I read that software bundling can be done in IBM Control Desk/Maximo (ICD) if using a discovery tool other than BigFix. What are the disadvantages of performing manual bundling in ICD rather than BigFix? Has anyone done a comparison of the bundling between the two?
We have the latest version of ICD and BigFix Inventory. The BigFix All Metrics Report and Audit Snapshot Report include only about a third of the part numbers we have installed from the Passport Advantage website.
In my opinion, if you are using BFI or ILMT then that is where bundling should be accomplished. These tools are specifically designed for this. They feature:
- A canonical software catalog, which governs software hierarchies, is used to normalize the discovered software
- Automated bundling rules which provide guidance for classification of discovered instances of software components to their parent products
Note also that, according to IBM’s Passport Advantage Agreement, ILMT (or BFI) must be used when a customer installs any eligible sub-capacity product in an eligble virtualization environment, unless the customer qualifies for manual usage tracking. In practice, this means the customer is not using full capacity licensing or it’s Enterprise servers have more than 1000 PVUs of physical capacity.
If you are to perform bundling in ICD, then you cannot integrate it with BFI/ILMT, since this will make the Software Catalog hierarchy read only. When non-IBM discovery tools are integrated with ICD, the software catalog is actually generated during the process of importing discovered software into ICD. These catalog entries must be manually edited to reflect the proper bundling of components into products. All these hierarchies must be researched, validated, and kept up to date.
There are several possible explanations for the “missing” part numbers in your reports.
- The instances of the components associated with these products must be discoverable:
- Are any instances actually installed?
- Are they installed on machines without BigFix Clients?
- Have the installations been modified such that they are not visible to BFI?
- Are they installed on systems in areas of the Enterprise that are not included in the BFI/ILMT deployment?
- The discovered instances of candidate components may be assigned incorrectly to other products in the Software Classification panel
The part numbers are only one element of the bundling process. Regardless of whether bundling is done using BFI/ILMT or ICD, proper classification of each discovered instance of a component requires knowledge and/or documentation from the IT staff responsible for installing the products. This work is unavoidable when the software is installed prior to the discovery tool’s implementation or when the software installations are not managed by a solid business process that includes the software asset management team.