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.
Printed Circuit Assembly
This Field Notice covers any PRE-2 with an RP2 board that has an assembly part number of 73-5544-09 and does not have deviation 72991 installed.
All 800-08916-09 PRE2 assemblies have a marginal timing condition that can result in a variety of memory corruption errors. The primary symptom of the marginal timing condition is a multi-bit ECC error in Main Memory.
The PRE-2 underwent engineer failure analysis (EFA) on return material authorizations (RMAs) from customers experiencing abnormal failure rates. Root Cause indicated a timing margin control circuit upgrade was required on all PRE-2 assemblies due to slow system controller devices. Cisco originally believed that only a select number of these assemblies, identified by serial number, were susceptible to the issue. It has since been discovered, with newer software images that have enhanced features and scalability, that the down rev 800-08916-09 PRE-2 is more susceptible to failures than previously thought.
Timing margins can be sufficiently increased by reconfiguring a phase-locked-loop (PLL) in the system controller device. Configuration is achieved through the use of external resistor pull-ups and pull-downs.
Cisco will upgrade all 800-08916-09 PRE-2 assemblies that have not had the timing fix (deviation 72991) previously installed. To determine if your product is eligible for upgrade, see the How to Identify Hardware Levels section of this Field Notice.
Upgraded units (version -12 or higher) have shipped since mid April 2004 and are not affected.
Single and multi bit ECC errors in main memory. System crashes. Incomplete crash files. Unexplained Bus Errors.
MULTI BIT ECC ERROR INDICATION is the most usual symptom.
The only failure mode that has been proven to be a clear indicator of a the hardware issue is the multi bit ECC error symptom. The ECC information registers near the bottom of the crashinfo indicate whether a multi bit ECC error has occurred. The lower two bits of the ecc_err_address register indicate whether the ECC error was single or multi bit. An "01" indicates a single bit error, a "10" indicates a multi bit error.
In the ECC information example below, the 0xA (1010) in (b/s 0x00011C9A) indicates that this was a multi bit error because the lower two bits are "10".
Affected PRE2-09 boards were returnable under an Umpire Upgrade Program. The program is now expired after a one year extension.
Should an eligible rev -09 board be still in service it can be returned under a standard RMA (fix-on-fail).
How to Identify Hardware Levels
This Field Notice covers any PRE-2 with an RP2 board that has an assembly part number of 73-5544-09 and does NOT have deviation 72991 installed.
Use the CLI show diag command to inspect the Part Number and Deviation Number of the PRE2. In this example the deviation has Not been installed and this unit IS eligible for the upgrade program.
PRE2#show diag Slot B: Primary PRE card RP EEPROM contents: Hardware Revision : 1.0 Part Number : 73-5544-09 <--- assembly part number Board Revision : B0 Deviation Number : 00 <--- deviation number Fab Version : 09 PCB Serial Number : CAT07400EM8 RMA Test History : 00 RMA Number : 0-0-0-0 RMA History : 00 Top Assy. Part Number : 800-08916-09 CLEI Code : IPC810KEAA FP EEPROM contents: Hardware Revision : 1.0 Part Number : 73-5545-04 Board Revision : C0 Deviation Number : 0-0 Fab Version : 04 PCB Serial Number : CAT074208PJ RMA Test History : 00 RMA Number : 0-0-0-0 RMA History : 00
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: