Hardware Management Console

[ Last updated: November 21, 2008 ]

The information in this Readme contains hints and errata information about the Hardware Management Console.

• Special upgrade instructions for HMC 6.1.0
• PTF-specific information
• HMC 6.1.3 Update package
• Enhancements and changes in HMC 6.1.2
• Enhancements and changes in HMC 6.1
• Security fixes in HMC 6.1.2
• National Language Support
• Known issues in HMC 6.1.0
• Security fixes in HMC 6.1.0
• Hints and tips for the HMC Desktop
• Education information

You cannot use the Install Corrective Service task to update to HMC V6.1.0. Instead, you must use one of the following upgrade methods. In addition, the only method by which any Version 4 HMC can be upgraded is the "Local media - DVD-RAM Upgrade Data" method.

HMC upgrade methods

Remote upgrade

Network upgrade

Local media - Hard drive Upgrade Data

Local media - DVD-RAM Upgrade Data

BIOS updates

If you have the following HMC machine types, you are required to obtain and install the latest BIOS, prior to upgrading to HMC V6R1.0:

In addition, IBM strongly recommends upgrading to the latest BIOS for your HMC model, regardless of the version of the HMC code in use.

To obtain the latest BIOS for the above machine types, please see the HMC Support for HMC BIOS updates pages.

This section lists the PTFs released for HMC V6 R1.x. The information for each PTF includes sections for enhancements and fixes, known issues (if applicable), and package information. The Package information section provides information to use during the download, installation and verification procedures for HMC corrective service/upgrade packages. You can also access this PTF-specific information by clicking on the "View" link for any package on the "Downloads" pages of the HMC web site.

Fix for HMC Version 6 Release 1.3.0 Service Pack 0.

This package includes fixes for HMC V6R1.3.0 SP 0 (PTF MH01012). You must install MH01012, along with PTFs MH01082, MH01110, MH01128 and MH01127 before installing this fix. You can also reference this package by APAR MB02896.

This package includes the following fix:

HMC V6R1.3 upgrade fix

HMC V6R1.3 (MH01012) must be installed before you can install MH01127. In addition, you must install PTFs MH01082, MH01110, and MH01128 before installing MH01127. You can also reference this package by APAR MB02828.

This package includes the following fix:

Security fix for HMC V6R1.3

You must install HMC V6R1.3 (PTF MH01012) and PTF MH01082 before installing this fix. A reboot of the HMC is required after you install this fix.

This package includes the following fix:

Security fixes for HMC V6R1.3

You must install HMC V6R1.3 and PTF MH01082 before installing this fix. A reboot of the HMC is required after you install this fix.

This package includes the following fixes:

• CVE-2007-5000: Fix Cross-site scripting (XSS) vulnerability in mod_imap and mod_imagemap
• CVE-2007-6388: Fix Cross-site scripting (XSS) vulnerability in mod_status.
• CVE-2008-0005: Fix Cross-site scripting (XSS) vulnerability in mod_proxy_ftp
• CVE-2007-6203: Fix Cross-site scripting (XSS) vulnerability with HTTP request containing invalid Content-length
• saveupgdata command fails with java.lang.NoClassDefFoundError

New DST updates and Code Update fix for E302F842, B1818A0D and B1818A0E. This PTF replaces MH01064. You must install HMC V6R1.3 before installing this fix.

This package includes the following fixes:

• Fix for Code Update failures with E302F842, B1818A0D and B1818A0E
• The following Daylight Savings Time (DST) updates:
  • Argentina: Changes to DST starting on 30 Dec 2007
  • Venezuela: Changes to offset starting 09 Dec 2007.
  • Brazil: Changes to DST starting 14 Oct 2007.
  • Australia: Changes to DST starting 06 Apr 2008.
  • New Zealand: Extension of DST by 3 weeks starting 30 Sep 2007.
  • US, Canada, Bermuda, and Bahamas: Adjusting the DST start and end dates in 2007 and beyond.
  • Western Australia: Providing support for DST starting on 03 Dec 2006.
  • Brazil: Adjustments for the 2006-2007 DST period.

HMC Security: Privilege escalation by some HMC commands for HMC V6R1.3. You must install HMC Version 6 Release 1.3 before installing this fix.

This package includes the following fixes:

• Security fix CVE-2007-5135: openssl off-by-one overflow.
• Fix for security exposure in some HMC commands.

This package updates your HMC from HMC V6R1.0 or HMC V6R1.1 or HMC V6R1.2 to HMC V6R1.3. You can also reference this package by APAR MB02024. MH01012 replaces MH01018, MH01031, MH00985 and MH00946.

Platform Management

• Modified daily file cleanup routine to allow certain core files to remain on the machine.
• Corrected a problem in the Restore Critical Console data where when the restore code was executing it was actually unpacking and overlaying restore code files corrupting its own process.
• Fix for saveupgfiles/getupgfiles generating JAVA errors.
• Fix for aca.log growing when running the lssyscfg.
• Fix for command controller log file growing without bounds.
• Corrected a problem that prohibited the user from setting the HMC���s time zone to Europe.
• Added support for USB keys with pre-configured multiple partitions.
• Added support for HMC model 7310-C06.
• Corrected a problem found when doing a backup critical console data to remote machine where the process was not checking the target directory to determine if sufficient space existed to contain the data. Now the check is done and the user is notified of the potential problem.
• Corrected backup critical console data to permanently exclude restore files from the backup.
• Corrected a problem where after disabling Service Processor Failover, the user interface still indicated that the failover state was ready and allowed the user to force a Service Processor Failover. The failover did not work since the disabling had been accomplished.
• Fixed a problem with the Guided Setup Wizard. The problem: When using the wizard remotely via WebSM the process would hang if an attempt was made to add a new user.
• Corrected a synchronization problem in the Time Manager process that would not allow new trace files to be created resulting in src E3690102.
• Corrected a problem where the user could not log on the the console with a password of more than 30 characters.
• Enhanced the View Console Logs task to better handle the decompression of compressed log files that may be damaged.

