benowen187

This is probably a straight forward "click this box" fix, but I'm pulling my hair out trying to find it... Basically we have WhatsUp! querying WMI for certain things (HDD space, CPU utilization etc.) which works perfectly under a local administrator account. However I want to change this to being a restricted user, so I've 1) Added the user with root permissions for "Remote Enable" and "Enable Account" in the WMI Control pannel (made sure that the permissions flow down) 2) Added the user to the DCOM "Remote Launch" and "Remote Activation" permissions for the machine. Now if I do a simple WMI query from the command prompt to get the OS name for example this works perfectly (if I modify the rights above, I get an access denied - so I know the rights are OK) However if I try and get information on HDD quotas etc.I get "No Instances are available" This is replicated in WhatsUP, as you can successfully connect to the machine via WMI but not "see" any instances. Anyone got any idea? - I'm assuming it’s a rights thing, but I’ll be damned if I can work out what additional rights it needs, and where I need to add them. Cheers for any help people can throw my way! Ben

Ok this is an odd situation I’m in, but I need help! - We have 2 AD forests in our environment, 1 for testing and 1 for development. All the user accounts are located in the dev forest, with a forest trust relationship between the 2 domains so users can use their dev user account to logon and make changes in test. All worked perfectly. The issue was that the test forest needed to be moved forward a month in time to September to test some procedures. So I made the relevant changes to stop the PDC emulator using NTP for its time source, changed its internal clock etc etc. And all the domain sprung forward 1 month. But the issue is now that the users cannot log in from the dev forest, the error they get is: "the current time on this computer and the current time on the network are different" Now I did expect this, as I know the threshold is 5 minutes for the clocks to be out to stop replay attacks. But this is configurable, so I changed the setting to exceed the 1 month that I have pushed the clocks forward.... but nothing, no one can still log in from the dev domain. Users in the test forest (admin accounts) are fine and they can log in no problem, but nothing from the dev forest. I'm assuming that the "Maximum Tolerance For Computer Clock Synchronization" Kerberos setting doesn’t apply for cross forests? Or something along those lines?? - Has anyone else every seen this if so help! Cheers Ben

Spot on with the lease issue, thanks a million, saved me a heap of time!! Cheers Ben

Thanks for all the info, I'm going to be busy testing this all out! Cheers again

Ahh yea that’s an idea, i'll give it a go - cheers! (i'm using the administrator account) You have lead me onto another dumb question now, which is how have you got the install files being copied to the local machine? I've tried creating a wim file using imagex, and then using the imageinstall section to push the files down, but this just seems to get ignored during the install? Sorry for the dumb questions here, I really appreciate peoples help. Cheers Ben

Anything, specifically office I’ve been trying to do but nothing works, it either skips the install completely or bombs out with an error. Is it my syntax that is wrong? Can I not just specify a unc path to an .exe? (and add the credentials of course) Sorry to be cheeky but do you have a example of one that works for you?? Many thanks Ben

I wonder if anyone can throw some light on this... Basically I have a Vista image deployed through WDS, which works fine. However I’m at the point now where I want apps installed as part of the procedure (Office etc) but I cant work out where to place the command to kick off the install. You can place the runsynchronous command in both the specialize and audituser sections, but neither seem to ever work. I've tried with a UNC path with credentials, and tried using the distribution share with local paths, but the app install is never called. - help! I'm sure its something simple that I have missed! Cheers for people’s help in advance Ben

I'm still completly unsure why when you assosiate a unattend.xml to the image that WDS stops joining the computer to the domain for you. However the issue I had with the computer wanting a name even though it already had one was fixed after I added the skipMachineOOBE to the xml. Previously I had only the skipUserOOBE set. Cheers Ben

Sorry if I’m being stupid here, but I cant get our Vista build to correct join the domain. Let me explain... If I use the WDSunattend.xml (assigned at the server level) to automate the disk partitioning, image selection etc. and then don’t assign an unattend at the image level to automate the windows welcome then my machine joins the domain as per the settings on the WDS client. However the minute that an unattend is added to the image to do anything (even tried a blank xml) then the machine doesn’t join the domain. I tried a few other things, and then added the domain join section in the image xml, this works and the computer object successfully is created in AD. However the machine will then always ask for a computer name, even though a random name has already been assigned for the domain join. Regardless of any setting i add into the machine name section I’m always prompted. Once you input the name then the workstation name is different to the domain workstation name and you are never able to log on. Help! I'm sure i'm doing something dumb here, but i'll be damned if I can find out what it is! I'd rather the domain join be handled by WDS but I dont know why as soon as you assign a image unattend it stops the computer from joining. But if this is by design I just need to know how to stop the machine asking me for a computer name when it already has one! Cheers for any light anyone can throw. Cheers Ben

Cheers! Ben

I couldn’t agree more, the thought of having SIF's per machine type is something I don’t what to touch. Is there another way that you know to get the drivers into RIS without using the MassStorage / OEMBootfile method? Cheers for all your help, its much appreciated!! Ben

Thanks for everyone’s help on this, I just wanted to clarify one more point. Since adding the SATA drivers the machines with SATA drives build perfectly, however the other machines all produce a STOP error after the file copy, just as the machine boots to start installing windows. Is this normal behaviour? Do you require another template file for machines that need additional controller drivers? Or has something gone wrong here? Cheers Ben

Brilliant, cheers for the information, definatly food for thought. I'm going to have a bash at this and I'll let you know! Cheers again Ben

Just a question this one to see if anyone has done this / knows how to. Basically we support 12 languages through RIS, and approx 20 different models of machines, all needing different S/W / drivers etc. At the moment through RIS you select the template you want based on the language of windows you want. Then once the machine has installed and booted for the first time another prompt appears (simple VB script) that prompts for the model of machine, which installs any model specific items. The annoying thing about this is that it’s not completely automated as you need to have that second prompt. So I was wondering if anyone knows how to have another template selection screen in RIS. e.g once you have selected for an English install, then another screen appears for the model of machine. So depending on what you select it points to the correct .sif file and builds the entire machine. Obviously I could have a screen that displays every possible build, but there would be 240 odd selections! Anyone had the need to do this before?! Cheers Ben

Now I feel like a right prat! - Yup I had put the TEXTMODE directory in the wrong place, as soon as I moved it then it all sprung into life. Thanks for all your help!! Ben