This content has been marked as final. Show 7 replies
you need to go look in the agent policy catalogue
you need to alter the agent policy settings to point first to the distributed repository on their own site (and you may want it to default back to the main one if this is not available so I would keep this as an option)
you then apply this policy only to the agents on their site.
the options are ping time (fastest responding to ping repos), subnet or specific list
If you don't mind, I'd like to piggy back on your question because I am doing almost the same exact thing only in my instance, Site B is on a different domain but still can talk to the ePO server at site A. When I tried to install epo 4 at site B even though I had no depoloy tasks enabled, it started to pull client servers from A site into 4.0. I had to go back to my Site A 3.6 epo server and manually repush the agent to all the effected nodes after I took the epo 4.0 box offline.
I am thinking its probably better even though its seperate domains to use just 1 epo server and setup a distributed repository at site B.null I am going to need the same help as Derby setting up the distributed repository and making a new framepkg file.
So if I am understanding how this works correctly, I will then have two different framepkg.exe files? One for each location?
firstly, many thanks for your reply toby.
i have changed the policy to include a configuration that first looks at the B site and then at the A site. i have only done this on the epo agent section - do i have to do similar things on the other software? the virus enterprise 8.5.0 for example?
James, what you/ve said on your reply is basically what i was after - a frame package for site A and a frame package for site B.null is that actually an option or do you only have one frame package and then apply different policies to it for the different groups in the model?
i think it's perfectly possible but it's just getting it round my (our) head(s)... happy
The best option is just to install your agents however you will and then differentiate them according to the agent policy you apply from the epo server to the different segments of the directory ( I assume of course that you do seperate them out in the directory)
There would be no need for different VSE policies unless you actually have some difference to apply.
That's what is confusing me also. I'd like to keep the two sites as seperate as possible. I'd like to know...
How do I get the site B nodes to use the repository at site B without effecting configigured and working nodes at site A? Is it a totally diffferent framepkg that gets created?
goes like this.
you have site A with 1 epo server that covers all your agents
at site B you have a distributed respo that collects the dats from site A
in your directory on the epo server you apply the following policy to your agents
POLICY 1: Use the epo server for DATS/updates
POLICY 2: Use the dist repos for DATS/Updates but go to EPO if this is down
(this would be the specific list with dist repos at top)
apply policy 1 to agents within the site a groups in directory
apply policy 2 to the agents in the site b groups of the directory
it all makes perfect sense now tony (i've just realised that in my first reply i called you toby...my apologies!)
i will set this in motion and see what happens.
again, many thanks for your help