Remote WebSM Client

• Fix for file permissions in the Linux WebSM remote client.
• Fix for showing correct time when viewing scheduled operations from a remote WebSM client.

System Plan

• System plan backward compatibility issue, System Plans generated from HMC's running V7 R3.1.0 code or later could cause an error when deploying on earlier HMC versions (i.e., V6). The error message of this issue would contain "Errors found parsing the internal plan file .csdml."
• Support for system plan deployment to include the expansion units of type 5787 and 5786.

Command Line

• Added man page support for install OS commands.
• Fix for lpar_netboot to support MAC address and physical location code for ping test
• Corrected an error message displayed during the activation of a partition with HCA adapter after a replacement of the adapter.
• Added support to display mount point in lsmediadev command.
• Enhanced rstupgdata to provide better check and error message.
• Fixed the chhmc command to finish the install form.
• Enhanced the lscod command to collect the CoD history log usage : lscod -m < managed system name> -t hist.
• Fixed lpar_netboot command giving erroneous usage message.
• Created a new command line task chhmcusr_and_passwd to provide authorized users the ability to change other user���s password from the command line.

Server and Partition Management

• Fixed vterm session timeout overnight
• Fixed vterm session reopen not prompting for login prompt
• Error message enhancement when activating a partition with HCA adapter after replacement of the adapter.
• Fixed running firmware commands from lpar-netboot.
• Fixed opening vterm sending extra characters to applications.
• Fixed managed system properties display problem on servers that support huge page and partitions are created with huge page and shutdown.
• Fixed delayed vterm error message for duplicate open sessions.
• Allow user to save customized 5250 console settings at the lpar level. Previously customized settings applied to all 5250 console sessions.
• Fixed changed virtual Ethernet MAC address generation.
• Fixed clearing of partition properties data after a Full System Activation with Huge Page Support.

Repair and Verify

• Corrected a problem on systems with redundant FSPs with failover enabled where an administrative failover initiated from the HMC or Repair and Verify could result in a system failure.
• Updated Repair and Verify code to interface with the new power firmware changes enhancements that allows better isolation of FSP UPIC cables failures.
• Removed the option to a 24" I/O drawer from a frame. An RPQ needs to be ordered by clients when I/O Drawers are relocated between systems. The user will receive the necessary instructions to perform remove with the RPQ.
• Corrected the procedure for the concurrent replace of an AMD (I/O Fan) on D/T5791 I/O drawer.
• Corrected the procedure for the VPD card exchange on an SF4
• Corrected an issue where servers would fail to power on after a system repair unless the FSP was reset.
• Amended the procedures to direct the user to the info center for the repair of a voltage regulator.
• Corrected procedures for the exchange of FRU���s within a CEC drawer in an ML system.
• Corrected and enhanced the concurrent replacement procedures for DASD back planes on D/T5791 I/O drawers.
• Removed the option to power on a single I/O planar in a Drawer Add function for 9118/9119 systems.

SFP

• Added new Java to address a hang in JVM.
• Fixed vague surveillance serviceable events opened.
• Fixed no ServiceEvents where the PA init hung.

Code Update

• Corrected a potential issue for systems with FSP���s and BPC���s where activation of PHYP and PFW code could be bypassed due to race conditions between the various components being updated.
• For systems with redundant FSPs code was added to detect conditions on systems with where the secondary FSP has never IPLed the machine. Prevent concurrent updating of the system when this "half-genesis" condition, because redundancy will be lost.
• Incorporated a change to provide a recovery method to concurrently activate PHYP and PFW firmware that may have been updated by not activated during a previous code update failure.
• Changed the handling certain ref codes that arise during code update to be logged as informational and not be called home. The refcodes are:
  
  E302F802, E302F803, E302F804, E302F805, E302F806, E302F808, E302F80E, E302F80F, E302F810, E302F811, E302F812, E302F813, E302F814, E302F815, E302F81D, E302F838, E302F848, E302F849, E302F84A, E302F84B, E302F851, E302F855, E302F85B, E302F87D, E302F886, E302F891, E302F897, E302F8A0, E302F8A1, E302F8F0, E302F8F1, E302F8F2,
  E302F8F4
• Corrected a problem that resulted in the execution of the command lslic -t sys -m -F --header failing with an error "HSCL8016 An unknown error occurred while trying to perform this command"
• Corrected a problem in systems with Redundant FSP configurations where the Accept or Reject of Licensed Internal Code operations would copy firmware from the incorrect flash side on the secondary FSP. New firmware would be copied over older firmware during Reject and older firmware will be copied over newer firmware during an Accept.
• Enhanced the code update process to commit firmware to the P side earlier in the update process thereby reducing the incomplete firmware updates due to potential errors or unexpected events later in the update process.
• Corrected a problem identifying and flagging fixes as deferred. Deferred fixes were considered to be those in level that the platform is IPLed at and ending with the level that is currently applied on the FSP, inclusive. However, because the deferred fixes in the current platform IPL level are already activated, they should not be included.
• Enhanced collection of user information during the code update process.

National Language Support

Fix for displaying/selecting key lock position for partition properties panel in Simplified Chinese.

Problem Analysis

• Enhanced Problem Analysis to initiate a query, when the primary HMC changes to the local console for a given managed system domain, to retrieve dumps that could have been ignored or delayed in their retrieval.
• Enhanced Problem Analysis to initiate to create an error rather than informational log when a service processor dump is initiated but cannot be retrieved by the HMC.
• Enhanced handling of connection failures created by the FSP to better correlate problems reported by FSP and HMC taking into account time differences in the logs that result from the communication delay.

Scheduled Operations

• Corrected a problem with the Dynamic Reconfiguration Scheduled Operation where it could not handle a memory parameter with a value of 0.
• Corrected Scheduled Operations to be compatible with Backup Critical Data scheduled operations created in HMC version 6.1 and prior.
• Corrected an error that prevented the Help button from working on the Scheduled Operations Detail Panel.

Service Agent

Changed Service Agent e-mail to conform to SMTP RFC and specify complete domain names in the "from" address.

Remote Support

Corrected a problem that prevented a VPN session from being established on i5/OS pass through using the Customize Inbound Connectivity task.

