Precedence sets defines the precedence value for a dataset in a Merge activity. Each source dataset is paired with one, and that Precedence set serves as the default precedence value for all classes and attributes in that dataset. The highest value among all source datasets determines the data that is written to the target dataset.
Within a Precedence Set, you can specify individual Precedences for either a class or an attribute, assigning them a precedence value higher or lower than that of the dataset. This prevents one source dataset from providing all the data in a merge. Class-level Precedences apply to all attributes of the class.
In the scenario mentioned - CI is created from Receiving Console with Name as abc and TokenId as 123 which makes it to ASSET <Production Dataset> as it is (Name - abc and TokenId - 123)
The above CI in ASSET is going to get a RE-ID - <example only RExxxxx001>
Now ,lets say the same instance/CI is discovered by some discovery mechanism and brings this CI in discovery dataset however with a different Name as xyz and TokenId as 123 and that there is RE job configured to reconcile CIs from Discovery dataset into ASSET .
1) Based on the Identification rules, RE would lookup CIs in Target dataset <ASSET> the discovery CI with TokenId - 123 gets assigned the RE-ID RExxxxx001 as there is match found based on TokenId <Identification Rule qualification based on TokenId>>
2) Now , during merge activity RE is going to look up for precedence weights assigned at dataset,class and attribute level to update values in Target dataset.
With Overall Precedence on ASSET assumed to be at 100 and that of Discovery set to 500
Name abc gets overwritten with xyz from discovery side
With Overall Precedence on ASSET assumed to be at 100 and for that of Discovery set to 50
Name abc gets retained in ASSET and xyz from discovery side gets ignored based on precedence .
However, the later would prevent discovery to update any values for existing matches found in ASSET Merge Precedences help precisely to choose and allow the data that needs to flow from one dataset<source which can be multiple> to another<Target>
so, setting attribute level precedence on Name and assigning it a weightage more than 100 will allow name to flow from discovery to ASSET
likewise , if you want name to retained in ASSET and rest of the values to be updated by discovery for such CIs which are in ASSET that made it through other sources you can set overall precendence for discovery more than 100 and set attribute level precedence for name lesser than 100 .
As you mentioned in your post ...
"In my case, I create some CIs by Receiving Console with a default CI Name, but after the CI goes into production (deployment)"
Not sure how did you create a CI into ASSET in first place which had the default name for the CIs.
If you have created a CI into ASSET with some manual Dataset then you will have to modify the precedence in such a way that the discovery dataset has higher priority then the manual dataset using which the CI was pushed into ASSET at first place.
Hi Anand Iyer, this is the idea, but, I'm a little bit lost with the steps to configure this precedence, could you help me?
Hi Anand Iyer, I don't know why, but I only recreated the merge activity, with the same rules, and now, it works, thank you for your help.