For use with Version 8 Release 8.4.0

Date: 26 September 2017
(C) Copyright International Business Machines Corp., 2017 All rights reserved.

Contents

The information in this Readme contains fix list and other package information about the Hardware Management Console.

• Description
• Package Information
• Known Issues
  
• General fixes
• Security fixes
• Installation

This package represents a service pack image that can be used to update your HMC from HMC V8 R8.4.0 to HMC V8 R8.4.0 Service Pack 3. You can also reference this package by PTF MH01652 and APAR MB04036. This image can be installed on top of HMC Version 8 Release 8.4.0 Recovery installation PTF MH01559 with or without additional PTFs installed.

Note: Service packs are cumulative and as such will include all the fixes for the PTFs released up to and including the last service pack(s) for this HMC version.  Please read the individual Readme files for each PTF to see the list of fixes.

• Fixed a problem where the HMC Management > Manage Certificate Revocation List task was not displayed on the classic GUI for custom HMC users.
  
• Fixed a problem that caused a DLPAR memory operation to fail with the error "HSCL03F9 Not enough memory resources to meet the allocation setting."  This problem only occurs if memory is being added to a partition and the new total amount of memory assigned to the partition exceeds the amount of memory currently available on the managed system.  In addition, this problem only occurs when the DLPAR operation is performed using the enhanced GUI, the REST API, or the chhwres command with the -o s option.
  
• Fixed a problem that caused the following error to be displayed when running a repair and verify procedure against a frame component: "Could not register as primary owner of server <MTMS>"
  
• Fixed a problem that caused the chhmcldap command to sometimes falsely report success (a return code of 0 and no error message) when setting an attribute when the actual result was that the command did not set the specified attribute. This problem only occurs when the specified attribute is missing from the LDAP configuration file on the HMC.
  
• Fixed an issue with pedbg log collection that caused ctsnap_out data to be missing and also could cause SRC E212E136 to be called home due to /tmp filling up with ctsnap data.
  
• Changed the lshmcldap command to show an empty string for the LDAP search scope attribute instead of "one" when it is not configured on the HMC.  When the scope attribute is not configured on the HMC, it could default to either "one" or "sub", which may lead to unexpected behaviour.
  
• Fixed a problem preventing server dumps from being offloaded to the HMC when the server is in the Incomplete state and resource utilization data collection is enabled for the server on the HMC.
  
• Fixed a rare error in the remove user operation that caused SRC E3550926 to be called home.
  
• Fixed a problem that sometimes causes the deploy of a system plan created using the System Planning Tool for some server models, including 8286-41A and 8286-42A, to fail with error "One of the I/O slots specified using the io_slots attribute has a missing or invalid DRC index.  The value specified was "null".  With this fix, the deploy will no longer fail since any I/O slots with a missing DRC index will be skipped.  Therefore, after the system plan has been deployed, ensure that your partitions have been configured with all of their I/O slots.
  
•  Improved the warning message displayed during a system template deploy operation when VIOS installation fails.
  
• Fixed a problem that caused the root filesystem to reach the 80% utilization threshold and SRC E212E136 to be called home as a result.  This problem only occurs when the "lshmc -h" command is run repeatedly.
• Fixed a problem that caused the removal of a virtual I/O adapter from a Virtual I/O Server (VIOS) to fail with the message "HSCLAB10 There is not a virtual NIC adapter in slot <slot number>." This problem only occurs when the VIOS is hosting virtual NICs and when the virtual I/O adapter being removed is in slot 128 or higher.
  
• Prevented the generation and call home of SRC E3551231 every time a CR5 or CR6 model HMC is restarted.
  
• Fixed a problem causing error 403 and error message "The User does not have the authority to access the object LogicalPartition" to be displayed when attempting to display the partitions of any server on the Enhanced+ GUI.  This error occurs when there is a powered off server connected to the HMC and that server has never been connected to any HMC while it is in Standby or Operating state.  To work around this problem, power on the server or remove it from the HMC.  With this fix, the partitions on a powered off server that has never been connected to an HMC while it is in Standby or Operating state will not be displayed, but the partitions on all other servers will be displayed.
  
• Changed the server firmware activation process to report an error when the activation of platform firmware of a partition fails due to an RMC error. Prior to this fix, the server firmware activation process appeared to finish successfully.
  
• Fixed an issue where installing HMC updates using the classic GUI appears to hang without completion. This happens only when http is updated and the classic GUI connection gets disconnected during installation.
  
• Fixed a problem that always caused the mkauthkeys command to silently fail when the -u option was specified. This issue prevents users from adding SSH keys for other users but does not prevent a user from adding an SSH key for the user they are currently logged in as.
  
• Fixed an issue causing the generation and call home of SRC E3321007 after updating the HMC. This issue also can cause data collection for other call home events to fail, preventing the events from being called home or from sending the necessary data.
  
• Fixed an issue that can cause lock contention on HMCs managing partitions that have simplified remote restart (SRR) enabled and that have the Allow Migration with Inactive Source Storage VIOS server option enabled. Symptoms include errors "HSCL025A Service processor lock failed" attempting other tasks such as DLPAR and "HSCLA9D7 The configuration data which is used during remote restart of the partition could not be updated."
  
