Power6 High-End Power Subsystem Firmware

Applies to:  9125-F2A

This document provides information about the installation of Licensed Machine or Licensed Internal Code, which is sometimes referred to generically as microcode or firmware.

Contents

1.0 Systems Affected

This package provides firmware for Power 575 (9125-F2A)  Servers only.  Do not use on any other systems.
 

The firmware level in this package is:

2.0 Important Information

HMC-Managed Systems

This firmware level requires  HMC V7  R3.4.0.

Go to the following URL to access the HMC code packages:

NOTE:   You must be logged in as hscroot in order for the firmware installation to complete correctly.


Installing Power Subsystem Firmware and System Firmware

Atention: 9125-F2A servers should be evaluated for ECA845 installation prior to performing firmware upgrades. Contact your service provider for more information about ECA845. If firmware must be upgraded prior to the installation of ECA845, use the following special instructions:

Firmware installations should be performed on an entire managed frame, and all managed systems
housed in that frame, at one time.  The following instructions can be used to install firmware on
one managed frame and all of the managed systems that it houses.  The instructions can be
repeated as many times as needed until all firmware has been installed on all managed frames.

1) Power off all managed 9125-F2A systems in the managed frame.

2) Reinstall the current system firmware level on all managed systems in the managed frame:
   
   The following command can be used to determine the current firmware level:

      lslic -t sys -m <managedsystem_1> -F activated_level

   The following command will retrieve and reinstall the current firmware level on
   one managed system:

      updlic -o a -m <managedsystem_1> -t sys -l <current_fsp_level> -r <repos>
      
      <current_fsp_level> is the level determined with the lslic command.

      <repos> is the location of the firmware such as "-r dvd", "-r ibmwebsite", etc.
      Additional parameters might be required, depending on the repository selection.
      If the firmware has already been copied onto the HMC, the HMC hard drive (-r disk)
      should be used as the repository.

   After the firmware has been copied onto the HMC, the HMC hard drive (-r disk) should
   be used as the repository to update the remaining managed systems:

      updlic -o a -m <managedsystem_2> -t sys -l <current_fsp_level> -r disk
      updlic -o a -m <managedsystem_3> -t sys -l <current_fsp_level> -r disk
      ......
      updlic -o a -m <managedsystem_N> -t sys -l <current_fsp_level> -r disk

   These commands can be run in the background in parallel to speed up the installation of the firmware.

   Wait for all updlic commands to complete before proceeding to step 3.

3) Clear any locks that are held on the managed frame by issuing this command:

      rmlock -e <managedframe>
   
4) Reboot the HMC.

5) Install and activate new firmware on the managed frame and all managed systems in that frame.
   
   Using the HMC GUI interface, click on "Updates" in the navigation (left-hand) pane.  Click on the
   checkbox beside all managed systems in the same managed frame, and click on "Upgrade Licensed
   Internal Code to a new release".

   After clicking "OK" on the readiness check panel, select the repository where the firmware files
   can be found and click "OK".  Select the new managed frame (02EP340_041) and managed system
   (01ES340_042) firmware levels from the selection panels that are displayed.
   
   Accept the license agreement and click on "OK" in the confirmation panels to start the firmware installation.  
   
6) Power on the managed systems.

3.0 Firmware Information and Description

Use the following example as a reference to determine whether your installation will be concurrent or disruptive.

Note:  The file names and service pack levels used in the following examples are for clarification only, and are not
             necessarily levels that have been, or will be released.

Power Subsystem firmware file naming convention:

     02EPXXX_YYY_ZZZ

NOTE:  Values of service pack and last disruptive service pack  level (YYY and ZZZ) are only unique
within a release level (XXX).  For example, 02EP330_067_045 and 02EP340_067_053 are different
service packs.

An installation is disruptive if:

              Example:  Currently installed release is EP330, new release is EP340                Example:  EP330_120_120 is disruptive, no matter what level of EP330  is currently
                                   installed on the system                  Example:  Currently installed service pack is EP330_120_120 and
                                     new service pack is EP330_152_130

An installation is concurrent if:

              Example: Currently installed service pack  is EP330_126_120,
                                 new service pack is EP330_143_120.
 

Firmware Information and Update Description

 
Filename Size Checksum
02EP340_132_041.rpm 14,336,531 16693
 
EP340_132_041

12/1/2010
 

 Impact:  Availability        Severity:  SPE

Changes:
  • Fix System Attention LED still flashing after Tres DASD repair
  • Fix lscfg command sometimes not showing all Tres power supplies
  • Fix part numbers for non-comminicating parts not showing up in error logs
EP340_123_041

05/19/2010
 

 Impact:  Availability        Severity:  ATT

Changes:
  • Change Tres DCA reboot defective callout threshold (from 2 total to >12 reboots in 24hrs)
  • Prevent levels disabled SRC after Tres DCA reboot
  • Turn off current balancing for Tres DCA level 7 to eliminate erroneous defective DCA SRCs
  • Fix SRC 140131BB called out wrong MDA
EP340_120_041

