Idea ID 2838201
The customers with ADID discovery and few DFPs say: it'll be easier for us to have only one IP to get the resulta from all agents, not only because it's hard to split our current PCs, but also because of future PCs which will be deployed by different persons from different teams.
I see that the job works the following way now: from DFP \sending folder it only takes the file for particular trigger UDA CI. So if we create one shared storage which will store all eriched files from all \sending folders from all DFP the following may happen (I think it is close to what we need):
- the agent gets installed and sends the callback to load balancer IP. Then we get the UDA CI created and linked to some random DFP A.
- the ADID job for that UDA CI gets started and waits for result file in \sending folder of DFP A (which is actually a shared folder with DFP B)
- the agent sends the results to load balancer and then to DFP B. They get enriched there and appear in the shared \sending folder
- The ADID job for our CI at DFP A gets these results and processes them
This way in reality our UDA CI is always processed by the same DFP, so there are no problems with autodeletion and so on.
But so far there's a problem - the reference to DFP in UDA properties gets changed, so on next run it will go to another probe. This is the main problem in this scenario. If we could get that property locked somehow (not by the enrichement rule which is not stable) - that would do the trick.
Another thing is the folder limitation - 65k of files as far as i know, so maybe ADID will need to check few folders.
It would be great if Micro Focus could take a look at this idea. Any solution for the problem i raised would be a good advantage for UCMDB SW
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.