Networking

• Updated Network Settings to better handle subnet masks when setting up a host route in the network configuration screens.
• Changed Network Setting to allow the user to include character ���@��� in the SNMP community name.
• Enhanced Network Settings to allow the user to set media speed to fixed value (100/Full or 100/Half etc) for the 7310/7042CR4 when auto-negotiation is not working (due to adapter or switch fixes).

Concurrent code update fix for HMC V6.

Description

During concurrent code update, Hypervisor (PHYP) and Partition Firmware (PFW) updates are normally activated dynamically without the need to power the system off and back on. However, due to an error in HMC processing, PHYP and PFW updates might not be activated during concurrent code update. This problem could happen if concurrent code update was performed using HMC V6R1.0, V6R1.1 or V6R1.2 on systems that have Bulk Power Controllers (BPC). This includes i5 model 9406-595 and p5 models 9118-575, 9119-590, 9119-595. No action is needed for other system types.

There are three scenarios where the problem is corrected automatically.

• The managed system has been powered off and on after the concurrent code update.
• The most recent code update was disruptive, such as a release upgrade.
• The most recent concurrent code update was to FSPs (Flexible Service Processors) only. BPCs (Bulk Power Controllers) were not updated.
  

  This problem cannot be detected by inspecting the system. It is recommended that the recovery procedure be performed for all systems which might be affected, to ensure that the latest firmware has been properly activated. The recovery procedure can be run safely for all managed systems. It will have no effect on systems where firmware updates have already been activated.

Recovery procedure

Firmware can be activated concurrently by using this procedure.

1. First update the HMC:
   
   Install PTF MH01031 if running HMC V6R1.0, V6R1.1 or V6R1.2.

2. Next, select one of the following methods to activate firmware:

Miscellaneous fixes for HMC V6

Description

This PTF replaces PTF MH00971.

If your HMC is running Version 6 Release 1.0 or Version 6 release 1.1, then you must install MH00912 before installing MH00985. If your HMC is running Version 6 Release 1.2, then MH00912 is already installed.

The first code fix deals with a MES (Manufacturing Engineering Service) upgrade of memory or CPU. This fix must be installed prior to the MES upgrade to prevent damage to the managed system profile data. If this fix is not installed, then after the MES upgrade the managed system state may show Incomplete or Recovery. Attempts to use the Recover Partition Data task may result in error.

The second code fix deals with a backup restore problem. This intermittent problem sometimes results in the failure of the Restore of the backed-up data.

This PTF includes an enhancement to the system plan deployment function for support of future products.

Notes:

1. If you install MH00985 on top of HMC Version 6 Release 1.0, and then upgrade to either HMC Version 6 Release 1.1 or HMC Version 6 Release 1.2, you must reinstall MH00985.
2. If you install MH00985 on top of HMC Version 6 Release 1.1, and then upgrade to HMC Version 6 Release 1.2, you must reinstall MH00985.

Fix Virtual Ethernet MAC address change after upgrade

Problem description

This PTF has been replaced by PTF MH00985.

If your HMC is running Version 6 Release 1.0 or Version 6 release 1.1, then you must install MH00912 before installing MH00971. If your HMC is running Version 6 Release 1.2, then MH00912 is already installed.

This optional corrective service provides a fix for customers who experience a problem with the MAC address of the Virtual Ethernet adapter after an upgrade from HMC Version 5 to HMC Version 6. Only customers who were using virtual Ethernet before the upgrade, and then re-activate their partitions after the upgrade will experience this problem.

For this scenario, the MAC address on the virtual Ethernet adapter in HMC Version 6 will be different from the MAC address that was used in HMC Version 5. For example, if you use the MAC address of the virtual Ethernet adapter for NIM, then you will need this corrective service. This corrective service will not affect you if you do not use virtual Ethernet or do not care about the change in the MAC address of the virtual Ethernet adapter.

NOTE: If you install MH00971 on top of HMC Version 6 Release 1.0, and then upgrade to either HMC Version 6 Release 1.1 or HMC Version 6 Release 1.2, then you must reinstall MH00971. Likewise, if you install MH00971 on top of HMC Version 6 Release 1.1, and then upgrade to HMC Version 6 Release 1.2, then you must reinstall MH00971.

Fix for Cluster Ready Hardware Server issues after upgrade

Customer impact

IBM recommends that all System p cluster customers using Cluster Ready Hardware Server on their HMC apply this interim fix when upgrading from Version 5 or early Version 6 levels to Version 6.1.2. This recommendation also includes anyone with the IBM High Performance Switch Network Management component who is upgrading to HMC V6.1.2.

This fix only applies to systems at HMC V6.1.2. Do not attempt to install this interim fix package with any other HMC level.

This fix will be included in a future HMC V6.1.x corrective service update package. This package can also be referenced by APAR MB02086.

Problem description

The HMC, when running in cluster ready hardware server mode, will lose its hardware configuration after upgrading from Version 5.x or early Version 6.x to Version 6.1.2 because the IBM.HWSVRRM resource manager core dumps.

With this fix, IBM.HWSVRRM does not core dump and allows the HMC to successfully display its configuration.

Defects fixed

This HMC update contains the following fix:

MB02086: Fix Cluster Ready Hardware Server issues after upgrade

Known issues and workarounds

None

Installation notes

Follow one of the standard corrective service installation instructions for applying this fix:

• Install a specific fix from removable media
• Install a specific fix from a remote system
• Install a specific fix from the Internet

After applying the fix, use the following information to verify the success of the update.

This package updates your HMC from HMC V6 R1 to HMC V6 R1.2. You can also reference this package by APAR MB01930. All enhancements and fixes included in this package are the same as for PTF MH00919

This package represents the Recovery images that can be used to upgrade your HMC to HMC V6 R1.2 or to install a clean version of HMC V6 R1.2. You can also reference this package by APAR MB01938.

