OS/400 V5R3 PTFs Can Corrupt Licensed Internal Code
June 19, 2006 Doug Bidwell
Unlike many of you, I spend a great deal of time–some would say an inordinate amount of time–dealing with PTFs for the OS/400 and i5/OS operating systems. And even a techie like me gets burned every now and then. Be careful with Cumulative Group PTF C6101530, which I call “cume” 6101 for short. If you follow the current installation instructions that are online at IBM‘s Fix Central site, you will be OK. But if you follow the instructions that ship with the PTF order, you could in trouble.
The basic problem is that IBM issued from PTFs, and they were defective. This is never a good situation, but all operating system vendors have this problem. It follows the Rule of Unintended Consequences: You fix one problem, and you cause an entirely different one because computers, their setups, and their applications are complex.
The following is an excerpt from SF98088: Defective PTFs, an IBM Preventive Service Planning document:
36. 16 May 2006 DEFECTIVE PTF(s): MF34369, MF38536, MF38576, MF39123, and/or MF39361 for V5R3M0 MF35091, MF38575 and/or MF38658 for V5R3M5 LICENSED PROGRAM = 5722999 APAR NUMBER: MA33716 USERS AFFECTED: All i5/OS users with PTFs MF34369, MF38536, MF38576, MF39123, and/or MF39361 applied for V5R3M0. All i5/OS users with PTFs MF35091, MF38575 and/or MF38658 applied for V5R3M5. If any of the above PTFs are temporarily applied, perform the following command to remove the PTFs: RMVPTF LICPGM (5722999) SELECT (MFXXXXX) RMV(*PERM) RMVDEP (*YES). If any of the above PTFs are permanently applied, a restore of the Licensed Internal Code will be required. LIC PTFs MF34369, MF38536, MF38576, MF35091, MF38575 and MF38658 are included in cumulative PTF Package C6101530. REASON DEFECTIVE: The Work with disk unit screens under Dedicated Service Tool (DST) and System Service Tool (SST) are in English, not the installed National Language Version (NLV). Under some conditions, the Work with disk unit screens become unusable for all languages. RECOMMENDATION: The defective PTFs MUST all be removed before applying the fixing LIC PTFs. Apply LIC PTF MF39611 when it becomes available for V5R3M0. Apply LIC PTF MF39614 when it becomes available for V5R3M5.
In that long list of stuff, here’s the sentence that has the gotcha: “If any of the above PTFs are permanently applied, a restore of the Licensed Internal Code will be required.” Superseded status also means that it was applied permanently. And, the gotcha is that putting on the latest cume over the 6101 cume will hose you as well.
There’s a whole bunch to tell you on this one, but, we’ll start with first telling you how to figure out if you’ve got an issue or not. Briefly, for those of you who have never done it, if you have to “restore the licensed internal code” (otherwise referred to as “slip the LIC”), you’ve got some anxiety in front of you. It’s simple, it’s well documented, and there’s lots of help out there to get you through it, but it is cause for worry. If you get into a situation with your disk drives that require the use of the screens in question, and, they come up corrupted, you then may be in a situation where you have to slip the LIC before you can trust what you see on the disk screens. This is not a pleasant position to be in, if you catch my meaning.
To determine where you are, use the command DSPPTF for each of the PTFs above (MF34369, MF38536, MF38576, MF35091, MF38575 and MF38658, all for 5722999 Licensed Program Product). If any of them come up with a status of superseded or permanently applied, you are probably into a slip the LIC situation. You can always contact your next level of support and get a confirmation.
Instructions on what is involved and steps to perform a restore of the Licensed Internal Code (which IBM may or may not call a slip of the LIC) can be found in the Backup and Recovery Guide (SC41-5304-07), Chapter 4.
If you have never done this procedure, or, if you’re the least bit anxious about doing this procedure, I urge you to seek the guidance of your next level of support for your iSeries or System i. If you are not experiencing an issue with this at this point, then I suggest you do the preparatory research, seek the guidance of your next level of support, and be patient. There is no need for panic in any of this; it’s just a matter of being prepared.
Now for the fun part: I will try to explain how this happened. The defective PTFs were first made available in Cumulative Group C6101530. At the time they were issued, they were not known to be defective. They were found to be defective about six weeks later. Many of us had installed cume 6101 by that time, and, several HIPER PTFs and other group PTFs as well. IBM posted the note in the install instructions for cume 6101 at that time to omit these PTFs when putting on cume 6101. So, the only people with issues were those that installed cume 6101 in that six-week period (yours truly and some of my customers), or those who ignored the installation instructions (yes, people do this). IBM also immediately issued cume C6142530, which automatically omitted the offending PTFs. So, today, if you have to go with a cumulative PTF update, do not put on cume 6101, put on cume 6142 instead. If you need to put on 6142, and you already have 6101 installed, putting on 6142 does not put you in any greater jeopardy.
Also, you might want to be cautious with a PTF called SF99503-530 DB2 UDB for iSeries Level 10. Level 11 is now available, so go with that instead. Level 10 has issues with defective PTFs as well.
Doug Bidwell is the president of DLB Associates, an IBM business partner that supports customers on the OS/400 platform, and creator of the System i PTF Guide, which we publish weekly as part of our Four Hundred Guru newsletter. You can reach him at firstname.lastname@example.org