• Fixed an issue that caused the updhmc –t nfs command to fail intermittently with the error "An error was detected while mounting the remote server. Verify the parameters have been entered correctly and try the operation again."
  
• Fixed a problem where using the chsacfg command to add an email address for event notification failed with the error "An invalid parameter value was entered.  The parameter -a is empty or not valid.  Please check your entry and retry the command.".  This problem only occurs when the domain name in the email address starts with a number.
  
• Fixed an issue that caused the chhmc command to enable or disable jumbo frames to always fail.
  
• Corrected the calculation of server level processor utilization shown on the enhanced GUI and returned by the REST API when Performance and Capacity Monitoring data collection is enabled.
  
• Fixed a problem on virtual HMCs that caused all Service Management > Transmit Service Information > Send Problem Reports tasks to fail without an error but did cause SRC E3D45069 to be generated.
  
• Fixed a Performance and Capacity Monitoring (PCM) data collection issue where after changing the user-defined name of a managed system PCM stops collecting VIOS data for network and storage traffic and displays "VIOS data not available".  This problem only occurs if PCM is disabled then re-enabled after the managed system's user-defined name is changed.
  
• Fixed an issue that caused repair and verify tasks and Licensed Internal Code update tasks to always fail for a powered off server which had NovaLink set as the PowerVM management master when the server was powered off.  Repair and verify tasks failed with the error "Could not register as primary owner of server."  Licensed Internal Code update tasks were not available on the GUI, and when run from the command line failed with the errors HSCE2124 and "HSCF0261E Could not start the update because this management console is not the master console.  Check to see if there is another management console program attached to the target server <server>."
  
• Fixed an issue that caused a chperfmon command failed message to be logged repeatedly in /var/log/messages for a server that was no longer managed by the HMC.  This problem only occurs if Performance and Capacity Monitoring (PCM) was enabled on the HMC for the server prior to that server being removed from the HMC.
  
• Fixed an issue that caused the generation and call home of SRC E212E134 to report that the HMC /var file system size exceeded the threshold limit of 80%.  This problem can only occur if the HMC is configured to transmit performance management information to a service provider and the HMC is managing a large number of partitions.
  
• Fixed a problem where old Elastic CoD information was being called home, resulting in out of compliance emails being sent out from the CoD project office.
  
• Fixed a rare issue where HMC performance degrades over time until the command server and/or web servers hang requiring an HMC reboot to resolve. Logs will show a large number of partition surveillance events and blocked threads and may also include out of memory errors. This issue can occur when network connectivity issues repeatedly occur between the HMC and large numbers of partitions over a short period of time.
  
• Fixed an issue where email notifications for HMC reported service events were not sent when the failing machine type and model differs from the underlying model type of the POWER server (for example the 5146-GL6 Elastic Storage Server systems).
  
• Fixed a problem that caused the following error to be displayed when trying to open the Virtual Storage Management GUI window:
       "There was a problem running the VIOS command. HSCL2970 The IOServer Command has failed because of the following reason:
        hdisk1,[VG NAME],???????,???????,???????,[PVID]
       Volume group must be activated; use activatevg command."
  This error occurs when a volume group on the Virtual I/O Server being queried is varied off.
• Fixed a problem that caused the WWPN and WWNN information for Fibre Channel adapters to be missing from system plans created on the HMC.  This problem only occurs for quad port Fibre Channel adapters.
  
• Prevent the generation and call home of SRC E3550925.  This SRC is generated when creating a Kerberos user and no remote user ID is specified or the remote user ID specified is not valid.
  
• Fixed a rare problem that caused a server to return to the Save Area Version Mismatch state immediately after the partition data for the server was successfully recovered.
  
• Prevent the call home of SRC E3690102.  This SRC is generated during the installation of an HMC service pack or iFix.
  
• Fixed a problem where an attempt to deploy a system plan created with the IBM system planning tool on a model MHE/MME fails with an error reported against two of the three system units: "System plan system unit 9119_MME-1-0, type MME not valid as planned. Validation errors occurred when compared to every MME type system unit found on the managed system. system unit 9119_MME-1-0 planned hardware in this system unit not found on the managed system in any unit of type MME."
  
  

Security fixes

• Disabled TLS 1.0 for HMC port 12443 in legacy security mode.
  
• Removed support for all ciphers that use a Diffie-Hellman modulus of 1024 bits or less from HMC ports 9920, 9960 and 12443.  This change was made to address the following vulnerability: SSL/TLS Diffie-Hellman Modulus <= 1024 Bits (Logjam).
  
• Fixed a security issue with HMC GUI help panels. 

Special Install Instructions

a. The enhanced GUI is not supported for install.  Installing this PTF using the Enhanced+ interface may hang.

b.  Classic GUI install from USB requires Service Pack 0 fix level MH01588 or later.   To install from USB at earlier levels use CLI updhmc command.  See the updhmc "man" page (e.g. man updhmc) for further information and  examples on using the command.

To use the classic GUI install:
1.    Log in again selecting the Log In option of "Classic".
2.    If already logged in to the HMC using Enhanced GUI, log off the HMC.
3.    Install using the normal installation instructions.

Installation instructions for HMC Version 8 upgrades and corrective service can be found at these locations:

Upgrading or restoring HMC Version 8

Installation methods for HMC Version 8 fixes

Instructions and images for upgrading via a remote network install can be found here: