March 2018 Newsletter

Greetings iTech Fan,

The March newsletter is a little late this month not because we have been watching the NCAA Basketball Finals, but because we are averaging about 10 upgrades a week. The entire technical team has been going at a whirlwind pace, and it just seems that everyone wants to have their OS upgraded from IBM i 7.1 before the last day of regular support which ends on April 30, 2018. Now, that’s a good thing to upgrade from 7.1, the bad thing was waiting until the last minute. If you need help with your upgrade, turn to the business partner that everyone else is turning to for upgrades, www.itechsol.com

In February IBM introduced the POWER9 family of scale-out servers, which are targeted towards most of the IBM i community. A very exciting announcement on hardware that should support the IBM i community for the next 3 or 4 years. This new lineup of POWER9 servers have just started to ship, if you would like more information please contact our sales team. Our first article examines some of the differences between POWER8 and POWER9.

The PTFs for Spectre and Meltdown have been released for IBM i, and all POWER7, POWER7+, and POWER8 FSPs. There has actually been a few sets of these PTFs, so make sure you have the latest. We have been putting these on customer’s machines, and its time to get them loaded onto yours. In general, we have noticed a 5% performance impact. Of course, your mileage may vary. Check out the last article on PTFs to see what you need, or if it’s easier just give us a call and let us do them for you.

What happened at Under Armour®? The athletic wear company announced last Thursday that an unauthorized party gained access to the data tied to its fitness app, MyFitnessPal. The breach, which took place late February 2018, affected approximately 150 million user accounts.

The related information that was accessed includes:

  • Usernames
  • Email addresses
  • Hashed passwords

The company is urging MyFitnessPal app users to update passwords immediately. What would happen if this was your IBM i? Are you protected or somewhat protected or worse unsure? You are responsible for your environment, wouldn’t you like the peace of mind knowing that you have safeguards in place? Contact Phil and let him show you how we can protect your IBM i.

We are continuing to grow and expand our technical team, and welcomed both Mike who joined us in January, and Amy who joined us this week. They are welcome additions to our staff, with their diverse skills and knowledge. We are growing to meet the needs of our clients, and we see that trend continuing going forward. Thanks for giving us the opportunity, we strive to exceed your expectations.

Don’t forget, in April we have the Northeast User Group Conference in Framingham, MA, April 23 through the 25th. Over 80 sessions to improve, grow, and enhance your IBM i skills. It’s a great price, at a very convenient location. If you haven’t yet registered, please visit their website www.neugc.org. There will be many iTech Solutions folks at the conference, presenting, exhibiting, and attending. Its a great chance to stop by and have a chat with us.

This newsletter has 6 articles. The first article is by Charlie on the difference between POWER8 and POWER9. The second article by Chris is on after upgrading from IBM i 7.1 to 7.3, users had problems connecting to file shares. The third article by Laurie detailing what you get with PowerVM Enterprise Edition, which comes at no charge with POWER9. The fourth article is one of my favorite commands, Change IPL Attributes. The fifth article lists some of the upcoming events in which iTech Solutions will be participating. The last article is for your reference with updated PTF information. Please note that for all 7.1 customers that are on the Quarterly or Semi-annual iTech Solutions PTF maintenance plan, we will be installing the latest PTFs as you are most likely now on Technology Refresh 11. For the 7.2 customers, we will be installing 7.2 Technology Refresh 8, and 7.3 will be Technology Refresh 4.

newsletter_image1.jpg

POWER8 or POWER9 for IBM i

On February 13, 2018 IBM announced the scale-out models of the POWER9 servers.  Scale-out servers are the smaller servers, with 1 or 2 sockets and as many as 24 cores, that most IBM i customers use.  The larger Enterprise class POWER9 servers are expected to be announced later in 2018.  The new servers are a replacement for POWER8 servers that have already been in the market for over 3 years.  I expect both POWER8 and POWER9 servers to be offered by IBM for at least a year.  If you need a new server to run your IBM i applications, while both POWER8 and POWER9 are available, which one should you choose?

