FCO A0335-1: Proactive - Scheduled: Red phosphorus in the PCI-e cable connecting CMUL to CMUU causes corrosion resulting in a short circuit on the DDC control signal and a system panic.

Asset ID:	1-73-1629497.1
Update Date:	2014-06-27
Keywords:

Solution Type FAB (standard) Sure

Solution 1629497.1 : FCO A0335-1: Proactive - Scheduled: Red phosphorus in the PCI-e cable connecting CMUL to CMUU causes corrosion resulting in a short circuit on the DDC control signal and a system panic.

Applies to:

Fujitsu M10-4S - Version All Versions to All Versions [Release All Releases]
Fujitsu M10-4 - Version All Versions to All Versions [Release All Releases]
Information in this document applies to any platform.
__________
Affected Parts: (FRU/CRU Part Number / Description)

7060817 - M10-4 CMUL CPU Memory Unit Lower
7060911 - M10-4S CMUL CPU Memory Unit Lower

Symptoms

When this issue occurs the server will experience a system panic.

Impact

This corrosion can lead to a system panic, which can not be recovered from until the CMUL is replaced.
This issue will not cause a thermal event.

Changes

Contributing Factors

IMPORTANT! This issue only affects 4 socket M10-4/M10-4S systems (systems that have a CMUU installed and connected to the CMUL).

Cause

Red phosphorus was used as flame retardant for a connector on pci-bp (cable that connects the CMUL to the CMUU). Chemical reaction between red phosphorus and humidity in the air causes corrosion and ionization. Corrosion is causing a short circuit on control signal for the DDC. When this happens the system will crash due to power failures from the DDC.

   Action 1: Isolate red phosphorus, isolation tape has been used at Fujitsu factory.(implemented on Oct-29th, 2013).
   Action 2: Change the flame retardant to bromine at cable vendor. (implemented on Nov-14th, 2013)
   Action 3: Fujitsu implemented the shipment of systems with bromine based connector cable on Nov-20th, 2013.
   Action 4: Fujitsu implemented the shipment of FRUs with bromine based connector cable on Nov-27th, 2013

Services Logistics rework of spares was addressed via GSAP 6233.

Solution

Target Completion Date: December 31, 2014

Workaround

No workaround available - see Resolution section below.

Resolution

Hot Swap? No

Only 4 socket M10-4/M10-4S systems listed in the affected Customer List up on APEX are affected. Many other M10-4/M10-4S systems, not containing a CMUU, are not affected by this issue. Due to limited availability of CMU_L spares, this FCO will be tightly scheduled depending on material availability in each Region. Please work with your Regional FCO Deployment Manager (identified below) for scheduling and a list of impacted systems in your area.

Proactively replace the CMU_L and PCI-e (pci-bp) cable as follows:

   M10-4: Replace CMUL PN 7060817 with new CMUL PN 7086555.
   M10-4S: Replace CMUL PN 7060911 with new CMUL PN 7086557.

Important: Replacing the CMUL requires the system be powered down.

Note: Both the CMUL and the PCI-e cable are included in the FRU.

Refer to M10-4/M10-4S Service Manual - Chapter 7 Maintaining the CPU Memory Units via below URL;

   http://www.fujitsu.com/global/services/computing/server/sparc/downloads/manual/m10-4s/

A Customer Ready Document (Customer Letter) has been developed to help communicate this issue to your customers.

For technical questions about this remediation or if you need to collaborate with TSC note to the following;

. The TSC Group Name is - SN-SPARC: Fujitsu M10
. The support alias is Support_M10_Series_www_grp@oracle.com

The Regional FCO Deployment Managers are

EMEA: mike.netz@oracle.com, jan-erik.olsten@oracle.com
JAPAC: chee-kin.ngai@oracle.com
LACR: marcos.omura@oracle.com, sergio.alonso@oracle.com
NAMER: dennis.cairns@oracle.com

Note: There are no impacted systems in Latin America.

Identification of Affected Parts (how to)

Affected parts can be identified using "showhardconf -M"

        Example:
        XSCF> showhardconf -M
   SPARC M10-4S;
            + Serial:2111203001; Operator_Panel_Switch:Service;
            .....
                CMUL Status:Normal; Ver:0101h; Serial:PP124604CM ;
                    + FRU-Part-Number:CA07361-D941 B9   /7060911              ; <------ Part #

This same information can be found in the snapshot file name xscf_commands/@sp@cli@usr@bin@showhardconf.out.

        Example:
   $ more @sp@cli@usr@bin@showhardconf.out
   SPARC M10-4S;
            + Serial:2111203001; Operator_Panel_Switch:Service;
            .....
                CMUL Status:Normal; Ver:0101h; Serial:PP124604CM ;
                    + FRU-Part-Number:CA07361-D941 B9   /7060911              ; <------ Part #

Hardware Remediation and Material Availability Details

Limited material will be available for this remediation activity. Please coordinate remediations with your Regional Deployment Managers and your local Logistics Manager.

Comments

Proactive FCOs have managed remediations requiring all remediations to be planned through the Regional Deployment Managers.

Several of the affected systems will be remediated directly by Fujitsu as the Authorized Service Provider. David Bayne is the contact for engaging Fujitsu in remediation activities. Fujitsu will open SRs with Oracle only in order to order parts for this FCO.

References

Manual: Fujitsu C120-E716-07EN
ECO: ECO0018356
GSAP: 6233

Contacts

Contributor: mike.cootware@oracle.com
Responsible Engineer: mike.cootware@oracle.com
Responsible Manager: Evan.Piercey@oracle.com

Attachments

This solution has no attachment