This package provides the following enhancements and fixes:

Hardware support

Support for new HMC hardware 7310-CR4

Server and Partition Management

• Fix for lpar netboot problem
• Allow dynamic partition remove operation of shared CPU till the minimum limit for non en_US local
• Fix to suppress debug messages when the mkvterm command is executed
• Fix for remote ASMI access
• Fix for displaying the hosted backing devices on a VIO server
• Fix for vslip connection requiring the HMC to be rebooted
• Corrected the display of the HMC name for the disconnect another HMC operation
• Fix for initiate system dump to handle error status indicating dump truncation
• Fix for system profile activation with many partitions
• Fix for 5250 launch problem
• Enhanced code to handle a change in Return Packet Filtering in Linux and allow VPN remote support connections between managed system partitions and power Hypervisor (PHYP).

Command line

• Fix to enable remote http access via the chhmc command
• Enhancements to "getupgfiles" to check for corrupted files and to do checksum validation
• Fix to add a Network Time Protocol server to the configuration file using IP address
• Fix for "getupgrfiles" failure if issued multiple times due to lack of space in the /hmcdump file system.
• Enhancement to allow -t udf on the mount command
• The following new commands have been added:

  monhmc
  This command provides a dynamic real-time view of HMC related subsystems and systems resources. Because this command uses underlying top, watch and df commands, you need to ensure that a pseudo-tty is allocated if you are using SSH to login to the HMC.

  chkmedia
  This command allows a user to test for media readiness on the HMC. Media devices that can be tested are DVD drive (DVD-RAM media only), internal diskette drive, USB diskette drive and USB flash memory device.

User Management

• Fix for "hmcviewer" role to be able to execute the lslic command
• Fix for remove user via the GUI

Platform Management

• Disabled default Apache page shown on HMC
• Corrected a problem where the HMC clock is wrong after the save upgrade operation.
• Fix for name resolution issue when accessing partitions.
• Fix for failure to bring up the CEC properties with out errors
• Fix for Western Australia Time zone changes.
• Fix for problems with USB memory key.
• Added ability to turn on or the Tip of the Day from the User Settings main screen.

Problem Analysis

• Corrected VPD problem on Defiant hmc that omitted HMC information from error logs.
• Corrected a problem in HMC network settings resulting from changing the managed system name and IP address where a reboot was not performed.
• Code was changed to correctly iterate through a set of sets. The typical remote machine removal procedure iterates through a set of managed systems. Rio Domain removal iterates through a set of sets. The Rio domain removal didn't correctly iterate through the set of sets to removal the actual system from it's domains.
• Changed the handling of surveillance error during code update of a single managed system handle by multiple HMC's. Instead of a Serviceable Event being opened with reference code of B3030001 and text that indicates that multiple HMC's detected a problem to multiple systems a and event with a reference code of B3030007 will be opened since a connection with only one system has been lost.
• Corrected a problem where extended error data could contain residual data from a previously reported problem.
• Changed code to disable the setting of virtualLED on partition surveillance errors
• Corrected a problem with HMC to peer HMC failures.

Scheduled Operation

Corrected a problem with Scheduled Operations to maintain scheduled times over daylight savings time changes.

Code update/System information

• Corrected a problem where incorrect code levels were displayed for I/O devices
• Corrected a problem with HMC to peer HMC failures.
• Enhanced Code Update to be able to handle processes where the location code of the power frame not available.

Service Agent

Corrected a problem where PM data is collected by Service Agent and xmitted to the SDR, the partition-name field in the XML is populated from the DNS hostname of the partition. In some customer scenarios, this data is missing or invalid, causing the partition-name to be null, making subsequent data correlation difficult or impossible.

Network settings

Enhanced Configure Network Settings to handle different media speeds.

Repair and Verify

• Added support for new managed system models 9407-515 and 9406-525.
• Corrected a problem with cold repair of redundant FSP's on ML12 and ML16 models.
• Corrected repair and verify procedures for exchange parts on Enclosure chassis (B&C).
• Corrected repair and verify procedures for HV4 VPD FRU exchange.
• Eliminated unnecessary steps for FSP cold repair on ML.
• Added steps to for FSP cold repair on ML to direct a user to flip the FSP reset switches.
• Corrected instructions for performing an FSP replacement on a 9119.
• Enhanced instructions for ML Cold Repair on ML to better instruct the user on powering down the managed system and to better indicate expected status when power on is complete.
• Enhanced the Voltage Regulator replacement procedure to handle systems with and without control panels.
• Corrected graphics for SF2 VPD card replacement procedure.
• Corrected code for handling BPD repair procedures.
• Corrected a problem with FSP cold repair on an ML when the primary FSP was located in the bottom drawer.
• Added and enhanced panels to improve the repair and verify processes for the AMD expansion unit or enclosure chassis.
• Enhanced H4 VPD card replacement procedures.
• Enhanced Squadrons H BPC replacement procedure panels.
• Corrected a problem with graphics missing from panels for the BPR to IBF cable exchange procedure.
• Enhanced System Processor RV instructions for SMP cable removal and instructions for Restore Service Processor Settings.
• Added support for U7045.SW4 Federation switches.
• Enhanced procedures for the repair of the U787F.001.DPM0N84 backplane.
• Enhanced panels for the exchange procedure for the PCI Adapter Card.
• Corrected a problem with an incorrect link to the InfoCenter for the removal and replacement of the 7311-D20 DASD.

This package is critical for customers in China that connect to the IBM service provider using a modem for remote service and support.

The phone numbers used to connect to the IBM service provider from China have changed. You must make configuration changes and install this package prior by January 31, 2007, to avoid a disruption in remote service and support from China. Refer to Enhancements and Fixes for a full description of the steps required to make this change.

You can also reference this package by APAR MB01940.

This package provides the following fix:

Update Modem Connectivity to the IBM service provider in China

As of February 1, 2007, the existing phone numbers used by the HMC to connect to the IBM service provider within China will be discontinued. Effective immediately a new, expanded set of phone numbers are available to connect to IBM in China.

Before your HMC can successfully connect to the IBM service provider using these new numbers, this package must be installed on your HMC. We recommend that you configure a minimum of two phone numbers.

The new North China phone number 16970 can be dialed from the following cities.

The new South China phone number 4006 744444 can be dialed from the following cities.