04/07/2010
 

 Impact:  Usability        Severity:  SPE

Changes:
  • Automatic recovery improvements of Tres I/O drawer DCA after reboot
  • Fail DCA-T24 redundancy check if older midplane PN (support for midplane pin length problem)
  • SRC 14011305 can occur in the window between when the planar is "intended on", but before the DASD is fully turned on
  • Refcodes 14012052 and 14010063 posted when DCA is fine but MDA is broken
  • Warning SRCs were preventing critical SRCs from being posted on the power/thermal subsystem
  • SRC 14012308 shows xxxx for part number and #### for MTM and serial number
  • No SRC posted for IBF Circuit Breaker in off position
  • False 14011302 I/O Planar SRCs Against B&C-IB
  • Fix calling out invalid fru pn
  • BPC reset caused by interleaving of fan control w/ verify master routine
  • Log Data in 140100F6 Informational SRC is Corrupted (don't put out a log for this SRC)
  • Memory leak in plic during FRU code updates
  • Cable callouts for BPA to DCA has one point too many to check
  • Location codes in R&V Repair panel not applicable to UPIC cable at fault
  • Shorted CEC DCA Fails Redundancy Check Due to Tripped SCBs
  • UEPO FRU Exchange, BPA elementLevelDeactivate failure rc=2816 (bug in EPO location code decoder)
  • Dual access errors for MDA-W are not posted
  • FIELD System failed with SRC 14011318 calling out incorrect P/N for Enigma Card
  • After HMC CCM Code update to Rel F336.1 obtained SRC 14011302 (skip EDFI on planar if planar type is zero)
  • B1813028 while 333->336 (improve procedure for cleaning up fifo to prevent leftover data)
  • Getting Location Translation code Failed while doing fill and drain (-C1 loc code suffix not being recognized)
  • SRC B181B108 reported while upgrading from 335 to 336 (BPC cmd response data corruption in plugin)
  • Tres Drawer Fan P/N is incorrect in call home PMH
  • After BPC F351.1 Code loaded on system via HMC CCM - SPCN can not talk to BPC's
  • BPC F351.2 Tres port card part number drops rightmost character in PEL
EP340_112_041

12/16/2009
 

 Impact:  Serviceability        Severity:  SPE

Changes:
  • Fix memory throttling operation
  • Enhancements to master DCA recovery during DCA servicing and code updating
  • Fix fault isolation for overvoltages due to voltage distribution shorts
  • Eliminate erroneous error reporting when the EPO switch is cycled quickly
  • Improve fault isolation for DCA shorted oring diode
  • Enhancements to master DCA recovery during BPA servicing
  • Eliminate erroneous SRC for Low Room Temperature and BPC reset during BPC service
  • Add BPA logs to CEC error logs to enhance failure analysis
  • Keep certain informational SRCs from blocking non-informational SRCs
  • Fix problem with concurrent add of drawer
  • Invalid part number for DCA callout
  • Redundancy check response toggling on DCA
  • I2C bus conflicts on I/O drawer
  • New I/O drawer command to Reset "other" I2C bus from planar to DCA
  • Prevent BPC reset if other BPC is not available
  • Some I/O drawer voltage levels are not turning on if master DCA has not been determined
  • Water pump firmware workarounds for motor decoupling from pump
  • Power off of B&C IB fails to power off 1/2 after planar failure on one planar
  • Fix erroneous defective BPR posting for loss of AC on one of two line cords on IH system
  • BPA fans go to high speed after loss of AC recovery
  • DCA s/n not listed in call home PMH or PEL data
  • Secondary cage/fru incorrect for SRC 14027098/reasonCode : 004f
  • Change Health Check to ignore 14019999 informational SRCs
  • Concurrent add of BPR/BPD fails
  • Invalid MDA PN 41V2522 in error logs
  • 12V Stby voltge glitch between concurrent DCA code updates
  • Improve I2C bus arbitration on I/O drawer
  • No part number and enclosure vital product data included in 14012308 SRC
  • DCA left partially code upated during concurrent code update
  • Invalid fru default part numbers for several parts (DCA, I/O midplane, MWU)
  • SRC 14010400 posted during code update
  • BPC deactivate causes reset of BPC
  • Location code for AMD/MDA UPIC incorrect and panel lacks illustration
  • Cable callouts for BPA to DCA has one point too many to check
  • Location codes in R&V Repair panel not applicable to UPIC cable at fault
EP340_090_041

08/20/2009
 

 Impact:  Serviceability        Severity:  SPE

Changes:
  • Change to prevent Service Processor crashes due to problem where defective Power Supply resets repeatedly
  • Fix for not reporting defective Bulk Power Regulators with internal communication problems
  • Fix for an intermittent problem where a change to the I/O Drawer ownership results in false 100090F2 SRC
  • Change to improve recovery handling of 5803 I/O Drawer power supplies
  • Fix for Bulk Power Fans running at high speed due to informational SRC being present in the Bulk Power Controller
  • Fix for problem that prevent the concurrent replacement of a unresponsive CEC power supply
  • Fix for problem causing false 1401131x SRCs
  • Fix for problem that leads to degraded performance on 5791 I/O drawers
EP340_078_041

