Creating a computer group based on text file (orcsv etc)

Content Authoring
1

Hi,

We have a large BigFix deployment (>10,000 endpoints) and have various groups already in place for patch management etc but now we have a new requirement to complete an ILMT report for TSM PVU sub-capacity reporting. We need to create a group covering the, approx 5000, systems that backup to a subset of the TSM servers.

The nodes don’t really have anything in common (computer name, IP, OS etc) so ideally this would be a dynamic group that somehow looked at ‘backs up to server X’, however I can only imagine would be possible by looking for the ‘TCPServeraddress’ entry in the tsm .opt file on these systems which I’m not sure is a realistic possibility.

Given the above I think that means we’re left looking at a manual group based on name or IP. We can extract the list of IPs of all active nodes from the TSM servers - is there a way we can import this into BigFix to create this group? If that is possible can we update the group by repeating the process with a new IP list?

Any tips on the above or any other suggestions for creating a group for ILMT to use would be appreciated,

Cheers.

2

I have an analysis at https://bigfix.me/analysis/details/2994716 to retrieve many TSM client data points, including the configured TSM server name from dsm.opt.

3

Hey,

Thats great, although it looks like it takes the dsm.opt location from the registry, am I correct there? If so do you have something similar for AIX, Linux etc OS please?

4

Sorry, I wrote it for Windows. It shouldn’t be hard to modify it to look in a Linux/AIX path though. Is the dsm.opt located in the same path on all of your platforms?

5

Its mostly the standard paths I believe so yes, we should be able to modify for the non-windows clients (that said your analysis has mostly returned so far but we won’t hold that against you, I think we’re on the right path :slight_smile: )

In broad terms creating an analysis that checks for the existence of a .opt file in the standard path(s) and then returns the TCPSERVERADDRESS from there seems like a good approach.