The following phone numbers will be accessible throughout China.

You may continue to connect to the IBM service provider electronically from China without installing this package if you can do either of the following:

Use an existing broadband connection to connect to IBM.��� You can configure either the Internet VPN or Internet SSL option.

OR

Configure your HMC to call home using an international call to an AT&T access point outside of China. The following numbers may be convenient alternatives. Note that international dialing codes must be added to call them from China.

1. Hong Kong

   3001-1700

2. Singapore

   6825-1800

3. Seoul

   02-3018-3073

Steps required to change phone numbers used to connect to the IBM service provider

1. Before you begin, install this package.
2. Log on the Hardware Management Console. In the navigation area,��� open Service Applications.��� Select the Remote Support Option, and then click Configure Outbound Connectivity.
3. Within the Customize Outbound Connectivity Pane, click the Local Modem Tab.��� Within the Local Modem tab, locate the Telephone Number box, and then select Add.
4. From the Add Telephone Number pane, make sure that China is selected as the Country or Region.��� Compare the phone numbers displayed to the list shown above.��� If the new numbers are displayed in the list, select a number from the list, and then click Add.���
5. If the newer numbers are not yet on your list, you may type the numbers in the Telephone Number field instead of selecting of one of the numbers displayed. Note that if you manually enter these numbers, there is an additional requirement that the System Address, as specified in the Customize Customer Information Pane, is set to China.���
6. IBM recommends that a minimum of two phone numbers be configured for each Call Home server.��� You may test your connectivity to each phone number from the Customize Outbound Connectivity Pane.
7. Once you have successfully tested your connectivity using the new numbers, you should remove the older numbers from the list of Telephone numbers.������ From the Customize Outbound Connectivity Pane, select the phone number you wish to remove, and then click Remove.
8. After you have completed all your changes, click OK from the Customize Outbound Connectivity Pane.

This package provides remote support and cold repair fixes for HMC V6. You can also reference this package by APAR MB01907.

This package provides the following fixes:

• Fix for VPN remote support connections to managed system partitions
• Fix for 5250 launch problem
• Fix for FSP repair going to Recovery when HMC is rebooted
• Fix for lpar netboot problem

Description

HMC Maintenance Package Version 6, Release 1.1. This package includes fixes previously provided in MH00833 and MH00856. If you are upgrading to HMC Version 6 Release 1.0, you must install this maintenance package.

You can also reference this package by APAR MB01797.

This package provides the following enhancements and fixes:

Server and partition management

• Fix for lshwres command to list the physical location of the backing devices mapped to a disk on a VIO server.
• Fix for RMC/IBM.lparCmdRM initialization that may cause the concurrent firmware update process on AIX and Linux partitions to partially succeed.
• Fix for saving the opticonnect information via the save profile function of a running partition.
• Corrected incorrect display of Barrier Synchronization Register.
• Update recovery state warning message to include concurrent merge profile option.
• Fix for chusrtca command not working when the -o e option is specified.
• Fix for 5250 launch problem after the installation of the PTF MH00688.
• Alleviates a problem for AIX and Linux partition users who force close vterm sessions with the "Close Terminal Window" option in the GUI or rmvterm command.
• Fix for multicast MAC address error during activation of a partition with virtual Ethernet adapters.
• Fix for redundant FSP function showing incorrect readiness status message.
• Fix for activation error when duplicate trunk adapter is set to desired.

Command line

• An option has been added to the chsacfg command to allow SNMP trap notification configuration.
• Options have been added to the lssacfg command to display SNMP trap notification settings and to display all defined SNMP traps
• A new command, rmprofdata, has been added to remove a profile data backup file from the HMC.

Platform management

• Added support for formatting a high-speed memory key.
• Corrected a problem where the editing of the clock format causes the HMC to be unusable.
• Corrected a problem where the RIO topology menu shows wrong model for 9119 59X.
• Enhanced Administrative SFP Failover by providing an additional check for HMC connection. Also added a warning message recommending verification of the secondary FSP connection before initiating a failover.
• Enhanced directory handling by the Backup and Restore task, after an upgrade and during remote restores.
• Corrected a problem during remote backup of critical console when users were viewing details of remote console data.

System plan

Enhanced mksysplan and cpsysplan to check and enforce System Plan file naming conventions.

Repair and Verify

• Fix for GUID change after the replacement of a HCA Adapter.
• Enhanced graphics for better FRU location reporting and repair guidance 7311-D10, 7311-D11, and 5790 expansion units.
• Enhanced panels for BPR exchange parts to show pictures of the IBF CBs and connections.
• Enhanced System Processor RV instructions for SMP cable removal and instructions for Restore Service Processor settings.
• Enhanced RV panels to better enable the user find the RIO/HSL cards and connectors.
• Enhanced RV for the CSM environment to handle the case where the HMC is not the DHCP server and HMC to FSP communication is determined by the CSM server. The GUI has been enhanced to instruct the user to assign replaced FSP or BPC units to the HMC via the CSM server.
• Enhanced RV code to check for unique enclosure and model numbers instead of unique enclosure types, only when checking for duplicate enclosures during add/replace/repair.
• Added code for better detection of ACDL failures during DCA repairs.
• Provided fix for errors detected during exchange of an Optical RIO GX bus adapter.
• Corrected the cold repair of systems equipped with 3 or 4 CEC drawers to instruct the user to remove power from all enclosures not only enclosures containing FSP cards. (Model 570)
• Enhanced RV panels to provide complete instructions for powering on A system and to indicate correct expected status when power on is complete.

Access control

Fix for access control that causes the authentication token to be removed during logout.

