployment Toolkit version: 6.2.5019.0 InstallSoftware 7/20/2015 2:37:28 PM 1420 (0x058C) The task sequencer log is located at X:\WINDOWS\TEMP\SMSTSLog\SMSTS.LOG. For task sequence failures, please consult this log. InstallSoftware 7/20/2015 2:37:28 PM 1420 (0x058C) ZTI deployment failed, Return Code = -2147467259 0x80004005 InstallSoftware 7/20/2015 2:37:28 PM 1420 (0x058C) Command line returned 2147500037 InstallSoftware 7/20/2015 2:37:28 PM 1420 (0x058C) Process completed with exit code 2147500037 TSManager 7/20/2015 2:37:28 PM 1320 (0x0528) dism.log 2015-07-20 14:37:21, Error CBS Failed to load offline store from boot directory: '\\?\D:\' and windows directory: '\\?\D:\Windows\' [hrESULT = 0x80070002 - ERROR_FILE_NOT_FOUND] 2015-07-20 14:37:21, Error CBS Failed to initialize store parameters with boot drive: D:\ and windows directory: D:\Windows [hrESULT = 0x80070002 - ERROR_FILE_NOT_FOUND] 2015-07-20 14:37:21, Error DISM DISM Package Manager: PID=1624 Failed initializing the session - CDISMPackageManager::RefreshInstanceAndLock(hr:0x80070002) 2015-07-20 14:37:21, Error DISM DISM Package Manager: PID=1624 Failed doing internal initialization - CDISMPackageManager::Initialize(hr:0x80070002) 2015-07-20 14:37:21, Error DISM DISM OS Provider: PID=1624 Failed to open the key Microsoft\Windows NT\CurrentVersion. - CDISMOSServiceManager::DetermineBootDrive(hr:0x80070002) 2015-07-20 14:37:21, Error DISM DISM Driver Manager: PID=1624 Failed to getting the boot drive from the os services provider. - CDriverManager::Initialize(hr:0x80070002) 2015-07-20 14:37:21, Error DISM DISM Unattend Manager: PID=1624 d:\win7sp1_gdr\base\ntsetup\opktools\dism\providers\unattendprovider\dll\unattendmanager.cpp:390 - CUnattendManager::Apply(hr:0x80004005) 2015-07-20 14:37:21, Error DISM DISM Unattend Manager: PID=1624 d:\win7sp1_gdr\base\ntsetup\opktools\dism\providers\unattendprovider\dll\unattendmanager.cpp:654 - CUnattendManager::InternalExecuteCmdLine(hr:0x80004005) 2015-07-20 14:37:21, Error DISM DISM Unattend Manager: PID=1624 d:\win7sp1_gdr\base\ntsetup\opktools\dism\providers\unattendprovider\dll\unattendmanager.cpp:603 - CUnattendManager::ExecuteCmdLine(hr:0x80004005) 2015-07-20 14:37:21, Error DISM DISM.EXE: DISM Unattend Manager processed the command line but failed. HRESULT=80004005 Only thing I can suggest is try this since your errors are similar to this guy Basically he recreated the image. http://en.community.dell.com/techcenter/enterprise-client/f/4448/t/19544668 I had a similar issue that we have still not resolved.. On Dell OptiPlex 3020 machines the video driver could cause the system to hang during the unnattented portion of setup. IE Setup is installing drivers 50% and it would stay at 50% until the same error code you got. Are your machines doing this? We would restart the TS on the machines that failed and about 10% of the time it would error again, so we would just restart the TS again until it worked. I was going to remove the video driver from that package but we wound up not deploying anymore of that model. Seems like a driver issue, but I think those would be in the unattended setup logs. One thing I have done before and I could have sworn it was an error 0x4005 was simply delete the driver package and recreate it. Other times we had to omit some of the drivers because they caused issues.

Did you install the optional components? https://technet.microsoft.com/en-us/library/dn387582.aspx http://www.niallbrady.com/2013/10/09/how-can-i-manually-add-winpe-5-boot-images-to-system-center-2012-configuration-manager-sp1-cu3/

 For example add a couple scripts "message.cmd" and "message.vbs" from a package and run "message.cmd" in the TS. message.cmd: start "%~dp0message.vbs" message.vbs: x=msgbox("Message text." ,0, "Title Text") Include both in package source. Never done it myself, so you would need to test. I called the vbs from the cmd script in hopes that it would not wait for the script to complete and not timeout processing the step.

I have several applications are using script installer deployment types. The appenforce.log always says running command line in user context, but I think they are actually running in the system context because they all work when non admin users are logged on. BTW I normally choose "Install for system" and "whether or not a user is logged on" Also these are device deployments, not user.

Right now I am deploying client updates using the CU package that is automatically generated by the CU server update. I have been reading about patching clients during task sequences and client push using the patch property. My question is, since the existing clients will still need to be updated using the CU package, if I create a task sequence method to update the client, will it try to run it again on those computers when they receive the CU package deployment? My CU packages are deployed to a blanket all x86 and all x64 collections. Will I need to make a collection that queries the client version? Thanks!

It might be better to call Dell. On systems with M.2 or mSATA the drive order (not boot order) perhaps should be handled by boot priority or CMOS settings instead of hard sata port #. On my system M.2 uses SATA port 5 and 6. I don't know what order they are presented to the OS, or if it can be changed via CMOS. I only have the one M.2 drive but it is in interesting question. Also some systems allow you to disable individual SATA ports in CMOS which could temporarily shift the remaining drive to drive 0 from the OS perspective.

We have a Windows 7 task sequence that has been working great. Suddenly now we are intermittently getting 0xC004B100 when activating. Manual attempts to activate fail with that error about 50% of the time. Is this Microsoft having issues? The error means "The activation server determined that the computer could not be activated" This just started happening yesterday. We have only noticed the issue on our dev VM.

I wanted to know if there is a way to support a loose deployment of conflicting applications. Let's say I have app A, which conflicts with app B, and I want to publish them both as available for one collection. There is no requirement I can set on app B that says app A = not installed. (that I know of) I may look into creating a custom global condition but not sure how that works yet. I know I can always check for conflicting apps during an install script. I also know I can be more granular with my deployments, and deploy them to separate collections. But I am curious if anyone else has ran into this? If so, how did you work around it?

If any OSD task sequence is deployed to all unknown computers, I would most certainly advise using a PXE password. You will need to decide on what works best for you. How many new systems will you be bringing in vs how many existing systems will need to run required task sequences. You can always delete the unknown computer deployment, remove the PXE password, deploy your required task sequences to other collections, and then when you get a big load of new systems, re-enable the PXE password and re-deploy your TS to unknown computers.