hi everyone,

We are in deployment phase of Zen11, especially for the deployment of the Agent to workstations (win7 x64 but no sp1). I have patched the agent with the last cumulative patch 1 that correct the random reboot bug. But I have some questions about the correct deployment method, because I have made many tests and I have encounter some issues and weird things. Here is my concerns:

1- what is the difference between the 3 kind of agent, and what is the best to use: network agent, complete agent and agent without the .NET. We have Win7 x64 with no sp1.

2- If I install a EXE package of the agent to a workstation and if at this time the server from which the package was downloaded is closed or not responding, the client machine will not being registered. I've tested it and the server must be online. The FIRST INITIAL connection between the client and the zen server address showed in the Properties seems to be linked. So if the server is down or crash, my client can't register correctly, otherwise I need to manually enter an new server IP with ZAC REG .... But if my client succesfully register the first initial connection, after that the client can communicate with any other zen servers (tested and works well when some of the zen servers are not online). Is there a way to test a zenworks connection before we install a agent?

3- it seems that we cannot make an image from a base machine that have already the agent, even if we uninstall it. There's still trace of the agent, and If I proceed anyway with sysprep, the cloned machines will have the same zenworks GUID after reboot..... and pointing to the same registered object. So I have found that a base machine MUST not have already the agent, this way when duplicated, no mess with duplicated objects in zen database. I know about the Agent Cleanup tool, don't know if it's safe to use on a base machine...

thank you for your support!