Apexes

+1 for phased approach, building numbers over time

Found this in your time stamped smsts log "Failed to run the action: Apply Operating System. The hash value is not correct. (Error: 80091007; Source: Windows)" Can commonly mean a mis-match in content version of that package. Try re-distributing the OS image, and once it's showing as successfully sent to DP try to image again That said, you also have another error when the machine is trying to save the image onto the HDD: "Failed to save the current environment block. This is usually caused by a problem with the program. Please check the Microsoft Knowledge Base to determine if this is a known issue or contact Microsoft Support Services for further assistance. The parameter is incorrect. (Error: 80070057; Source: Windows)" This can sometimes mean a fault with the machine's HDD, or it could be a raid controller, uninitialized SSD etc. In this case i'd suggest trying a different machine to image on.

So it's not actually running as a user at all then. Has the content been downloaded into the cache, and can you run the exact command line you're using in SCCM from a CMD prompt in the programs cache directory? Error 1619 you're getting Microsoft lists as the below ERROR_INSTALL_PACKAGE_ 1619 This installation package could not be OPEN_FAILED opened. Verify that the package exists and that you can access it, or contact the application vendor to verify that this is a valid Windows Installer package.

Does anyone have experience swapping http to https with PKI certs in SCCM 2012. Aware of the process and requirements, but interested in any effects this may have on clients, or if anyone has first hand experience with it

+1 for package approach

If you have no maintenance window set on the collection you're trying to run the program on, then if you deploy a required deployment to it, and run a machine policy evaluation on that collection, i don't see why it wouldn't start within 1 - 2 minutes of running that collection cycle.

Oh i thought you meant you'd set it to run as user. It'll be in the settings for the deployment type under the user experience tab, nmake sure you've got installation behavior set as install for system

Check out a machine that's not reporting correctly in the console, and look at the bottom for the last hardware scan date. Also check in your client settings that apply to these machines - how often do you have hardware inventory set to run ?

Not sure why you'd want to install as a standard user, you'll need administration rights for this sort of install. To suppress the reboot put /norestart on your msiexec command line.

Yeah it's annoying. Other option is you could put it into a task sequence, and put in a TS step to check if .net is installed - if it's not, then install it, but if it detects it is, it'll skip the step and run to the main application install. Kind of a similar way of doing dependancies

yep - for troubleshooting see if it can install as a package first off, use the same command line. out of curiosity - what command line are you using? Application model can be picky with certain apps

I use VM's running server 2012 on VMWare - however we've got servers in all branches for business functions meaning i can utilize space on these boxes, some sites only have a 5mb wan link for smaller ones - 20 machines or less. to be cost effective, you're looking at a desktop to host a distribution point. However, a desktop opreating system, i.e. Windows 7 - won't support PXE or Multicast. PXE is required to boot from network and image. you need the windows deployment services role which is only part of server operating systems. see here for site server pre-requisites: https://technet.microsoft.com/en-in/library/gg682077.aspx#BKMK_SiteSystemRolePrereqs If you're looking at primarily imaging, you'll need a server OS, that's not to say that you can't install a server os on a desktop machine, obviously depends on your licensing setup with Microsoft. I've only got experience in using enterprise gear for servers, so all our site servers have HP DL380's running VMWare to host our site servers along with distribution points in those.

See here: http://mikesconfigmgrblog.blogspot.com/2014/07/microsoft-net-framework-45-not.html Possibly an issue with it being in application model. Try it as a package to test

