damn web forms...
I just did a whole writeup to answer your question and when I posted it I got an unexpected error! I lost all of my post!
I have the doc you need; I'll try to post it again. If I can't get the document attached just email me and I'll send it to you.
the document you need is titled:
I'm trying to post it as an attachment.
sample command line arguments without all of the explanation I gave previously; image file contains two partitions and they are being laid down individually so as to control the size of them on the target computer.
/r /f:i:\%imagepath% /sp:1 /d:0 /clr /m /rs:%cpartitionsizeGB%000 /rb:0
/r /f:i:\%imagepath% /sp:2 /d:0 /m /x
you'll likely focus on the /d: switch as that controls which drive you are targeting. I'm assuming that you have a single image file with a single partition going to a single disk. If you have multiple partitions you can specify the specific one you want using the /sp: switch
to answer 1 - the latest manual is here - http://www.terabyteunlimited.com/downloads/ifw_en_manual.pdf
question 2 - if you want to retain the d drive, when you capture your image. Only capture the first partition. This can be done with the /d switch which specifies what partition to capture.
example: REMEXEC24=ldrun H:\imagew.exe /b /uy /comp:10 /d:0@1 /rb:0 /max:698MB /f:"""i:\%Computer - Device Name%"""
/d:0@1 means select the first disk and select the first partition only
so when you lay down your image just use the normal command:
REMEXEC16=ldrun i:\imagew.exe /r /o /d:0 /f:"""i:\image.TBI""" /rb:0
and it will only deploy the image to the partition that you specified in your capture and d: drive will stay intact.
This is a requirement for our company for "REIMAGES" vs "NEW HARDWARE" so i know it works for me. Let me know how that works out for you.
EDIT, my post was assuming you didnt have an image already created. If you already have the image created with both C: and D: partitions then as AspenSkier stated, you might look at the /sp switches to specify what to deploy.
Thanks for all the posts folks. Much appreciated. Couple more questions which I hope are already considered:
1. My 2008 R2 64-bit image does not contain a D: so we're good there, but it does contain the 100MB SYSTEM Partition created by a server 2008 install. This is captured in the image and is deployed with the image. I'll happily modify my image or how it's captured based on recommendations, but will this be an issue for Imagew v2 while leaving D: intact and using Diskpart to do a "Delete Volume" of C:? Is deleting the C: volume before restoring the image even necessary?
2. The target servers each have a C: and D: volume. C: is always 30GB and D: is expanded to fill whatever's left of the disk. I build my images in VMWare with a 40GB dedicated hard drive size and a 30GB partition for the OS, not including the SYSTEM part. I used to capture my reference images with a 20GB partition (Win XP) so there were no issues depploying to different drive sizes or partition sizes. Is this still an issue with Imagew v2 or do I not need to worry about the size of the partition in my image?
I will have a lok at the manual for imagew v2. Thanks for any more thoughts or suggestions on these questions.
The tool is still a sector based imaging tool so the initial partition size is still important as it needs to be small enough to fit the partition onto the disk before expanding it.
I've always found it best to consider things as partition numbers rather than drive letters because in your case there is a partition that would never get a drive letter. You will need to delete the partition that holds your C: drive before you restore the new one. Be careful on those partition numbers as it could be a very easy thing to delete the wrong one!
Re: partition numbering, I actually started down that path but realized LANDesk's provisioning diskpart actions only allow you to enter a Partition number, where as using an actual Diskpart command script allows you to specify a partition number or a volume letter to be removed. C: will always be C:, but depending on who installed the original server OS there could be some one-offs in partition numbering. Hopefully they are all consistent, but given I need to keep the D: volume I'd rather reference by letter if possible.
In a typical deployment where LANDesk is allowed to wipe the drive (Provisioning action), the SYSTEM partition from my image gets laid down as Partition 1 and the C: drive is partition 2. I then run FIXWINDOWS to clean up the driver lettering etc. This is where my confusion begins with keeping D: intact, as in the target servers the D: volume is Partition 2. How can imagew v2 account for this?
correct on the note about imagew being sector-based...when building images I always build a system partition large enough for what I am trying setup, but undersized from where it will be deployed. For Win7 clients I think I'm still using a 25GB-30GB partition size on the image and deploying it to drives that are at least 80GB where maybe 50GB goes to the deployed system partition.
Regarding the restoration of a single partition using ImageW...you may run into a case where your command lines are not explicit enough to prevent the new partition from overwriting the existing stuff that you want to keep. In this case you're talking about your 100mb boot partition and the data partition (D).
Recently I was needing to setup a dual boot winXP/win7 machine. needed to have a 100MB boot partition for the o/s loader, 20GB XP partition, 30GB win7 partition, and finally have the rest of the disk space taken up by a 4th data partition. I was building this from two different image files using ImageW.