Code update

• Enhanced code update code by adding retry logic around calls to GET_ACDL_STATUS to handle cases where GET_ACDL_STATUS times out. There can be up to ten retries, with a 60 second wait between them.
• Corrected a problem where incorrect messages were being displayed on progress panels when multiple CECs in the same frame were selected for update. (Model 575)
• Corrected a problem during a multi-system LIC accept on systems that have a single power frame associated with multiple CECs. (Model 575)
• Corrected an issue where a frame lock was obtained but not released during a "remove and activate".
• Enhanced the message pertaining to deferred fixes to clarify when a user must IPL the system immediately after an update, or if an IPL can be postponed.
• Corrected and improved Code Update for handling of I/O microcode updates.
• Enhanced code update so that errors that are detected during informational operations (that is, firmware is not being updated) will not create Serviceable Events.
• Corrected a problem found during concurrent updates where deferred fixes were identified to be in the fixpack but no deferred fixes were displayed in the details.

Problem Analysis

• Enhanced the Manage Serviceable Events user interface to better display text from operating system messages.
• Corrected a problem where after a full HMC reinstall, Service Focal Point was reporting operating system errors that had been previously reported, commented and closed via SFP Manage Serviceable Events.
• Corrected an issue where frame power errors were not being reported because of the inability to properly associate a failing BPA with a CEC.
• Problem Analysis logic was amended to handle cases when two different reference codes are reported to ASM simultaneously. Logic was changed to report the error with the older time stamp to better isolate the problem and identify proper FRUs rather than the reference code with the higher priority.
• Corrected a problem where SRC B3030001 was being used instead of refcode B3030007 to indicate an HMC has lost connectivity to more than one managed system.
• Corrected a problem where the sequence of serviceable events gets out of sync on peer HMCs which could result in the inability to handle or close serviceable events through the Manage Serviceable Events task.
• Enhanced Problem Analysis to handle multiple surveillance errors during a code update to a managed system. Only one will be reported with the with the others being marked as duplicates.
• Enhanced Problem Analysis to better handle FSP failovers.
• Enhanced Problem Analysis to better detect the primary HMC in a dual HMC environment.
• Enhanced performance of extended error data collection.

Service Agent

Corrected a problem with user interface to display error messages.

Networking

Corrected a problem where the Customize Outbound Connectivity and Customize Inbound Connectivity panels were always disabled when accessed remotely.

Miscellaneous

• Fix for IBM Director not displaying the HMCs.
• Fix for pedbg to collect the correct DHCP config file.

Description

This package provides OpenSSL and Apache2 security fixes for HMC V6R1.0. You can also reference this package by APAR MB01869.

This package provides the following security fixes:

1. CVE-2005-2700, CVE-2005-3357, CVE-2006-3918, CVE-2006-3747: Apache2 security fixes
2. CVE-2006-4339: OpenSSL RSA signature evasion

Description

The PTF MH00816 corrective service package contains updates to the InfoCenter that pertain to HMC Version 6 Release 1.0. This package is also referenced by APAR MB01647.

This package contains InfoCenter updates for HMC 6.1.

Description

This PTF provides a command monkacpid, that allows processor usage of the kacpid kernel daemon. This PTF can be referenced by APAR MB01845.

This PTF can only be applied on HMC V6R1.0. It must be applied before applying the required maintenance package.

A number of internal modems, when installed in HMC Machine Type 7310-C04, are affecting the performance of the HMC. If you have such an HMC configuration, install this PTF, and then run the monkacpid command from either

• a restricted shell terminal at the HMC, or
• a remote client over SSH.

This command produces output similar to the following. The output is refreshed every 4 seconds.

 Every 4.0s: /usr/bin/top -s -b -n 1 -p 6                Thu Sep 14 20:29:30 2006

  top - 20:29:31 up 1 day,  5:14,  5 users,  load average: 0.04, 0.01, 0.00
  Tasks:   1 total,   0 running,   1 sleeping,   0 stopped,   0 zombie
  Cpu(s):  1.6% us,  0.3% sy,  0.0% ni, 97.2% id,  0.9% wa,  0.0% hi,  0.0% si
  Mem:   1032992k total,  1005644k used,    27348k free,    36820k buffers
  Swap:  2040244k total,        0k used,  2040244k free,   192952k cached

 PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
   6 root      15 -10     0    0    0 S  0.0  0.0   0:00.00 kacpid

If the number in the %CPU column is greater than 10.0, please contact IBM Hardware support to order a replacement modem. Please reference Retain Tip H187630.

Description

The PTF MH00833 corrective service package is a required Maintenance fix that must be installed on HMC Version 6 Release 1.0. It replaces the previously released PTF MH00781.

This package addresses the following issues:

• Fix problem activating partition that contains shared Ethernet adapter
• Fix problem with missing cron entries after an upgrade, resulting in full /var filesystem
• Fix to recover redundancy between primary and secondary FSPs
• Fix for incorrect display of ports on HMC Network configuration menu, Secure Web Access
• Fix to re-enable Service Focal Point problem analysis logging after HMC reboot
• Fix for a problem that causes code update to terminate when multiple managed systems are targeted, and one or more of the systems is currently at the level of code being installed
• Fix for a problem that caused system with internal modems to fail as follows:
  • when test of a phone number is performed from the Customize Outbound Connectivity panel. The test will hang and the user may see the message "Test beginning. Waiting for connection availability. Request has been added to the queue." in the Test Telephone Number screen.
  • when a call home request is made from Problem Analysis. The request will not complete and other call home requests will build up in the Waiting Request queue. This was evidenced by entries in the Active Request queue on the Manage Remote Support Facility Requests panel.
  • when the command "ls -alt /dev/modem" is issued and the response is " No such file or directory"

Description

The PTF MH00781 corrective service package has been replaced by PTF MH00833, which is now the required Maintenance fix that must be installed on HMC Version 6 Release 1.0

This package addresses the following issues:

• Fix for incorrect display of ports on HMC Network configuration menu, Secure Web Access
• Fix to re-enable Service Focal Point problem analysis logging after HMC reboot
• Fix for a problem that causes code update to terminate when multiple managed systems are targeted, and one or more of the systems is currently at the level of code being installed
• Fix for a problem that caused system with internal modems to fail as follows:
  • when test of a phone number is performed from the Customize Outbound Connectivity panel. The test will hang and the user may see the message "Test beginning. Waiting for connection availability. Request has been added to the queue." in the Test Telephone Number screen.
  • when a call home request is made from Problem Analysis. The request will not complete and other call home requests will build up in the Waiting Request queue. This was evidenced by entries in the Active Request queue on the Manage Remote Support Facility Requests panel.
  • when the command "ls -alt /dev/modem" is issued and the response is " No such file or directory"

HMC 6.1 Recovery Media images.