For this article we are focusing on the POWER9 1-socket model (S914) and the 2-socket model (S924), although there are other POWER9 models available that can run IBM i.  These are the models that most IBM i customers use. The picture below is a 2-socket S924.

S924 no cover.jpg

Other than the label, from the outside POWER9 servers look exactly like POWER8 servers with just one notable exception, the lack of an internal DVD drive. It appears that IBM is moving away from software distribution on DVD’s. Instead the preferred method is electronic download to a PC and then copy to a USB Flash Drive. IBM Power Systems servers have had USB ports for quite some time, but IBM i had limited use of them. USB IPL support was added to IBM i with V7.1 TR6 to perform a D-Model IPL on POWER7 and above machines. You can download IBM i images from the Entitled Systems Support (ESS) website, copy with a program from the website to a USB flash drive, do a D-mode IPL on a managed system, and then install LIC. DVD’s are supported on POWER9, they will just have to be external, not internal, and take up some additional rack space.

Inside the cabinet, POWER9 servers also look much the same as their POWER8 counterparts. There are the same number of internal disk bays (18). POWER9 S914 servers have 8 adapter slots in the system unit, which is 1 more than the similar POWER8 S814 model. On a 6-core POWER8 or POWER9, all 18 disk bays are available. POWER8 or POWER9 4-core servers are restricted to a maximum of 10 internal disk drives. POWER9 servers have a mix of PCIe GEN3 and GEN4 adapter slots while POWER8 servers only have PCIe GEN3 adapter slots. The main difference is performance; POWER9 servers are delivering about 40% more CPW than their POWER8 counterparts. Memory is also different. POWER8 scale-out servers used buffered DDR3 and DDR4 memory, while POWER9 models use I/S (Industry Standard) DDR4 memory. The buffered memory is a little faster, but also much more expensive, 22 to 44%. The lower memory costs will be an advantage for IBM i customers, IBM i loves memory, the more the better. POWER8 servers allow the attachment of more external adapters through “Fan Out” modules in the Gen3 I/O expansion draws, but that does not seem to be a significant issue these days as fewer adapters are needed with more and more I/O functions being built into the motherboard.

IBM is touting the POWER9 servers as being “Cloud Ready”, but what does that mean? In order to deploy a server in a cloud environment, physical resources like CPW, memory, disk and Ethernet need to be virtualized. PowerVM Enterprise Edition software is included at no charge with all POWER9 servers, although software support for it is a separate charge. PowerVM provides a layer of virtualization between partitions running on Power servers and the physical hardware, making POWER9 servers Cloud Ready. PowerVM is required if you are using an HMC to manage your Power server. There are two PowerVM editions on POWER8, Standard and Enterprise. On POWER9 servers, only Enterprise is available. The standard edition provides virtualization and HMC support. The Enterprise Edition provides all the function of the Standard Edition plus Live Partition Mobility (moving an active LPAR from one server to another) and Active Memory Sharing (memory pools that are dynamically shared by LPAR’s).

POWER8 vs. POWER9 Differences

  • Processor Performance POWER9 is 40% faster
  • Memory POWER9 is a little slower, but 22% to 44% less expensive
  • DVD POWER8 – internal, POWER9 – external
  • Internal Adapters POWER8 – 7 or 11, POWER9 – 8 or 11 (1-socket or 2-socket)
  • External Adapters POWER8 – 2 groups of 6 adapters, POWER9 – 1 group of 6 adapters
  • Minimum OS Supported POWER8 – V7.1, POWER9 – V7.2
  • Minimum HMC Code POWER8 – V8.810.0, POWER9 – V9R1.910
  • PowerVM Enterprise Edt. POWER8 – Optional, POWER9 – No charge
  • NVMe Flash Boot Adapter POWER8 – Not available, POWER9 – Optional (not supported by IBM i)
  • OpenCAPI GPU Accelerate POWER8 – Not available, POWER9 – Standard (not supported by IBM i)
  • Meltdown/Spectra PTF’s POWER8 – Optional, POWER9 – Pre-installed
  • Overall cost POWER8 & POWER9 servers cost about the same for similar configurations

