4 Replies Latest reply on Jan 16, 2020 10:09 AM by Daniel Hudsky

    HasImpact, ImpactWeight, ImpactDirection attributes

    Alex Genström
      Share This:

      Dear community I need your help with understanding the following behavior.


      As we all know, BMC Discovery (we have version 11.3) has the ability to detect the impacted components such as for example Clustered systems. Let’s imagine we have a cluster of 2 hosts. This basically gives us a hint, that if one of those 2 nodes will go down, another one shall take over and the cluster shall in theory be operable, however with 50% of its capacity and therefore the impact shall be visible in the application model (if we add the cluster to it). That’s would be pretty logical.

      Also, I found the following in the documentation: Syncmapping block - Documentation for BMC Discovery 11.3 - BMC Documentation where you actually see the following text:

      The Impacted and ImpactWeight attributes can be used to create impact relationships. They specify impact direction and percentage respectively. For example, an ImpactWeight of 25 might be appropriate when representing an impact relationship between a BMC_ComputerSystem and BMC_Cluster, where there are four computers in a cluster. The value assigned to Impacted must be the Source or Destination CI appearing in the same definition.


        Source  := computersystem,

        Destination  := cluster_ci,

        Name  := "CLUSTEREDSYSTEM",

        Impacted  := cluster_ci,

        ImpactWeight := cluster_rel.impact_weight


      This is basically what we are trying to achieve here, however I could not locate where in the OOTB syncmapping Discovery maps Weights.. I started to dig further down and found another community post where a BMC employee stated that BMC Discovery "Discovery synchronization mapping does not set impact weight": Clusters and weight

      More than that, it has been an article that describes the behavior of CMDB Normalization rules that sets the value to ImpactWeight attribute: Atrium Impact Simulator, AKA AIS and the Weighted Cluster mistery


      So here I can clearly see a missmatch in documentation vs few topics and answers within them. In my opinion there is only one way to determine Weights in clustered systems - via Discovery. However, there are OOTB rules in CMDB that are active and that sets ImpactWeight to 100 to basically everything. Just configuring those rules doesn't make any sense, as we can't decide on the impact of the device if we have only that device information and nothing else.. This makes Impact Analysis impossible to achieve.


      I am lost with above and need some clarification. Has anyone succeeded with Application Modelling that could use Impact Analysis in CMDB in order to simulate the impact ? If yes - would you please share your expereince with me?


      Thanks in advance.

        • 1. Re: HasImpact, ImpactWeight, ImpactDirection attributes
          Andrew Waters

          The documentation you reference is talking about sytncmapping blocks as defined in TPL not the default CMDB sync mappings. the standard mapping make not attempt to determine any impact weighting.

          1 of 1 people found this helpful
          • 2. Re: HasImpact, ImpactWeight, ImpactDirection attributes
            Alex Genström

            Thanks Andrew,


            Can I tractate your answer as I will need to build the custom syncmapping for to populate those attributes in order to have the proper impact weighting in CMDB? Then the CMDB Normalization rules have to be disabled, else they will update the weights values to default that is 100... Do I understand you correctly? Also strange, as I thought that Discovery does that OOTB..

            To simplify - I need to know what would be recommended way of populating ImpactWeight attribute in CMDB? How to do it (extended CDM mapping?)? And what has to be done in CMDB-side (disabling/modifying the Normalisation Impact rules)?

            • 3. Re: HasImpact, ImpactWeight, ImpactDirection attributes
              Lisa Keeler

              To change the relationship weighting or such, you must write a syncmapping Override pattern ... with the override keyword.


              See my sample override by searching “sample override”


              Sent from my iPhone

              2 of 2 people found this helpful
              • 4. Re: HasImpact, ImpactWeight, ImpactDirection attributes
                Daniel Hudsky

                Override pattern would have to account for the number computer systems in the cluster and then assign it a proper distributed value.

                100 / x = weight value. Even if the pattern can be written in that way, one cluster point may have more weight than another.

                A use case for this is something close to Remedy AR Server Group where systems that are clustered would be for user facing Asset Operators versus Admin only servers that are still in the same cluster. It is bit more complicated than a simple calculation.

                2 of 2 people found this helpful