Description

The HMC 6.1 Recovery Media images can be used to upgrade your HMC to 6.1, or to perform a clean installation of HMC 6.1.

Updating to HMC V6 R1.0 requires an Upgrade by means of Recovery Media or the Network. There are no updates to be used with the Install Corrective Service task. An upgrade from HMC Version 4 to HMC Version 6 can only be done by first saving the upgrade data to DVD-RAM media, re-installing the HMC with HMC V6 R1.0 without selecting the Upgrade option, and then using the new rstupgdata command to copy upgrade data onto the HMC. The data copied will be automatically restored on the next reboot.

HMC 6.1.0 enhancements for the Hardware Management Console include:

Server and Partition Management

• Allows dynamic partition remove operation of the CPU to continue even if the CPU resources are out of sync.
• Provides enhanced locking to help dual HMC lock contention issues.
• Provides fix for Lpar creation profile summary showing incorrect virtual IO adapter numbers.
• Provides fix for lshwinfo command to show single records of environmental information for a managed system in a frame.
• Returns appropriate error message when chsyscfg is used to change the name of a profile to a name that already exists.
• Returns appropriate error message if duplicate VLAN ID is detected during activation of a partition.

System plans

• Support for deploying plans created by the System Planning Tool.
• Support for deployment of system plan as created by the mksysplan command.
• Support for creating a system plan from the WebSM interface.
• Fix for mksysplan to allow it to produce a system plan from a managed system that has a space in its name

Platform management

• Enhanced the clean up script to delete heapdumps that fill up the / filesystem.
• Enhanced the lshmcusr command to show the attribute pwage (password age)
• Changed Hot plug of USB Floppy drive to require floppy media in the drive to detect the device.
• Added new Daylight Savings Time support

HMC Command Line

• The following commands have been added for for HMC upgrade purposes

  saveupgdata
  Allows Upgrade Data to be saved to disk, or to disk and DVD-RAM.

  getupgfiles
  Allows upgrade files consisting of HMC code to be copied from a server onto the HMC disk partition. This command should ONLY be used prior to performing an upgrade to a version higher than HMC version 5. This command obtains all the necessary files required to perform an upgrade, and then puts them in a special disk partition on the HMC. These files are currently published on the HMC support web site and also reside on the DVD. This command was introduced in HMC V5R2.1 to allow remote upgrade of the HMC to a higher version.

  rstupgdata
  Allows previously saved upgrade data to be restored from DVD-RAM following a new installation on the HMC. Upgrade data must have been previously copied onto DVD-RAM by running the Save Upgrade Data task or saveupgdata command, prior to performing an installation. You must reboot the HMC after running the rstupgdata command to allow Upgrade Data to be restored automatically.

• An option has been added to the chhmc command:

  chhmc -c altdiskboot -s enable --mode upgrade

  These new options enable the alternate disk boot/upgrade option. Running the chhmc command with these options, after running the getupgfiles command, prepares the HMC to use the new code copied by the getupgfiles command on the next boot.

• An option has been added to the bkconsdata command to allow a remote directory to be specified. Also, an option has been added to remove the backup task lock to allow the backup task to be re-executed immediately.
• An option has been added to the bkprofdata command to allow an existing backup file to be overwritten.
• To fix a security exposure, the bkprofdata command has been changed to allow the profile data backup file to be written only to the following locations:
  • Removable media
  • Your home directory
  • The /tmp directory
  • The default profile data backup directory

  In addition, for future compatibility the operation of the bkprofdata command has been changed to accept only the file name to write the backup file to the default profile data backup directory. The command no longer accepts the directory name for the default profile data backup directory. That is, if you have been using the following command:

  bkprofdata -m <ManagedSystem> -f /var/hsc/profiles/<SerialNumber>/backup010106

  you must now use the following:

  bkprofdata -m <ManagedSystem> -f backup010106

• The lssysconn command has been enhanced to display the physical location code for the service processor.

Licensed Internal Code (LIC) update

• Enhanced Licensed Internal Code update performance and usability by adding:
  • Performance restructure
  • Improved Progress Panel information
  • Customer-controlled Power Interlock feature
• Enhanced LIC update so that specified code repositories are used. The process previously used the hard drive if any files were found there, regardless of user selection.
• Added support for firmware update from a DVD-RAM
• Added support for "View I/O Levels" for targeted I/O devices.
• Enhanced messages for successful code updates with deferred fixes present in the fixpack.
• Added support to the HMC to gather and display the levels of code for both FSPs (Models 570, 590 and 595) and both BPCs (Models 590 and 595).
• Added support to Code Update to use a local disk as a repository for firmware upgrades (new GA levels)
• Added support to the GUI to warn that a firmware down level (down-grade) will be disruptive. A message was also added to inform the user that if the destination level of firmware level matches the level on the P-side, the same function can be performed concurrently via the remove and activate function.

Scheduled Operations

• Added support to Scheduled Operations to perform back up of critical console data to a remote server.
• Added the option to Install and Activate latest Concurrent LIC level as a Scheduled Operation.
• Corrected a problem where a Scheduled Operation was attempting to perform an unsupported operation during its error handling routine. A repeating scheduled operation that encountered this error would not run again until the next HMC reboot.

Repair and verify

• Enhanced the fault isolation priority algorithm for MUX card analysis.
• Corrected a problem where a service lock was not freed after a failed FSP repair.
• Enhanced GUI content and panel ordering usability.
• Corrected a problem that caused power errors to be logged after replacing a power supply.
• Corrected the Repair and Verify GUI to include instructions for setting the Enclosure Serial Number when performing an I/O backplane replacement in an P575 system.
• Corrected a problem where the control panel was not selectable from the list of removable FRU types (Model 570).
• For InfiniBand HCA replacement procedures, information was added to the RV panels to direct the user to InfiniBand HCA replacement procedures documented in the "Clustering systems using InfiniBand Hardware" manual.
• Added support to Repair and Verify for the repair of FSP devices on model 570.