So it looks like POWER9 is the no-brainer choice if you need a new server to run your IBM i applications – cost is about the same as POWER8, but you also get: 40% more performance, PowerVM Enterprise Edition and pre-installed protection from Meltdown/Spectra vulnerabilities.  POWER9 servers should also have IBM hardware support and IBM i Operating System support longer into the future than POWER8 servers, making POWER9 a better long term investment.  In what circumstance would you want to choose a POWER8 server?  I can only think of a couple of reasons why: 1) Your application is not supported by V7.2 or V7.3; or 2) You need more adapters on your server than POWER9 can handle.  POWER9 Scale-Out servers will start shipping by the end of March, 2018, But, orders are being taken now.  How many do you want? Contact our sales group to start working on a configuration for you.

newsletter_image2.jpg

After Upgrading from IBM i 7.1 to 7.3, PC users weren’t able to connect to file shares on IBM i.

After a recent upgrade from IBM i 7.1 to 7.3, a customer was having problems with their users connecting from their PCs and accessing file shares on the IBM i.  We thought we would share this with you.

First, make sure the Windows domain user and the IBM i user match in profile and password, and the windows user was ‘unlocked’.  If that wasn’t the problem, it could be caused by another issue going from 7.1 to 7.3.  Note the default connections at 7.1 is SMBv1 (Server Message Block), and the default at 7.3 is SMBv2.

Problem

QNTC connections might be established even though using wrong case domain passwords at SMBv1. The same connection requests are correctly rejected when using SMBv2.

Diagnosing the problem

The following situation has been noted by IBM, and appears to be the case in many situations.

A working QNTC environment was migrated from IBM i OS 7.1 (SMBv1) to IBM i OS 7.3 (SMBv2). After migration, all QNTC connections began to fail with the following message:

CPDB053 Error exchanging security information for user &2 on network server &5. The message was returned with Error Class 0, and Error code 3401.

Error code 3401 (CPE3401) means permission denied.

QPWDLVL was set to 0 on the IBM i.

The NetServer domain was a valid Windows domain.

On the PC side, the users were Domain Users.

The same Domain user accounts were still able to access QNTC from other IBM i partitions that were running OS 7.1.

It was found that the Domain User Accounts had passwords where the alphabetic characters were upper case. QNTC should never have been able to make a connection (at any OS/400 version or at any SMB version), because with the IBM i at QPWDLVL 0, all QNTC passwords are sent with lower case alphabetic characters.

A password verification tool was used to confirm that the IBM i was correctly sending QNTC passwords using lower case alphabetic characters.

Resolving the problem

This incorrect behavior is an issue with the Domain logon accepting the wrong case password for SMBv1 connections. Anyone wanting to address the root cause should contact Microsoft.

If this same (incorrect) behavior is desired at OS 7.3, then QNTC must be set to use SMBv1. We don’t recommend this, as this leaves you vulnerable.

Adding the QIBM_ZLC_SMB_VERS environment variable (with a VALUE of 1) at the system (*SYS) level, configures QNTC to use SMBv1 rather than SMBv2. Which is not recommended. The following command is used to make the change:

ADDENVVAR ENVVAR(QIBM_ZLC_SMB_VERS) VALUE(1) LEVEL(*SYS)

Once the QIBM_ZLC_SMB_VERS has been added, the user must sign on to a new emulation session in order to use SMBv1.

QNTC will use SMBv1 at this time.

NOTE: Using SMBV1 is not expected to be satisfactory for a long term solution, because if SMBv1 were to later be disabled on the Network, then QNTC would have to be changed back to SMBv2 and some other method used instead. Therefore, the use of SMBv1 should be considered a temporary circumvention and one of the following options should be considered for a long term solution:

  1. Switch to using lower case alphabetic characters on the Domain accounts. This might not be a long term solution either, because many companies are switching to more complex passwords which require a mixture of upper and lower case alphabetic characters. Mixed case never has worked for QNTC.
  2. Implement QPWDLVL 2 or 3 on the IBM i, after careful investigation to confirm that this will work well with all computers that need to connect with the i.  This is not a simple change, and you must understand the considerations that have to be planned for when making this change.
  3. Implement EIM/NAS for IBM i (kerberos) on the network.

