Discovery

[Robert P_]

I have a Dell server record that Discovery re-classifies it to Out-of-band device when it discovers the iDrac. It doesn't create two records, it keeps changing the same one from Windows Server to Out-of-Band device.

[debendudas]

Hi @Robert P_ ,

A server might be misclassified as an Out-of-band device if Discovery initially fails to connect using WMI or SSH but successfully connects using SNMP. The SNMP data might then identify it as an iDRAC or similar device, leading to reclassification as an out-of-band device. 

How to Manage Reclassification:

1. Review Discovery Logs:
   Check the Discovery logs to see which protocols and methods are being used to identify the device and why it's being reclassified.
2. Verify Credentials:
   Ensure you have the correct credentials for the primary identification method (e.g., WMI for Windows servers).
3. Check Discovery Schedules:
   Review the discovery schedules to see if there are specific schedules or protocols that are causing the misclassification.

If this solution helps you then, mark it as accepted solution ‌‌✔️ and give thumbs up 👍! 

[AJ-TechTrek]

Hi @Robert P_ ,

As per my understanding , solution will be below and might help.

Why this happens as per my understanding,
* Both the Windows server (class: cmdb_ci_win_server) and the iDRAC (class: cmdb_ci_out_of_band) can report the same serial number or BIOS UUID during Discovery.
* The Identification and Reconciliation Engine (IRE) uses those fields to decide if the data belongs to an existing CI or should create a new one.
* If both devices share the same serial / BIOS UUID, IRE thinks it's the same CI → so it updates / reclassifies the server record to Out-of-band device.
This is why Discovery ends up changing the CI class rather than creating two separate CIs.

How to fix it (ServiceNow best practice):
1. Adjust Identification Rules
Go to Identification Rules:
* For cmdb_ci_win_server (Windows Server)
* For cmdb_ci_out_of_band (iDRAC)
Make sure:
* The Windows Server rule uses:
* BIOS UUID, serial number, and hostname or IP address (production network)
* The Out-of-band device rule uses:
* iDRAC IP address or MAC address
* Avoid relying solely on serial/BIOS UUID for both classes.
This way, IRE can distinguish:
Main server (prod IP, host name)
vs.
iDRAC (mgmt IP, iDRAC MAC)

2. Discover them separately
* Use Discovery Schedules / Discovery Behavior to:
* Discover Windows server on production IPs.
* Discover iDRAC on management subnet or dedicated IPs.
* Never discover the iDRAC and production interface in the same sweep if possible.

3. Patterns and classification
* Make sure the classification logic in patterns is correct:
* Windows server pattern should classify device as cmdb_ci_win_server.
* iDRAC SNMP or WS-MAN pattern should classify device as cmdb_ci_out_of_band.
* In the iDRAC pattern → ensure it only applies to the mgmt IP.

4. Add relationship
* In Discovery patterns or via post-processing:
* Add a Manages or Managed by relationship:
* Out-of-band device → Manages → Windows Server
* This way, you keep them as separate CIs but logically connected.

5. Validate
* Run Discovery first on production IP → confirm you get Windows Server CI.
* Then run Discovery on iDRAC IP → confirm you get Out-of-band CI.
* Confirm two separate sys_ids, and the class doesn’t flip.

Summary / Checklist:
1. Adjust Identification Rules → use IP/MAC + hostname to distinguish.


2. Discover iDRAC and server separately.


3. Check classification in patterns.


4. Add relationship between the CIs.


5. Validate in lower environment before deploying.

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