How to add same entry point in multiple Service Maps

[Community Alums]

I have requirement where we have few batch job servers supporting multiple application services.

When we create a Service Map for those Applications, We need to add those Batch Job servers as separate entry points. Since the Batch Job servers does not have any connectivity with Main Application.

To include these batch job servers as part of those applications, I need to add these server entry points in each applications.

We could add different ports at each time. But i suppose this will be a bottleneck in future.

If any one of those Batch job servers has Issue, We need to see the impact in all the Applications on Event Operator workspace.

Please let me know if anyone does this kind of requirements.

[Bill Iffland · Accepted solution]

Hello,

Normally and by design the same EP (class and value) must be unique.

However, I can see 3 possible approaches, assuming you are building Mapped Application Services [cmdb_ci_service_discovered].

1) In addition the HTTP/TCP for the main part of you application service, you can simply add CIs manually, i.e. the windows or linux servers that hosts Batch Jobs. The same CI can be manually created (added) as EP to more than 1 map.

 2a) Create a "re-usable" Manually Created "Mapped Application Service" using the option above and then add the this map as a  "Connected Business Service" using the manually created UI.  This approach can simplify the maintenance of move/add/change.  One consideration is that in Event Management the Connected Service's severity impacts the parent service. Impact tree will show severity, but will need to drill into the child service to see the Alerts.

2b) Instead of Manually Created "Mapped Application Service" use a CMDB Group to collect the Batch Job Servers to be used by Dynamic CI Group.

3) (Advanced)  If you want use the Batch Job "process" instead of the Server, leverage AFP to create a new Application Discovery Pattern to add Batch Job process instances (mirroring Tomcat for example) to CMDB with Horizontal discovery. Then use TCP endpoint as EP.  This will have the process on the map and the Runs on relationship will automatically add server to Map and Impact tree. Use the above approaches to containerize the Batch Jobs.

And depending on the Application Service Groups used for Operator Workspace, Option 2 can be a standalone Service along side your Application Service Maps.  Doing so,  you can tell by inspection of Batch Jobs Service if it is a root cause versus seeing all red on your Application Service Maps.

[Community Alums]

This is really helps. Thank you Bill!.