If you need a hand with this, or want someone to make this change on your system, then contact iTech Solutions.

newsletter_image3.jpg

What does PowerVM Enterprise Edition Bring to the Table?

One of the differences between POWER8 and POWER9 hardware is the edition of PowerVM that is included.  The POWER9 announcement states that all POWER9 hardware will come with PowerVM Enterprise Edition included.  We thought it would be helpful to explain what you get above and beyond the Standard Edition.  There are two main features which are included in the Enterprise Edition; Live Partition Mobility and Active Memory Sharing (AMS).

What is PowerVM?

PowerVM is a virtualization solution which is required, if you plan to create partitions on your system.  PowerVM and Power Systems make it possible for you to share system resources across applications, or environments.  Before partitioning was available, companies had separate Power Systems for different workloads.  The result was a lot of wasted system resources.  Now with partitioning we can share memory, processors and I/O resources allowing for maximum resource utilization.

What’s in PowerVM Enterprise Edition?

POWER8 was shipped with the PowerVM Standard Edition, which met the needs of most shops that partition their hardware.  However, with a new model of hardware being rolled out, migration becomes an issue.  Companies have to decide when they can migrate based on when they can take the system offline; if they can take it offline.  For those companies who have high availability needs, Live Partition Mobility offers a huge benefit, migration without shutting down the applications.

Live Partition Mobility allows you to migrate your workload without affecting the users.  The partition can be moved from one system to another, while the users and their applications remain active.  This is also helpful when you need to perform system maintenance.  You can move the partition to a different system, perform the maintenance and move it back.  Without the users ever even knowing.

Another feature you will gain from having the Enterprise Edition is Active Memory Sharing.  Active Memory sharing allows memory to be shared in a pool which can be accessed by all the partitions.  This happens dynamically, without your intervention.  In addition there is Active Memory Deduplication. This feature also helps optimize memory utilization by deleting duplicate memory pages.

We Can Help

Live Partition Mobility is a great feature, but like all IT projects requires good planning.  There are some prerequisites and you really need to have external storage defined.  iTech has the experience, knowledge and know-how to help you take advantage of all the features that your IBM i has to offer. Call us for help with your migration project today and let us show you how we can help you, contact iTech Solutions.

newsletter_image4.jpg

Change IPL Attributes

I have been at a few customers recently, who didn’t know that you can IPL to a restricted state just by using the Change IPL Attributes command, CHGIPLA. In fact, there are quite a few things that this command can set and perform, so I thought it was of interest to include this in an article. The CHGIPLA command is used to change attributes or functions on the next IPL. I like to use this when I don’t want the system to startup after an IPL, I use the last parameter Start to Restricted State, and I change its default of *NO to *YES. This prevents any of the subsystems from starting, and I come back to a restricted state after an IPL. I find this very handy to run this command after I have loaded all my PTFs, and now I want to IPL to apply them all. Then when the IPL is complete, I can check to make sure everything applied, and then perform my save. I think you will also find some great uses of this command when you want to IPL but not have the system restart after the IPL.

Here are all the parameters for CHGIPLA.

Restart type (RESTART)

Specifies the point from which the initial program load (IPL) restarts when RESTART(*YES) or RESTART((*YES *IPLA)) are specified on the Power Down System (PWRDWNSYS) command. Specifying *SYS rather than *FULL can reduce the time required to restart the system. The initial (shipped) value for this parameter is *SYS.

*SYS -The operating system is restarted. The hardware is restarted only if a PTF that requires a hardware restart is to be applied.

*FULL – All portions of the system, including the hardware, are restarted.

Keylock position (KEYLCKPOS)

Specifies the keylock position. A change to this attribute takes effect immediately. The following restrictions apply when the keylock position is being changed:

The keylock position cannot be set to MANUAL for security reasons.

If the keylock position is currently SECURE, the keylock position cannot be changed.

If the system has a keylock and the key is not inserted, the keylock position cannot be changed.

*NORMAL – The keylock position is changed to NORMAL.

