Hi there,
We are struggling with Peer to Peer Multicast and are hoping for some suggestions.
We are attempting to deploy our standard Windows 10, sysprepped image. This deploys quite happily when going via unicast.
We are trying to deploy to our test lab, which has 4 machines in it. Our live lab environment will see us deploying to 80-100 machines via multicast (we currently do this for our Win7 Image via GSS 2.5).
The Distribute Disk Image task is setup with the following parameters:
- Data Transfer Mode - Ticked
- Peer to Peer Multicast using Ghost selected
- Max clients to participate in Multicast session - 3 Clients.
- Min clients required for Multicast to being - 3 Clients.
- Connection timeout: 4 Minutes.
Question 1: We haven't been able to figure out what to set the min and max clients to, e.g. For our lab of, say 4 machines should Min clients and Max clients be:
-- Min 4, Max 4
-- Min 3, Max 4
-- Min 3, Max 3
We he have found if we set to 4 in our test lab, it eventually times out and starts a unicast job to the 4 machines.
Setting the Min to 3 and the Max to 3 or 4 seems to work and the multicast job starts. We are assuming here that for the Min clients, you take the total number of machines you want to image via Multicast and subtract one (essentially allowing for one to be the master?) Some guidance here would be great!
Now that we can get the Multicast tasks kicked off, and a master is picked and the clients align themselves with the master, our next issue relates to the actual imaging which fails on all clients apart from the master.
On each client, ghost closes and leaves the automation environment. The GHOSTERR.TXT file on the GSS 3.1 server reports the following (I'll just include the first few lines):
----------------------------------------------
Date : Mon Aug 15 13:38:25 2016
Error Number: 28000
Message: Corrupt ID header encountered
Version: 12.0.0.8023(Jan 19 2016, Build=8023
OS Version: Professional (Build 9200)
Command line arguments: -CLONE,MODE=RESTORE,SRC=N:\Images\Windows 10 UEFI\SYSPREPPED\Working34\Working34.gho,DST=1 -sure -batch -mp -session=54d9d492-c810-486c-8f07-07820ee9d8fa -ip:224.2.0.19:403 -clients=3 -threshold:3 -connecttimeout=4
Active Switches :
AutoName
----------------------------------------------
Corrupt ID header seems to indicate an issue with the image. But we know the image is fine (and will deploy out on the master PC and via any unicast jobs), why are all the clients failing? and does anyone where we best start looking? Is it something in Ghost? Is it something on the network? I can't find anything related to Ghost Error 28000 other than some old GSS 1.1 Sybase database issue which I doubt would be applying to our scenario.
We are at a bit of a loss and would appreciate any guidance. If it matters, we are using GSS 3.1 MP2.
Regards,
SIMST.
