DeployStudio Forum

thequietman44 · 2008-11-10 20:59:28

I just set up DeployStudio Server 1.0rc9 and I'm trying to restore a Leopard image to a MacBook.

DeployStudio is running on an iMac (10.5.5), the repository is located on an OS X server running 10.5.5 Server, and I'm trying to restore from a Netboot image hosted on that same OS X server. The image is a 10.5.4 build (compressed, ASR-ready) that I've been deploying with NetRestore just fine.

The MacBook boots from the Netboot image fine and up comes DeployStudio Runtime asking me to choose a workflow. I choose "Restore a master on a volume" (one of the default workflows), hit play, choose my image from the list, then hit play to begin the restore. I get the following output in the log:

Code:

2008-11-10 12:24:59.988 Runtime[95:6733] -> testing disk '/dev/disk1'
2008-11-10 12:24:59.989 Runtime[95:6733] Refused: boot disk
2008-11-10 12:24:59.989 Runtime[95:6733] -> testing disk '/dev/disk0'
2008-11-10 12:25:00.858 Runtime[95:6733] Accepted: non boot disk
2008-11-10 12:25:01.874 Runtime[95:6733] "/usr/sbin/diskutil" partitionDisk /dev/disk0 1 GPTFormat "Journaled HFS+" _DS_DEV_DISK_X_ 1000T 2>&1
2008-11-10 12:25:02.229 Runtime[95:6733] Started partitioning on disk disk0
2008-11-10 12:25:02.231 Runtime[95:6733] Creating partition map
2008-11-10 12:25:32.425 Runtime[95:6733] Partitioning encountered error on disk disk0: Could not unmount disk (-10000)
2008-11-10 12:25:32.428 Runtime[95:6733] Could not unmount disk for zeroing
2008-11-10 12:25:32.429 Runtime[95:6733] "/usr/sbin/diskutil" mountDisk /dev/disk0 2>&1

"Interesting," I think to myself, "I suppose I could just format the disk myself before restoring," but even with the "Erase target for block copy restoration" option turned off, it still tries to format the disk before restoring. Am I doing something wrong on the DS Server end? Runtime end? Or is this really an inexplicable error?

I've tried partitioning/formatting beforehand using Disk Utility, tried several different images (including one I created with the "Create a master from a volume" workflow), but I get the exact same output in the log and the restore fails.

Any insight would be greatly appreciated.

admin · 2008-11-10 21:51:09

Hi,

did you create the netboot set on a client computer?
If no, you'd better go for it.

Partitionning failures are a known bug but we've found no way to reproduce it consistently.
Most of the time, partitioning with disk utility.app and relaunching the workflow fixes the issue.

It as nothing to do with the server side and is probably a missing resource in the netboot set.

Let us know if a netboot set based on Mac OS X client fixes the issue.

admin · 2008-11-11 01:34:09

It sounds like you have checked the option "restore the image on the first drive available".
It means that the Runtime will take the first non boot disk and partition it to get a suitable volume to restore the disk image.

In order to prevent the partitioning step, you need to select a volume (not a device) and turn on the erase volume option.

thequietman44 · 2008-11-11 14:36:43

Okay, created a Netboot set from the machine I'm trying to image and the partitioning works now. However, it gives me an error about the file checksum:

Code:

2008-11-10 18:23:09.190 Runtime[104:10b] Restore action:
2008-11-10 18:23:10.574 Runtime[104:3ffb] Looking for the first available disk:
2008-11-10 18:23:10.575 Runtime[104:3ffb] Boot device='/dev/disk1'
2008-11-10 18:23:10.575 Runtime[104:3ffb] -> testing disk '/dev/disk9'
2008-11-10 18:23:11.035 Runtime[104:3ffb] Refused: protocol=Disk Image
2008-11-10 18:23:11.036 Runtime[104:3ffb] -> testing disk '/dev/disk8'
2008-11-10 18:23:11.512 Runtime[104:3ffb] Refused: protocol=Disk Image
2008-11-10 18:23:11.513 Runtime[104:3ffb] -> testing disk '/dev/disk7'
2008-11-10 18:23:11.964 Runtime[104:3ffb] Refused: protocol=Disk Image
2008-11-10 18:23:11.965 Runtime[104:3ffb] -> testing disk '/dev/disk6'
2008-11-10 18:23:12.425 Runtime[104:3ffb] Refused: protocol=Disk Image
2008-11-10 18:23:12.426 Runtime[104:3ffb] -> testing disk '/dev/disk5'
2008-11-10 18:23:12.895 Runtime[104:3ffb] Refused: protocol=Disk Image
2008-11-10 18:23:12.896 Runtime[104:3ffb] -> testing disk '/dev/disk4'
2008-11-10 18:23:13.347 Runtime[104:3ffb] Refused: protocol=Disk Image
2008-11-10 18:23:13.347 Runtime[104:3ffb] -> testing disk '/dev/disk3'
2008-11-10 18:23:13.816 Runtime[104:3ffb] Refused: protocol=Disk Image
2008-11-10 18:23:13.817 Runtime[104:3ffb] -> testing disk '/dev/disk2'
2008-11-10 18:23:14.281 Runtime[104:3ffb] Refused: protocol=Disk Image
2008-11-10 18:23:14.282 Runtime[104:3ffb] -> testing disk '/dev/disk1'
2008-11-10 18:23:14.283 Runtime[104:3ffb] Refused: boot disk
2008-11-10 18:23:14.283 Runtime[104:3ffb] -> testing disk '/dev/disk0'
2008-11-10 18:23:15.120 Runtime[104:3ffb] Accepted: non boot disk
2008-11-10 18:23:16.090 Runtime[104:3ffb] "/usr/sbin/diskutil" partitionDisk /dev/disk0 1 GPTFormat "Journaled HFS+" _DS_DEV_DISK_X_ 1000T 2>&1
2008-11-10 18:23:16.427 Runtime[104:3ffb] Started partitioning on disk disk0
2008-11-10 18:23:16.429 Runtime[104:3ffb] Creating partition map
2008-11-10 18:23:34.905 Runtime[104:3ffb] Formatting disk0s2 as Mac OS Extended (Journaled) with name _DS_DEV_DISK_X_
2008-11-10 18:23:37.867 Runtime[104:3ffb] Finished partitioning on disk disk0
2008-11-10 18:23:37.938 Runtime[104:3ffb] /dev/disk0
2008-11-10 18:23:37.939 Runtime[104:3ffb]    #:                       TYPE NAME                    SIZE       IDENTIFIER
2008-11-10 18:23:37.939 Runtime[104:3ffb]    0:      GUID_partition_scheme                        *111.8 Gi   disk0
2008-11-10 18:23:37.940 Runtime[104:3ffb]    1:                        EFI                         200.0 Mi   disk0s1
2008-11-10 18:23:37.942 Runtime[104:3ffb]    2:                  Apple_HFS _DS_DEV_DISK_X_         111.5 Gi   disk0s2
2008-11-10 18:23:37.943 Runtime[104:3ffb] "/usr/sbin/diskutil" mountDisk /dev/disk0 2>&1
2008-11-10 18:23:38.302 Runtime[104:3ffb] Volume(s) mounted successfully
2008-11-10 18:23:38.306 Runtime[104:3ffb] '/tmp/DSNetworkRepository/Resources/Masters/HFS/FacLapLeopardVan.dmg' size: 7461.3 MB
2008-11-10 18:23:38.306 Runtime[104:3ffb] '/Volumes/_DS_DEV_DISK_X_' size: 114145.4 MB
2008-11-10 18:23:38.781 Runtime[104:3ffb] "/usr/sbin/disktool" -A /dev/disk0s2 2>&1
2008-11-10 18:23:38.809 Runtime[104:3ffb] Adopting volinfo database device /dev/disk0s2
2008-11-10 18:23:38.812 Runtime[104:3ffb] Displaying volinfo database device /dev/disk0s2.  permissions = 1
2008-11-10 18:23:38.813 Runtime[104:3ffb] Setting boot device to '/dev/disk0s2'.
2008-11-10 18:23:39.096 Runtime[104:3ffb] "/usr/sbin/asr" restore --source /tmp/DSNetworkRepository/Resources/Masters/HFS/FacLapLeopardVan.dmg --target /dev/disk0s2 --puppetstrings --noprompt 2>&1
2008-11-10 18:23:39.531 Runtime[104:3ffb] XSTA    start    97    client
2008-11-10 18:23:39.533 Runtime[104:3ffb] XSTA    setup
2008-11-10 18:23:39.582 Runtime[104:3ffb]     Validating target...done
2008-11-10 18:23:39.583 Runtime[104:3ffb] XSTA    metadata
2008-11-10 18:23:41.727 Runtime[104:3ffb]     Validating source...done
2008-11-10 18:23:42.882 Runtime[104:3ffb]     Retrieving scan information...
2008-11-10 18:23:42.883 Runtime[104:3ffb] The source image is missing file checksum information.
2008-11-10 18:23:42.883 Runtime[104:3ffb] Please rescan it with the --filechecksum option.
2008-11-10 18:23:42.898 Runtime[104:3ffb] Could not retrieve scan information - No such file or directory
2008-11-10 18:23:42.900 Runtime[104:3ffb] Command failure: XSTA    fail
2008-11-10 18:23:42.908 Runtime[104:3ffb] Restoration failed
2008-11-10 18:23:43.106 Runtime[104:3ffb] -> Restore action completed.
2008-11-10 18:23:43.107 Runtime[104:3ffb] Restoration failure (elapsed time: 0.57 minutes)