*AUTO – The keylock position is changed to AUTO.

*SECURE – The keylock position is changed to SECURE.

Hardware diagnostics (HDWDIAG)

Specifies whether certain hardware diagnostics should be performed during the IPL. The list of diagnostics is pre-determined by the system and cannot be modified by the user. On a partitioned system, this IPL attribute can only be changed from the primary partition or the hardware management console. The initial (shipped) value for this attribute is *MIN.

*MIN – The minimum set of hardware diagnostics is run.

*ALL – All hardware diagnostics are run.

Compress job tables (CPRJOBTBL) 

Specifies when the job tables should be compressed to remove excess unused entries. The Display Job Tables (DSPJOBTBL) command can be used to determine the number of unused entries in the job tables. Excessive unused entries can indicate that job tables need to be compressed. However, compression can take a significant amount of time during IPL. The initial (shipped) value for this attribute is *NONE.

*NONE – The job tables are not compressed during any IPL.

*NORMAL – The job tables are compressed during normal IPLs only.

*ABNORMAL – The job tables are compressed during abnormal IPLs only.

*ALL – The job tables are compressed during all IPLs.

Check job tables (CHKJOBTBL) 

Specifies when particular damage checks on the job tables should be performed. The initial (shipped) value for this attribute is *ABNORMAL.

*ABNORMAL – The job table checks are performed during an abnormal IPL only.

*ALL – The job table checks are performed during all IPLs.

*SYNC – The job table checks are performed synchronously during all IPLs.

Rebuild product directory (RBDPRDDIR)

Specifies when the product directory information should be fully rebuilt. A full rebuild is required during the IPL following the install of the operating system. This attribute is overridden during that IPL. The initial (shipped) value for this attribute is *NONE.

*NONE – The product directory information is not fully rebuilt.

*NORMAL – The product directory information is fully rebuilt during a normal IPL only.

*ABNORMAL – The product directory information is fully rebuilt during an abnormal IPL only.

*ALL  – The product directory information is fully rebuilt during all IPLs.

Mail Server Framework recovery (MSFRCY)

Specifies whether Mail Server Framework recovery is done during abnormal IPLs. The initial (shipped) value for this attribute is *NONE.

*NONE – Mail Server Framework recovery is not done during IPL. Recovery is done when Mail Server Framework starts.

*ABNORMAL – Mail Server Framework recovery is done during an abnormal IPL.

Display status (DSPSTS)

Specifies whether the status of OS/400 IPL steps is displayed at the console during IPL. Status is not displayed during install IPLs or when the console is not powered on. The initial (shipped) value for this attribute is *ALL.

*SYS – The status of OS/400 IPL steps is displayed during attended IPLs and abnormal IPLs.

*NONE – The status of OS/400 IPL steps is not displayed during IPL.

*ATTEND – The status of OS/400 IPL steps is displayed during attended IPLs.

*ABNORMAL – The status of OS/400 IPL steps is displayed during abnormal IPLs.

*ALL – The status of OS/400 IPL steps is displayed during all IPLs, excepted as noted above.

Start TCP/IP (STRTCP)

Specifies whether the Start TCP/IP (STRTCP) command is automatically submitted at the completion of IPL and when the controlling subsystem is restarted to bring the system out of the restricted state. The STRTCP command is not submitted during install IPLs or when the system is starting to the restricted state. See the STRTCP command help for more information. The initial (shipped) value for this attribute is *YES.

*YES – STRTCP command is automatically submitted.

*NO – STRTCP command is not automatically submitted.

Clear job queues (CLRJOBQ)

Specifies whether or not to clear the jobs from all job queues. The initial (shipped) value for this attribute is *NO. This attribute is reset to its initial value after each IPL.

*YES -The job queues are cleared during the IPL. This attribute is then reset to *NO.

*NO – The job queues are not cleared.

Clear output queues (CLROUTQ)

Specifies whether or not to clear all output queues, thus removing all spooled output from the system. The initial (shipped) value for this attribute is *NO. This attribute is reset to its initial value after each IPL.

*YES – The output queues are cleared during the IPL. This attribute is then reset to *NO.