Save Upgrade Data/Backup

• Added a feature to allow customers to backup HMC critical console data to specify a remote directory for disaster recovery. This feature also allows you to maintain multiple versions.
• Enhanced Restore from FTP site to account for differences in operating systems.
• Corrected a problem during the restore of Critical Console data saved before HMC V5R2.0

Network connectivity

• Added code to limit the range of IP addresses that may be assigned in the 192.168.x.x class B DHCP Server Range IP range.
• Added Internet SSL connectivity options to the Guided Setup Wizard.

Problem analysis/Call Home

• Enhanced Serviceable Events to include the partition id for Operating Systems failures.
• Enhanced handling for Surveillance errors and problem closures.
• Corrected a problem in a dual HMC System Configuration where errors were being called home twice.
• Enhanced dump handling to allow dump data to be retrieved from systems that are no longer managed by the HMC.

RMC

• Enhanced to prevent HMC interactive performance degradation after Linux or AIX images are cloned.
• Corrected an issue where a Linux or AIX partition should not be shut down using the OS option after an LPAR had its hostname and node id changed.

RIO Topology

Enhanced the View RIO Topology tool to include the Leading and Trailing Port status.

Service Focal Point (SFP)

• Enhanced the panel for 'Administrator Service Processor Failover' to eliminate the chance that a user may inadvertently initiate FSP Failover to occur.
• Changed code to display the time in the HMC's time zone when using the remote WebSM client.

The following table lists the languages and locales supported in HMC 6.1.

• Help Search is not supported. Help Find is supported.
• Printing is not supported.
• The numerical keypad on some keyboards doesn't work. Use the normal numerical keys instead.
• When your POWER5 system is shipped to you it is in a non-HMC managed mode by default. Once you attach and set up an HMC to manage your Power5 system, the default non-HMC managed mode exits. The procedure to return to the default non-HMC managed mode is documented on the InfoCenter web site under Resetting the server to a non-partitioned configuration

• To launch the browser, right click on the desktop to display the menu. Select "Net", and then "Browser".
• To start a restricted shell terminal, right click on the desktop to bring up the menu. Select Terminals", and then "rshterm".
• Currently, HMC only stores four (4) service processor dumps and four (4) platform system dumps per managed system.
• Changes of a partition profile do not apply immediately to the partition. The changes on the partition only take effect when activation is done through the profile.
• To view keylock positions and SRC values when performing Operator Panel Service Functions through Service Focal Point, use the Server and Partition plugin.
• To view system event logs, you must login as a user with the hmcpe role. From Service Applications, select Service Focal Point -> Service Utilities -> Actions -> View Problem Logs.
• HMC Version 4 Release 2 or higher is required to manage pSeries Power5 servers.
• HMC has enabled UTF-8 locale support. This additional support is currently only available through the command line. To enable the new locales, execute the following command: chhmc -c <locale> -s modify -l <locale name>
• HMC has enabled additional keyboard support. There are multiple pages for the keyboard configuration. During the keyboard configuration, enter 98 to go the next page and 97 to go to the previous page.
• If the HMC is used in a Cluster 1600 environment, see the following PDF-formatted document for scaling limitations: IBM eServer 1600 Facts and Features
• Refer to the following page for the recommended HMC and Server code levels for the currently supported Power5 releases. Power5 Code Matrix: https://esupport.ibm.com/customercare//sas/f/power5cm/home.html

The following customer courses will be available from Resource Link for the Product Announce on May 4, 2004.

• How to Use the eServer i5 and p5 Hardware Management Console
  This course explains how to install and configure the Hardware Management Console (HMC) for the Model 520/570. It also covers the HMC's basic operations by exploring the general user interface. Partitioning is discussed in detail, with explanations showing both command line and GUI support.
• eServer i5 and p5 - Physical Planning for Installation
  This course discusses considerations for the physical planning for installation of eServer Models 520 and 570. It then provides the user with links to obtaining the step-by-step installation procedures.
• Resource Link Highlights
  This course provides an overall look at the major Resource Link (RL) areas and how to use them. These areas are Planning, Education, Library, Forums, Fixes, Problem Solving, Services, Customer Initiated Upgrade, and Tools. It also provides step-by-step procedures in using the Subscription function.
• Performing Licensed Internal Code Maintenance
  This course discusses about maintaining the software that enables hardware such as the service processor on your eServer POWER5 system.

Access to these courses requires an IBM Registration ID and Resource Link Access.

To obtain an IBM Registration ID

1. Go to http://www.ibm.com/servers/resourcelink
2. Select "Register" under New users
3. On the My IBM Registration, fill in an e-mail address for IBM ID and password, and the Security question and answer and the Country of residence and then click Continue and fill in the rest of the User information and click Submit.

Resource Link Access for New Users

1. Go to http://www.ibm.com/servers/resourcelink
2. Select "Sign in" enter the IBM ID and password you used to register above
3. Select "Customer"
4. Click "Submit" Once submitted it will take an hour before the access takes effect

To view a course

1. Sign in to Resource Link
2. Select Education in the navigation bar on the left
3. From the Education page, select "eServer i5 and p5 courses"
4. Select "How to Use the eServer i5 and p5 Hardware Management Console"
5. Select "Performing Licensed Internal Code Maintenance"
6. Double-click the link to open the course.

Notes:

1. The courses are browser based. For optimal viewing, we recommend Microsoft Internet Explorer 6.0 or higher with your display set at 1024 x 768. To set your display, go to My Computer -> Display -> Settings
2. Course pages can be book marked for easy retrieval.
3. Course simulations open in a separate window. Use the X in the upper right corner of the window to close the simulation.
4. If there are Quick Time videos in the course, you need to have the Quick Time viewer installed. A link to the free viewer is on each Resource Link course page and on the related page within each course.

Documentation, installation guides, cumulative PTF history, best practices, and related support information:

HMC documentation

• HMC Notices
• POWER code matrix
• FLRT: Fix Level Recommendation Tool