Hi SAM Superheroes,
I read and watched a lot of content about how discovery models are created and normalized, but I'm still a little confused. I ran an experiment based on this ServiceNow video (https://www.youtube.com/watch?v=ePi5WBqJcOI) and uploaded software installation records (see below).
The resulting discovery model came back as Match Not Found.
Am I foolish to expect SAM Pro to do the heavy lifting of automatically normalizing common software models (e.g. MS Project)?
What am I missing? Do I need to turn anything on for the automatic normalization process to work in my instance?
I tried it for other software installs and it was Match Not Found as well...
I was expecting this to match against ServiceNow provided software library and thus come back as Normalized or at least Partly normalized.
Could somebody help me understand why it is not?
I was able to reproduce the issue with the data you've provided in my PDI.
The solution is quite simple, but I hope that the community can provide us with more information about the behavior. Compared to other "MS Project" installations with the same version if found one difference in the wording, which has an impact on the normalization.
Microsoft Project Professional 2016 - en - us (your data)
Microsoft Project Professional 2016 - en-us (SAMP demo data)
When replacing the space between "en - us" the installation and the linked discovery model is normalized as expected.
Perhaps, @Daniel Slocum can give us more insights and tell us why in the first example at least the publisher was not normalized.
Best, Dennis
in a first test in my PDI I could reproduce the problem with the ones you showed. In comparison with other installation data for "MS Project", I noticed the following:
Result with your data is the same - "no match"
Compared to other SAMP demo data, it seems to be that the "correct spelling" of the display name is the solution
Microsoft Project Professional 2016 - en - us 16.0.7766.2096 (your data)
Microsoft Project Professional 2016 - en-us 16.0.7766.2096 (SAMP demo data, Difference: no space between "en-us")
Perhaps @Daniel Slocum can give us more details about the logic for discovery models based on installation data in relation to the display name and why in this example at least the manufacturer was not normalized.
Best, Dennis
I was able to reproduce this issue with the your data.
I've compared your data with the SAMP demo data and there is a small difference in the notation.
Microsoft Project Professional 2016 - en - us 16.0.7766.2096 (your data)
Microsoft Project Professional 2016 - en-us 16.0.7766.2096 (SAMP demo data, same model)
The difference is only the space between "en-us". Changing this the data is recognized and normalized correctly
Perhaps, @Daniel Slocum could you please share more details about the normalization process in relation to the "Display name" and why not at least the publisher was normalized.
Thanks and Best,
Dennis
I was able to reproduce the issue with the data you've provided in my PDI.
The solution is quite simple, but I hope that the community can provide us with more information about the behavior. Compared to other "MS Project" installations with the same version if found one difference in the wording, which has an impact on the normalization.
Microsoft Project Professional 2016 - en - us (your data)
Microsoft Project Professional 2016 - en-us (SAMP demo data)
When replacing the space between "en - us" the installation and the linked discovery model is normalized as expected.
Perhaps, @Daniel Slocum can give us more insights and tell us why in the first example at least the publisher was not normalized.
Best, Dennis
