This content has been marked as final. Show 34 replies
I use RIS here for all of my images and have not had any problem with the imaged machine checking back in correctly. What happens is, I usually install the image to the machine and once the agent checks in with its new name, it shows up in lost and found. Once it shows up there I simply cut and paste it to the appropriate group.
It's not a problem.
Sorry to bring a post back from the dead, but it appears that with the advent of EPO 3.5 and Patch 1 that this doesn't work how it used to. All my newly imaged computers aren't showing up in my directory. I tried clearing out the GUID and then putting the image on, but it looks like the GUID isn't regenerating itself. Anyone else having this issue?
I'm with you on this one...
I have a image that currently is really messing with us. The agent will appear - then disappear (even if you are looking at the computer it will change to another one), checkin under the wrong group (IP filtering is in use) if there is a computer already setup for it to check in as it will make a new one in a different group, etc....
If you check the All Connecting computers it shows up right were it should be, but does not display in the directory tree in the console.
I hope more poeple speak up on this, because it is really strange... and I hope we can find a way to solve the problem.
I was having the same problem with 3 imaged machines. Indeed, the GUID or Agent ID is the problem. Just reinstalling the agent wont regenerate this ID. Youll have to do a small registry tweak first.
Here is how I resolved the issue on Win 2000 Pro machines.
First I went into the registry and deleted this key: (As always, backup your registry first)
HKEY_LOCAL_MACHINE\SOFTWARE\Network Associates\ePolicy Orchestrator\Agent
This key identifies the agent to the server and makes it unique as it communicates.
Next, rerun framepkg.exe to reinstall the agent. Reboot when its complete..you dont have to wait for the Agent icon to appear on the taskbar.
After the reboot....open up the agent and verify that you have a new Agent ID, then click Collect and send props to force the agent to connect to the server. On the server, the machine should fall in place where its supposed to go.
Hope this helps!
Ouch, having to do that to 500 PCs might be an issue..;) Any other thoughts? We recently did a massive reimage (pushed out SP2) and only, maybe 1/2 of the clients re-registered with the EPO box.
Now, everything is getting policies, everything is getting updates, (currently), however only at most 1/2 of my Workstations is being shown in the EPO list. Note, EPO server is running patch 1, clients are currently patchless. (was waiting for everything to be put into place before patching)
Yeah, I can see where that might be a problem. Ive never attempted this before, but if there was only a way for you to remove the above mentioned registry key via a login script, then the agent should regenerate a new ID when the machine boots up & checks in with the EPO server.
Another thing to consider as well: Hope that Mcafee will release a patch that will resolve this issue. This problem seems to have appeared after the release of Agent 3.5 and Epo 3.5....if a fix does come along, then your agents will correct themselves.
I have since recreated our main image for our base user setups to not include the Agent ID key.. so after reimaging a new machine, Agent starts up with a new ID and shows up in the proper group on the EPO server.
Well if all that it needs is a reg key, and not reinstalling agent, its possiable we could toss the del *regkey* into Novell and have it run on login to a small local group, and see what happens.
I will try this, see if it works... I hope this is not going to be a problem... if so I'll just install an older version of the agent on our images and let them upgrade once they check in... I do not know if that will work though....
A patch would be the best!