Initially I ran into a case where ImageW was prompting me inside of WinPE that I was about to overwrite my boot partition with my XP partition...this was because the XP parition was originally built without the 100MB offset and so when ImageW was laying it down it thought that it was supposed to go where the boot partition was. This is exactly the kind of challenge you face when using a sector based imaging tool. I'm not sure what I did but I was able to eventually get it to work. Maybe using the /m switch? Perhaps combined with using an offset? I really don't remember.
While this is only background for you, I hope it helps.
you say that the target server D drive has a partition number of 2...and that the image you are laying down has the boot partition assigned to partition number 1...
?prior to doing all of this, do your servers have A) C & D only or B) boot, C & D?
For what its worth, ImageW doesn't care about a destination partition number...it only cares about a source partition number in the source image file and a destination drive. /SP: is the source Partition switch followed by 0,1,2,etc and /d: is for the destination drive followed by 0,1,2, etc
Assuming you answered A above, I don't think it really matters if you end up with a final partition order on the drive of 1, 3 (c), 2 (d).
Thanks. I gave you the win for the thread for the last little bit about imagew not caring where the restored partition winds up on the drive, but if what you say is accurate you also solidified a worry of mine which is having to do any future "diskpart-ing" with static scripts. What it looks like is I'm going to wind up with a partition setup of 1, 3 (c), 2 (d) like you said. That makes my OCD go nuts, but it may be something I'll have to live with.
I'll give the /sp commands a shot and see how we do.
1 of 1 people found this helpful
I deploy Windows 7 with system partition and can reimage with D: drive in intact. I will just explain how I do my imaging and maybe it can help you out.
We decided our Win7 Images would be 100mb system reserved, 75Gig C: Drive and the Remaining on D: Drive. All of our new Windows 7 machines come with a minimum of a 150gig hard drives.
I created the image with 2 partitions:
- 100mb System Reserved
- 40gig C: Partition (I created this 40gig because we had some testers with smaller hard drives than our standard and I made it that way to accommedate them and the smaller the partition is better for a sector based tool like imageW)
I captured the image as a full disk split into CD file splits just incase someone didnt have DVDs and its easier to replicate to my distribution points.
Capture command was = H:\imagew.exe /b /uy /comp:10 /d:0 /rb:0 /max:698MB /f:"""i:\%Computer - Device Name%"""
For Deployment we have two choices:
A New Machine Image and a Reimage Choice
A New Machine Image is a used for fresh new machines or if someone wants to start over fresh and completly wipe out their drives.
New Machine Image Deployment
Key items in this deploy are:
- diskpart /s X:\LDClient\WIPECLEANDISK0.TXT to wipe the disk
- deploy using default settings - i:\imagew.exe /r /o /d:0 /f:"""i:\Win7.TBI""" /rb:0
- Expand the partition to the correct size and create and formate the D: Drive
- Diskpart Script
select disk 0 select volume 2 extend size=35840 select disk 0 create partition primary select disk 0 select volume 3 assign format FS=NTFS Quick
Key items in this deploy are:
- I remove diskpart /s X:\LDClient\WIPECLEANDISK0.TXT action from the script
- I deploy the image twice: 1 for the system reserved, 1 for the boot partition goes like this:
i:\imagew.exe /r /o /fd /sp:1 /d:0 /f:"""i:\win7.TBI""" /rb:0
i:\imagew.exe /r /o /fd /sp:2 /d:0 /f:"""i:\HDSImages\Win7\win7.TBI""" /x /rb:0
The /sp specifies the partitions to lay down and all the second command i use /x to expand the 40gig partition back to the 75gig which it was previsiouly. The D: Drive remains intact.
- I wait for FixWindows to fix the drive letters
That is how I do it and It works out quite well for us.
If I could only give more than one correct answer
As of the time when you must have been writing this I am about half way done doing it exactly as you have laid it out. It was the /sp that clinched it.
Out of curiosity, although I'll find out here in a minute, do your partitions wind up being out of sequence when you re-image? I guess they wouldn't if you're already starting with 3 partitions, but coming from a Server 2k3 or XP machine I suppose they would. I was looking into the /e switch for Image given this description in the hopes that I could preserve the SYSTEM=Part1 C:=Part 2 D:=Part 3 convention. I think I'm on track now but I would really like it if the partition numbering would follow the general flow of the drive. Normally I don't have to retain the Data )D:) drive so this is the first time I've been asked to leave anything behind.
Use this parameter to have Image for Windows move the partition table entry of the restored partition to the same location in the master partition table as it appeared on the source drive. Image for Windows will move the existing partition table entry to another location rather than overwriting it.
Default if omitted: Image for Windows does not move the partition table of the restored partition.
Thanks to all who posted, much appreciated.