*NO – The output queues are not cleared.

Clear incomplete joblogs (CLRINCJOB)

Specifies whether or not to delete the joblogs for jobs that were active at the time of the last system power down. The initial (shipped) value for this attribute is *NO. This attribute is reset to its initial value after each IPL.

*YES – Thejoblogs are deleted during the IPL. This attribute is then reset to *NO.

*NO – The joblogs are not deleted.

Start print writers (STRPRTWTR)

Specifies whether or not print writers should be started at IPL time. The initial (shipped) value for this attribute is *YES. This attribute is reset to its initial value after each IPL.

*YES – Print writers are started at IPL time.

*NO – Print writers are not started at IPL time. This attribute is then reset to *YES.

Start to restricted state (STRRSTD)

Specifies whether or not the system should be started to the restricted state, which means that only the system console will be active. The initial (shipped) value for this attribute is *NO. This attribute is reset to its initial value after each IPL.

*YES – The system is started to the restricted state. This attribute is then reset to *NO.

*NO – The system is not started to the restricted state.

The CHGIPLA command I find very useful for compressing job tables (see iTech Newsletter January 2012), not starting Print writers, or IPLing into a restricted state.  This command certainly makes my final four.

We have pointed this out to a few customers during some recent health checks. The iTech Solutions Health Check is to identify problems you might have, best practices you might not be following, identify performance issues, security issues, etc. Yet, sometimes it just takes someone to see the very simple things that you are doing wrong. For help or to have an iTech Solutions Health Check contact our sales staff.

NewsletterApril2013.png

Events

April 23 – 25, 2018 – Northeast User Group Conference, Sheraton Framingham, MA

Visit the iTech Solutions booth in the Expo and learn how we can help you get the most out of your Power Systems running IBM i. In addition, hear Steve Pitcher and Pete Massiello speak on Various System Administration topics.

Pete will be presenting:

  • What you need to know when upgrading to IBM i 7.2 & 7.3
  • Tip and Tricks to Improve System Performance & Save Disk Space
  • Cool Things in IBM Navigator for IBM i to Help you be a Rock Star Administrator
  • Step by step guide to IBM i Hosting IBM i.

Steve will be presenting on:

  • Anatomy of a Security Failure
  • IBM i and our False Sense of Security
  • Rapid Fire Admin

Monday April 30 to Friday May 4, IBM Technical University Hilton Orlando Resort, Orlando, FL

  • What you need to know when upgrading to IBM i 7.2 & 7.3
  • Tip and Tricks to Improve System Performance & Save Disk Space
  • Cool Things in IBM Navigator for IBM i to Help you be a Rock Star Administrator
  • Step by step guide to IBM i Hosting IBM i.

Wednesday May 9, 2018 – Long Island Systems Users Group – Westbury Manor, Westbury, NY

Visit Charlie at our booth and hear Pete Massiello on an IBM i Experts panel.

May 20 – 23, 2018 – COMMON Annual Conference & Expo, POWER-UP18 at Marriott River Center, San Antonio, TX

Visit the iTech Solutions booth in the Expo and learn how we can help you get the most out of your Power Systems running IBM i.

Pete will be presenting:

  • What you need to know when upgrading to IBM i 7.2 & 7.3
  • Tip and Tricks to Improve System Performance & Save Disk Space
  • Cool Things in IBM Navigator for IBM i to Help you be a Rock Star Administrator
  • HMC, IBM i, FSP, & Firmware: Putting All the Pieces Together.
  • Step by step guide to IBM i Hosting IBM i.

Steve will be presenting on:

  • Anatomy of a Security Failure
  • IBM i and our False Sense of Security
  • POWER Panel: Partnering for Success
  • YIPs Roundtable
  • Rapid Fire Admin

Yvonne will be presenting on:

  • Introduction to Software Testing
  • Development Lifecycles and How to Test them

Sept 27, 2018 – VTMUG Technical Conference – Double Tree Hotel Burlington, VT

Visit Laurie & Paul at our booth and learn how iTech Solutions can help you with your IBM i. Pete Massiello will be speaking on various Systems Management sessions to be determined.

