Find your people. Pick a challenge. Ship something real. The CreatorCon Hackathon is coming to the Community Pavilion for one epic night. Every skill level, every role welcome. Join us on May 5th and learn more here.

Creating Software Installation Records

RahulY00040
Tera Contributor

3 weeks ago

Hello everyone,

 

I’ve come across an interesting integration scenario and would like your input. We are receiving device details along with associated software installation data from a source system. Each device can have 100+ software installations. This data is being ingested into a staging table via the integration.

Our objective is to:

  1. Identify or match the CI using device attributes such as Serial Number, FQDN, and related identifiers.
  2. Once the CI is successfully identified, create the corresponding Software Installation records in the cmdb_sam_sw_install table.

 

 

Since we will have one row per software installation, I want to avoid repeated IRE calls to retrieve the device sys_id for each record. Does anyone have suggestions on how to handle this more efficiently?

My current approach is as follows:

  1. Configure the Import Set to run asynchronously.
  2. Execute a Scheduled Job first to:
    • Group records based on device identifiers such as Serial Number and FQDN.
    • Make a single IRE call per device to retrieve the corresponding sys_id.
    • Update that sys_id on all related rows in the staging table.
  3. Once the above step is completed, trigger the Transform Job to map and insert the data into the target table.

I’d appreciate your review and any feedback or alternative suggestions.

 

Thanks in advance,

R

0 Helpfuls
  • 860 Views
6 REPLIES 6

Naveen20
ServiceNow Employee

3 weeks ago

Break into batches of 10K–20K rows per import set and setWorkflow false to avoid BR's

RahulY00040
Tera Contributor
In response to Naveen20

2 weeks ago

Thanks for your responses, Naveen. Alternately, I was thinking of letting IMPORT sets run and all records would be ignored (I will condition it to ignore if it's missing the sys_id), later, configured scheduled Job (to lookup CIs using IRE) will update the rows with sys_ids and then rerunning the transform. What you think?

 

Regards,

Rahul

0 Helpfuls