Field Notice: FN - 64123 - B200-M4 with UCSB-MRAID12G Might Unexpectedly Reboot. Software Upgrade Recommended.

NOTICE:

THIS FIELD NOTICE IS PROVIDED ON AN "AS IS" BASIS AND DOES NOT IMPLY ANY KIND OF GUARANTEE OR WARRANTY, INCLUDING THE WARRANTY OF MERCHANTABILITY. YOUR USE OF THE INFORMATION ON THE FIELD NOTICE OR MATERIALS LINKED FROM THE FIELD NOTICE IS AT YOUR OWN RISK. CISCO RESERVES THE RIGHT TO CHANGE OR UPDATE THIS FIELD NOTICE AT ANY TIME.

Revision History

Products Affected

Problem Description

A B200-M4 blade equipped with a UCSB-MRAID12G RAID controller might unexpectedly be rebooted by Unified Computing System Manager (UCSM).

Background

Shallow discovery - A nondisruptive inventory check of the blade with use of Out Of Band (OOB) communications.

Deep discovery - A disruptive inventory check of the blade which requires a boot to the Processor Node Utility Operating System (PNuOS) preboot environment.

In normal operation, hardware information is periodically collected from each blade and component. While the information is collected, UCSM might receive invalid data from the MRAID12G controller and store the Vendor and Model attributes as blank values. 

When a shallow discovery occurs on a blade in this state, UCSM might read the correct Vendor and Model of the RAID controller. The difference between the stored and newly read Vendor and Model attribute is flagged as a hardware change and results in a disruptive deep discovery of the blade.

Common operations that will cause a shallow discovery include:

• UCSM upgrade 
  
• UCSM cluster lead operation 
  
• Cisco Integrated Management Controller (CIMC) reset 
  
• Input/Output Module (IOM) is reset or replaced 
  
• The cable between an IOM and the Fabric Interconnect (FI) is pulled/replaced 
  
• FI is rebooted

Problem Symptoms

Prior to the unexpected reboot, output from the show server inventory expand command output is missing Vendor and Model attributes. All other attributes will be populated: 

   RAID Controller 1:  
            Type: SAS  
            Vendor:  
            Model:  
            Serial: LSVXXXXXXXX  
            HW Revision: C0  
            PCI Addr: 01:00.0  
            Raid Support: RAID0, RAID1  
            OOB Interface Supported: Yes  
            Rebuild Rate: 30  
            Controller Status: Optimal

After a deep discovery completes, the UCSM show server inventory expand command output now shows the correct Vendor and Model attributes:

  RAID Controller 1:  
           Type: SAS  
           Vendor: Cisco Systems Inc  
           Model: UCSB-MRAID12G  
           Serial: LSVXXXXXXXX  
           HW Revision: C0  
           PCI Addr: 01:00.0  
           Raid Support: RAID0, RAID1  
           OOB Interface Supported: Yes  
           Rebuild Rate: 30  
           Controller Status: Optimal

If an unexpected reboot has already occurred, this condition can be confirmed via the Data Management Engine (DME) process logs:

Note: The UCSB-MRAID12G RAID controller is an optional/removable component on the B200-M4. If no attributes are populated in the show server inventory expand command output for Raid Controller 1, then it is not installed and you are not impacted.

Example:

RAID Controller 1:  
            Type: SAS  
            Vendor:  
            Model:  
            Serial:  
            HW Revision:  
            PCI Addr:  
            Raid Support:  
            OOB Interface Supported: No

Workaround/Solution

In order to prevent incorrect Vendor and Model attributes from causing additional deep discovery reboots, you must upgrade the UCSM infrastructure firmware to a fixed release.

When you upgrade from an impacted release to a fixed release, blades must be verified for this condition. If Vendor and Model attributes are missing, the UCSM upgrade will still trigger a deep discovery reboot. It is important to verify all Vendor and Model attributes are present directly prior to an upgrade.

This issue is fixed in UCSM Infrastructure bundle 2.2(3g), 2.2(4b), 2.2(5a), 3.1(1e) and later releases.

Users can run this UCS powertool cmdlet in order to get a list of B200 M4 blades and check whether UCSM is missing Vendor and Model attributes for the storage controller.

Prerequisite

Install PowerShell and the UCSM PowerShell module on a Microsoft Windows system. See Cisco UCS PowerTool Suite - PowerShell Modules for Cisco UCS Manager, Cisco IMC, UCS Central.

Launch PowerShell, connect to UCSM, get the list of all B200 M4 blades with Storage controller details, and disconnect the session.

From the returned output, blades which do not have the Vendor and Model attributes populated are susceptible to go through an unexpected deep discovery reboot during the next shallow discovery.

If a Windows host with PowerShell is unavailable, Secure Shell (SSH) to the primary FI, enter this command, and review for any blank Vendor and Model attributes:

show server inventory storage | grep n 18 p 2 "UCSB-B200-M4"

For affected blades, perform a decommission, and re-acknowledge the server. After the server completes discovery, repeat this process in order to verify if UCSM is updated with proper storage controller attributes.

CDETS

To follow the bug ID link below and see detailed bug information, you must be a registered customer and you must be logged in.

For More Information

If you require further assistance, or if you have any further questions regarding this field notice, please contact the Cisco Systems Technical Assistance Center (TAC) by one of the following methods:

• Open a service request on Cisco.com
• By email
• By telephone

Receive Email Notification For New Field Notices

Cisco Notification Service—Set up a profile to receive email updates about reliability, safety, network security, and end-of-sale issues for the Cisco products you specify.