CI Classification Issue in CMDB – Virtual Machine Identified as Deployed Marketplace Product
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
07-16-2025 03:48 AM
Hello Community,
We are part of the Discovery team and have encountered an issue during CI classification in CMDB.
We received a change task (ctask) where the Configuration Item (CI) is a virtual machine. This was confirmed through the Discovery logs, which correctly identified the CI as a virtual machine instance. However, upon reviewing the CMDB, we noticed that the CI is classified under the "Deployed Marketplace Product" class instead of "Virtual Machine Instance".
We would like to understand:
- What could be the possible reasons for this misclassification?
- Are there specific identification or classification rules that might be causing this behavior?
- As the Discovery team, what steps can we take to correct this and ensure accurate CI classification going forward?
Any insights or guidance on how to troubleshoot and resolve this issue would be greatly appreciated.
Thank you!
- Labels:
-
Orchestration (ITOM)
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
07-31-2025 07:43 AM
Hi @Shraddha Dube ,
As per my understanding what’s happening (summary):
* Discovery successfully identifies the device as a virtual machine in the logs.
* But in CMDB, it gets created (or updated) with class = Deployed Marketplace Product instead of Virtual Machine Instance.
This is almost always due to:
* How classification + identification work together.
* Incorrect or missing data → misfires the classification phase.
Possible reasons for misclassification
1. Classification pattern or rules mismatch
* Classification is based on:
* Patterns
* Probes (for older discovery)
* Specific attribute values or OIDs
* If the pattern logic wrongly matches some field (e.g., a marketplace tag, AMI ID, image name) → the system chooses the wrong class.
2. Identification Rule issue / CMDB Hierarchy
* Identification rules + CMDB hierarchy determine where in the class hierarchy the CI goes.
* If your identification rule or reconciliation logic:
* Matches an existing CI in Deployed Marketplace Product class
* Or doesn’t have enough attributes to choose between subclasses
→ it keeps or moves the CI into the wrong class.
3. Marketplace data or tag in the discovered data
* If the discovered payload has attributes like:
* ami-id
* image-id
* marketplace-product-code
* The pattern may see this and wrongly classify as Deployed Marketplace Product.
Steps to troubleshoot & fix
Step 1: Check classification logs
In Discovery logs (classification phase):
* Find out:
* What criteria caused the system to pick Deployed Marketplace Product?
* Check for:
* Classification: trying pattern.
* Classification: selected class.
Step 2: Review the classification pattern
Go to:
Discovery → Patterns → Classification patterns
Find the one matching Virtual Machine Instance and Deployed Marketplace Product.
* See if:
* There’s an overly broad pattern that triggers on fields like AMI ID.
* The VM pattern is too narrow or missing conditions.
Step 3: Check Identification Rule
Go to:
Identification Rules → cmdb_ci_vm_instance
* Confirm:
* It’s active.
* It has correct identifying attributes (e.g., instance_id, name, serial number).
* If the wrong rule is matching (e.g., for marketplace product), update / adjust.
Step 4: Check existing CI history
If the CI already exists in CMDB as Deployed Marketplace Product:
* Discovery / IRE may be updating that record instead of creating a new Virtual Machine Instance.
* Check sys_audit or sys_history on the CI:
* Was it first created as marketplace product?
Step 5: Fix / improve patterns
* Adjust classification pattern:
* Add conditions: only classify as marketplace product if certain strong attributes exist.
* Adjust VM pattern:
* Broaden to include your vendor's specific image or tag.
Step 6: Retest
* Run Debug Discovery Pattern:
* Choose the CI's IP and MID Server.
* Confirm:
* Classification → correct class chosen.
* Exploration → correct data populated.
As Discovery team, what you can do:
1. Analyze logs: Understand why classification is picking the wrong class.
2. Update / refine patterns: Adjust pattern logic to make classification smarter.
3. Review identification rules: Ensure correct attributes and class hierarchy.
4. Validate before production: Test with Debug Pattern before re-running large discovery.
5. Communicate with CMDB data team: Align on what data is required to distinguish virtual machines vs marketplace products.
Please appreciate the efforts of community contributors by marking appropriate response as Mark my Answer Helpful or Accept Solution this may help other community users to follow correct solution in future.
Thank You
AJ - TechTrek with AJ - ITOM Trainer
LinkedIn:- https://www.linkedin.com/in/ajay-kumar-66a91385/
YouTube:- https://www.youtube.com/@learnitomwithaj
Topmate:- https://topmate.io/aj_techtrekwithaj (Connect for 1-1 Session)
ServiceNow Community MVP 2025