I don't think it's officially supported, but as long as you can install RDC and run IIS then i don't see why it wouldn't work

i think you're slightly confusing it, enterprises do indeed image over the network, but specifically the LAN, not the WAN. if i was to image a machine over our WAN at a branch office, it would transfer in the region of around 18-20GB of data, our site link is only 10mb at that branch - this would cause the latency to go through the roof for users and affect business processes. Example of what i manage - a 10,000 client environment, with 20 branch offices. I design the image and task sequence on our primary site in our data center, and distribute this content out to all site distribution points which act as a local cache. Site boundaries are configured in SCCM to tell a client what server it retrieves it's content from. e.g. if my IP is 10.7.4.4 - i'd go to a specific site distribution point which is local to me. If i went to another office and got the IP of 10.8.3.3 - SCCM would tell me to swap my content location to the local server there, to avoid traffic going across the WAN. Having setup a distribution point you distribute your content once. When you've sent that content, you can then image every machine on your site if you like, and the traffic would be local only, no WAN transfer of content. Think of SCCM as a filtering down system. Content and Data starts at the top, then trickles down to child sites below it. Client computers below the child sites then connect in to them to get data on a local area network only. You can setup an SCCM distribution point on a Windows 7 OS and run it on a desktop with minimal specs, a distribution point is just a glorified file server that hosts content for clients to retrieve locally

I wouldn't recommend imaging over the WAN, bad idea. However, if you've got half decent connectivity at these sites, then get a distribution point setup on site and hook that into your current SCCM primary. You'll be able to image across the LAN from that DP rather than WAN on your primary. If you're looking at customizing images based on locations, then you'll want to integrate MDT into your SCCM task sequences. Check in the how to guides for this forum, there are some novice guides on building out your first image and then using task sequences to deploy them to sites.

Error code 8 is not enough space from what I've looked up, but that'd contradict your previous screenshot of the disk space on the drives, and the fact that you can distribute other content to the server. Sorry if it sounds like i'm teaching to suck eggs - but that server definitely has enough space on it's host drives when you log on to it? I'm at a loss as to where i'd even go now if i was having this problem, from your descriptions i've managed to fix similar issues to this by getting shot of the .ini and content files in the SCCMDataLib, but you've tried that and still nothing. Short of removing the DP role, and re-installing and sending content again - i'm not sure where you'd go from that.

Unsure if this would make any difference, but I've got my office uninstalls to run with /config - rather than /configure I'd also be inclined to turn that installation into a package, and then run the command line in the package inside that task sequence, rather that directly from a share. if carrying on with same method, try adding the full path before the uninstall.xml line. e.g. "\\spsccm01\source$\Applications\Office 2013\setup.exe" /config "\\spsccm01\source$\Applications\Office 2013\uninstall.xml" with the Exformatics - try adding /quiet instead of /q I know it sounds stupid, but that worked for me once with a task sequence on a specific installer, it complained of /q rather than /quiet :|

is the client recognizing the install in execmgr.log ? is the machine requesting policy correctly? when was the last policy request listed as showing int he console?

I have this problem until this day, application model in OSD is terrible - yet it deploys fine on a standard desktop I've taken the route of using packages only for OS deployments, since doing so all my builds have been working 100%

Be curious to see your clientlocation.log and locationservices.log - along with clientidmanagerstartup.log as Garth has mentioned, it already sounds boundary related

Odd - the only thing i can think that'd spawn a similar message would be if enable pull distribution point was set on the DP to look elsewhere for it's content source, but that being said other content works fine so i'm at a loss :| Maybe create a whole new package for the content you need and send that out to all DP's to use instead?

Check out ClientIDManagerStartup.log & PolicyEval.log - you may find some clues as to what's happening there. Also check ClientLocation.log and verify they're picking up a MP successfully

I've had similar before, try this, it may or may not work. 1. Remove the content from the DP in configmgr console, give it 10-15 minutes to reflect as showing as removed 2. Log onto your affected site DP and go into your SMS source location drive, and into SCCMContentLib 3. In this folder go into PkgLib, find any .ini file references to your package ID COR00129 and delete them (There may be some with hex codes appended too) 4. Go back up a folder into DataLib, again - delete any folder with the title of COR00129 in 5. Re-add the content in ConfigMgr console and monitor the distribution

Is your site Database replication working as expected? no degraded links under Monitoring > Database replication?