@Nisha30 We did couple of tag-based Service mapping. Here are the details.
Step 1: All the CI's to be mapped had to be discovered first. For example, there is a Service for 'People soft' for PRODUCTION environment and there are 3 servers running for it ( let's name: Server 1, Server 2 and Server 3).
Step 2: After these are discovered, make sure, their respective Key Values are also imported OR auto-populated. If you check with the application (Peoplesoft team), there are some tags mapped for every Server in in the infra (These are available readily for cloud based ones.) Here is an example:
These values have to be brought into Servicenow cmdb_key_value.LIST table.
Step3: CI Tag Categories need to be created. This is a kind of normalisation of the key values. In this example, anything with 'App', 'Application','Appl','AppName' will be considered as 'Application'.
Step 4: Create 'Tag based Service Families' With a meaningful name and add the above created CI Tag Categories to this family.
Step 5: Now the 'Manually update candidates' related link to be clicked.
Step 6: There will be a pop-up window with the Service name as 'prd:peoplesoft'. It's the environment-application combination (families we selected in Step 4).
Select that particular application by clicking on check box and then select 'Map selected' button.
Step 7: Click on OK
Step 8: Now a new section named 'Mapped Services' would show-up. click on the link 'View map'.
Step 9: A map would be loaded with only the entry point. Read the information message, which is asking to re-calculate the Service.
Click on the gear icon to go back to application.
Step 10: Now, click on the 'Recalculate Service' related link.
Step 11: After the page refreshes, click on 'View Map' button.
Step 12: You can see the application 'prd:peoplesoft' as the entry point and 3 Servers mapped underneath.
Tag based mapping is easily understood as, considering each CI as JSON Object with various Key-value pairs in it. The map purpose is to connect these Objects.
I can't write blogs, so made all the steps here. Do let me know if you are struck in any step.
@Nisha30 We did couple of tag-based Service mapping. Here are the details.
Step 1: All the CI's to be mapped had to be discovered first. For example, there is a Service for 'People soft' for PRODUCTION environment and there are 3 servers running for it ( let's name: Server 1, Server 2 and Server 3).
Step 2: After these are discovered, make sure, their respective Key Values are also imported OR auto-populated. If you check with the application (Peoplesoft team), there are some tags mapped for every Server in in the infra (These are available readily for cloud based ones.) Here is an example:
These values have to be brought into Servicenow cmdb_key_value.LIST table.
Step3: CI Tag Categories need to be created. This is a kind of normalisation of the key values. In this example, anything with 'App', 'Application','Appl','AppName' will be considered as 'Application'.
Step 4: Create 'Tag based Service Families' With a meaningful name and add the above created CI Tag Categories to this family.
Step 5: Now the 'Manually update candidates' related link to be clicked.
Step 6: There will be a pop-up window with the Service name as 'prd:peoplesoft'. It's the environment-application combination (families we selected in Step 4).
Select that particular application by clicking on check box and then select 'Map selected' button.
Step 7: Click on OK
Step 8: Now a new section named 'Mapped Services' would show-up. click on the link 'View map'.
Step 9: A map would be loaded with only the entry point. Read the information message, which is asking to re-calculate the Service.
Click on the gear icon to go back to application.
Step 10: Now, click on the 'Recalculate Service' related link.
Step 11: After the page refreshes, click on 'View Map' button.
Step 12: You can see the application 'prd:peoplesoft' as the entry point and 3 Servers mapped underneath.
Tag based mapping is easily understood as, considering each CI as JSON Object with various Key-value pairs in it. The map purpose is to connect these Objects.
I can't write blogs, so made all the steps here. Do let me know if you are struck in any step.
Hi @Mannapuram This was so detailed and explaining in such a step by step really i got now the concept .
Thanks a million for this explanation. Really helped in few minutes.
Just a question: In cmdb_key_value.LIST
(If I understood correctly) all possible values in infrastructure has to me made entered . Example: we have Environment = Dev, QA, Test, Prod, so this means i have to entry in below way.
Since within Infrastructure the team has somewhere called "Env" and some where "Environment".
CI can be empty here ?? since how do i know what CI will be discovered prior running discovery. There will be thousands of CIs discovered after running discovery. OR is it post discovery runs this table is auto populated by itself . Then we start tagging by looking what has been discovered?
SO all possible entries to be present in key table
Thanks
@Nisha30 values in cmdb_key_value.LIST can be manually entered but the CI should never be empty. For each CI - there are multiple entries in this table for various Key-Value pairs. For the above example,
Server 2 - is a CI from CMDB - this is having entries with 2 Key-value pairs ( 1. Environment - Prod, 2. App - Peoplesoft).
But Ideal way to fill this table would be to do it as part of filling child tables in transformation after discovery runs. This table can be manually filled/automatically filled only after the required CI's are discovered.
The Screen shot you pasted is only worried about variations in the Key - like Environment or Env. This is taken care when creating category (Step 3). The screen shot should be similar to below for each CI.
Out of thousands of CI's, you can search for your application name in 'Value' if this table is automatically updated. If you are doing manual entries, then you need to know all the CI's related to this application. Usually, this is not possible, that's the reason, we try to pull these details from the infra.
You can get the name of the CI from Infra and compare it with the name in CMDB. You will get the exact CI.