Oct 15 – 17, 2018 – COMMON Fall Conference & Expo, Pittsburgh Marriott City Center, Pittsburgh, PA  Booth #20

Visit the iTech Solutions booth in the Expo and learn how we can help you get the most out of your Power Systems running IBM i. In addition hear Pete Massiello, Yvonne Enselman, and Steve Pitcher speak on a variety of IBM i System administration topics.

Oct 15 – 18, 2018 – Jack Henry Annual Conference – Gaylord Texan Hotel, Grapevine, TX

Visit the iTech Solutions booth in the Expo and learn how we can help you get the most out of your Power Systems running IBM i

newsletter_image5.jpg

Release levels and PTFs

People are always asking me how often they should be performing PTF maintenance, and when is the right time to upgrade their operating system. I updated this article from last month with the current levels of PTFs. Let’s look at PTFs. First, PTFs are Program Temporary Fixes that are created by IBM to fix a problem that has occurred or to possibly prevent a problem from occurring. In addition, some times PTFs add new functionality, security, or improve performance. Therefore, I am always dumbfounded as to why customers do not perform PTF maintenance on their machine at least quarterly. If IBM has come out with a fix for your disk drives, why do you want to wait for your disk drive to fail with that problem, only to be told that there is a fix for that problem, and if you had applied the PTF beforehand, you would have averted the problem. Therefore, I think a quarterly PTF maintenance strategy is a smart move. Many of our customers are on our quarterly PTF maintenance program, and that provides them with the peace of mind of knowing their system is up to date on PTFs. Below is a table of the major group PTFs for the last few releases. This is what we are installing for our customers on iTech Solutions Quarterly Maintenance program.

7.3 7.2 7.1 6.1 V5R4
Cumul Pack 18025 18032 17192 15063 12094
Tech. Refresh  4 8 11
Grp Hipers 54 117 218 210 204
DB Group 9 21 43 33 33
Java Group 7 15 30 41 34
Print Group 3 13 31 49
Backup/Recov. 18 40 73 61 57
Blade/IXA/IXS 1 16 30 15
HTTP 13 26 52 46 36
TCP/IP 3 10 17 22
Security 25 60 80 60  33
High Availability 5 8 15  5
Hardware 14 31 41  17
Open Source 6 6 6

The easiest way to check your levels is to issue the command WRKPTFGRP. They should all have a status of installed, and you should be up to the latest for all the above, based upon your release. Now there are more groups than the ones listed above, but these are the general ones that most people require. We can help you know which group PTFs you should be installing on your machine based upon your licensed programs. Here is a nice tidbit. The Cumulative PTF package number is broken down as YDDD, where Y is the year and DDD is the day it was released. Therefore, if we look at the cumulative package for 7.3, the ID is 18023. We can determine that it was created on the 23rd day of 2018, which is January 23, 2018. Look at your machine and this will give you a quick indication of just how far out of date in PTFs you may be.

HMCs

If you have a Hardware Management Console (HMC,) you should be running:

Model Release Service Pack End of Service
HMC (7063-CR1)

Power Based

V9R1.910 No PTFS Not Announced
HMC (7063-CR1)

Power Based

V8R8.7
  1. MH01707
  2. SP1 MH01726
  3. MH01749
Not Announced
HMC (CR7 & above)