Is there some special checksum that needs to be added to the image besides the ASR scan? How?

Thanks!

Last edited by thequietman44 (2008-11-11 14:38:32)

thequietman44 · 2008-11-11 15:39:03

Ahh, since I had disabled the "Erase target for block copy restoration" while troubleshooting the partitioning error it was trying to do a file-level restore. The image does not have a file-level checksum, only block-level, since I didn't want to allow file-level restores in NetRestore.

Checked the "Erase target..." option and it does a block-level restore just fine now.

So now the only question I have is, do I have to create a Netboot set for every type of machine I want to image? That might get a little tedious.

admin · 2008-11-11 23:40:48

It takes longer to create a disk image with DeployStudio because block/file checksums and multicast reordering are done by default.

You shouldn't need to create a netboot set for each type of computer (only for new hardware).
We've just noticed that weird things could happen with netboot sets based on Mac OS X Server.

thequietman44 · 2008-11-11 23:52:49

As a rule I don't create Netboot sets on the server because I've had compatibility issues in the past.

The original Netboot set (with the partitioning error) was created on an Aluminum iMac 2.0GHz running the client version of 10.5.5. The new Netboot set (no errors when partitioning) was created on a MacBook 2.1GHz running the client version of 10.5.4.

I could try creating the Netboot set on the MacBook after updating to 10.5.5 to make sure the issue wasn't with 10.5.5 if you'd like.

DeployStudio Forum

Announcement

#1 2008-11-10 20:59:28

Restore HFS Fails with partitioning error

Code:

#2 2008-11-10 21:51:09

Re: Restore HFS Fails with partitioning error

#3 2008-11-11 01:34:09

Re: Restore HFS Fails with partitioning error

#4 2008-11-11 14:36:43

Re: Restore HFS Fails with partitioning error

Code:

#5 2008-11-11 15:39:03

Re: Restore HFS Fails with partitioning error

#6 2008-11-11 23:40:48

Re: Restore HFS Fails with partitioning error

#7 2008-11-11 23:52:49

Re: Restore HFS Fails with partitioning error

Board footer