06/22/2009
 

 Impact:  Function       Severity:  SPE

New Features and Functions:
  • Support for 5803-001 I/O Drawer
Changes:
  • Fix to report planar power on failures in Infiniband drawers)
  • Fix to eliminate multple SRCs for P1-C1-J04 cable unseat
  • Fix to post correct SRC for Bulk Power Distributor port fault
  • Fix to prevent both Power Supplies being called defective for Critical Internal Overtemperature
  • Fix to only post SRC 1401008x When All BPRs on Same Line Cord Have Phase Fault
  • Fix to make behavior consistent when deactivating Bulk Power Controller from opposite side
  • Change to retry writing voltage setpoint data to power supplies on intermittent communication failures
  • Fix to post correct failure code for Critical Overvoltage fault on both power supplies
  • Fix to prevent erroneous SRC 14012C51 after install and activate of 9125-F2A node
  • Fix for erroneous SRC 100090F2 during firmware update
  • Fix SRC reason code description errors
  • Fix to prevent erroneous SRC 14010083/4 (Loss of Input Power TO BPA) on 9125-F2A with less than 4 Bulk Power Regulators
  • Fix to ensure coreect status reporting of firmware update failures in 9125-F2A power supplies
  • Change to turn off 9125-F2A power supplies for overtemperature and not for a fan fault
  • Fix to prevent indicating I/O drawer being reported as UNKNOWN_IO type
  • Improve logging of command history in Bulk Power Controller error logs
  • Fix for Service Guide calling out wrong side of frame for SRC 14017195
  • Change to post 14018131 SRC after UEPO switch is used
  • Fix to remove cross power cable from redundancy check considerations to improve serviceability
  • Fix to prevent fast UEPO cycles from posting defective power supply SRCs
  • Fix to post SRC for dead/missing Bulk Power Fan controller
  • Fix for incorrect SRC description message for SRC 14010361
  • Fix for incorrect FRU on SRC for undervoltage on a single power supply
  • Enhancement to power supply error detection
  • Fix to add more delay before checking FRU response after activation to prevent error codes during service
  • Fix to prevent fails during deactivate of Modular Water Unit
  • Fix to prevent erroneous Low Low Water SRC against Modular Water Unit after code update
  • Add miscable checking for Modular Water Units
  • Fix to prevent incorrect SRC 14012a85 after P1-C1-J04 cable pull
  • Fix to prevent posting of extraneous SRCs during power line disturbances
  • Fix to prevent erronous posting of SRC 14012051 for power supply cabling
  • Fix to prevent BPC reboot from malformed LED cmd
  • Fix to allow firmware update to Bulk Power components in large 9125-F2A configurations
  • Fix for incorrect customer voltage and current readings displayed on the HMC
  • Fix to prevent erroneous SRC 14012052 during System Power Off
  • Fix for incorrect 9125-F2A SRC Posted for defective node fan
  • Fix to prevent extraneous SRC 1401036x for Fan Inlet Critical fault
  • Fix for incorrect status of Bulk Power Hub health indicator after firmware update
  • Fix to update enclosure VPD cache on both BPCs to prevent invalid readings
  • Fix to allow 9125-F2A node to power on after node maintenance
EP340_041_041

01/16/2009

Impact:  Function        Severity:  ATT

New Features and Functions:
  • Support for concurrent processor node addition, as well as cold node repair or removal.
  • Enhanced fault reporting for the Integrated Battery Feature.
  • Overcurrent protection for the driver card on the Fill/Drain Tool.
  • Modular Water Unit performance enhancements

Other Changes:
  • Fixes for false System Reference Codes during UEPO cycling.
  • Fixes for false System Reference Codes during AC power outages.
  • Fix for I/O fans not going to the proper speed after concurrent I/O power supply replacement.
  • Fixes for false System Reference Codes during Concurrent Firmware Maintenance.
  • Fixes for various other false System Reference Codes.
  • Fix for problem syncing the HMC Service Lock between both Bulk Power Controllers.
  • Fix for BPC reset/reload problem on frames with large F2A configurations.
  • Fix for F2A CEC power supply failure during Concurrent Firmware Maintenance.
  • Fix for failing false 'transition to Low Power Mode failed' message when AC power is lost on one side.

4.0 How to Determine Currently Installed Power Subsystem Firmware Levels

You can view the frame's current firmware level on the Advanced System Management Interface (ASMI) Welcome pane.  It appears in the top right corner.   Example:  EP330_029.

5.0 Downloading the Firmware Package

Follow the instructions on the web page. You must read and agree to the license agreement to obtain the firmware packages.

Note: If your HMC is not internet-connected you will need to download the new firmware level to a CD-ROM or ftp server.

6.0 Installing the Firmware

The method used to install new firmware will depend on the release level of firmware which is currently installed on your server. The release level can be determined by the prefix of the new firmware's filename.

Example: EPXXX_YYY_ZZZ

Where XXX =  release level

Instructions for installing firmware updates and upgrades can be found at  http://publib.boulder.ibm.com/infocenter/systems/scope/hw/topic/ipha1/updateschapter.htm