Power Based

V9R1.910 No PTFS Not Announced
HMC (CR7 & above) V8R8.7
  1. MH01706
  2. SP1 MH01725
  3. MH01748
Not Announced
HMC V8R8.6
  1. MH01655
  2. SP2 MH01690
  3. MH01742
10/31/2018
HMC V8R8.5
  1. MH01617
  2. SP3 MH01689 (must be installed from classic GUI or command line)
  3. MH01721
  4. MH01730 (can be installed on any SP, required for upgrade to 8.8.6 or 8.8.7).
05/31/2018
HMC V8R8.4
  1. MH01560 (must be installed from command line using UPDHMC)
  2. SP3 MH01652 (must be installed from command line using UPDHMC)
  3. MH01730
  4. MH01729 (must be installed from classic GUI or from command line; can be installed on any SP, required for upgrade to 8.8.6)
11/30/2017
HMC V8R8.3
  1. SP3 MH01619
  2. MH01717
07/31/2017
HMC  V8R8.2
  1. SP3 MH01583
  2. MH01688
10/31/2017
HMC (CR4 last release) V7R7.9
  1. SP3 MH01546
  2. MH01587
  3. MH01687
12/30/2016
HMC V7R7.8
  1. SP1 MH01397
  2. SP2 MH01432
  3. MH01570
10/31/2015
or V7R7.7
  1. SP3 MH01379
  2. SP4 MH01415
  3. MH01516
02/28/2015
HMC C03 V7R3.5
  1. SP4 MH01277
05/31/2014

If we have a model listed above in the HMC column that is the highest level of firmware that model of the HMC can be upgraded to.

  • Note that release 8.8.x does not support any POWER5 servers.
  • Version 7.7.9 is not supported as of 12/30/2016 and cannot be installed on HMC models C03, C04 or CR2.
  • If an HMC is used to manage any POWER7 processor based server, the HMC must be a model CR3 or later model rack-mount HMC or C05 or later desk side HMC.
  • HMC V8R8.1 is supported on rack-mount models CR5, CR6, CR7 and CR8; and on desktop model C08. These listed models meet or exceed the V8R8.1 minimum memory requirement of 2GB however 4GB is recommended.
  • If you want to manage a POWER8 machine, you need to be on at least HMC 8.8.1

Some notes on the new HMC release V8R8.6 that just came out:

  • Will be the last release to support POWER6.
  • Will be the last release to allow ‘classic’ UI login.
  • Will be the last release that supports the model CR5, CR6 and C08.
  • The HMC must be at version V8 R8.4.0 or later to be upgraded to HMC V8 R8.6.0. This requirement is enforced during installation.

 

If you have a Flexible Service Processor (FSP) your firmware should be:

Machine Processor Model Version End of Service
POWER5 or 5+ 520, 515, 525, 550, 570 SF240_418_382 11/30/2012
POWER6 940x, M15, M25, M50 EL350_176_038 01/31/2017
8203-E4A, 8204-E8A, 8204-E4A EL350_176_038 01/31/2017
MMA, 560, 570 EM350_176_038 01/31/2017
9119-FHA EH350_176_038 01/31/2017
POWER7 8231-E1B, 8202-E4B, 8231-E2B, 8205-E6B, 8233-E8B, 8236-E8C AL730_154_035 (last release)

AL730_157_035  **Includes Meltdown and Spectre Fixes **

08/09/2017
9117-MMB, 9179-MHB AM780_091_040 (last release)

AM780_096_040 **Includes Meltdown and Spectre Fixes **

8231-E1C, 8202-E4C, 8205-E6C AL740_163_042 (last release)

AL740_165_042 **Includes Meltdown and Spectre Fixes **

9117-MMC, 9179-MHC AM770-119_032 (last release)

AM770_122_032 **Includes Meltdown and Spectre Fixes **

POWER7+ 8231-E1D, 8202-E4D, 8231-E2D, 8205-E6D AL770_119_032

AL770_122_032  **Includes Meltdown and Spectre Fixes **

8408-E8D, 9109-RMD AM770_119_032

AM770_122_032  **Includes Meltdown and Spectre Fixes **

9117-MMD, 9179-MHD AM780_091_040

AM780_096_040  **Includes Meltdown and Spectre Fixes **

POWER8 8408-E8E, 8284-21A,  8284-22A, 8286-41A, or 8286-42A SV860_127_056 (If HMC requires HMC 8.8.6+)

SV860_138_056  **Includes Meltdown and Spectre Fixes ** (If HMC requires HMC 8.8.6+)

9119-MHE or 9119-MME SC860_127_056

SC860_138_056  **Includes Meltdown and Spectre Fixes **

 

If you need help with upgrading your HMC or FSP just give us a call. We will be happy to perform the function for you or assist you in doing it. Contact Pete Massiello.

Leave a Reply