3

Ignoring HOLDDATA instructing me to do so, is there a specific module that would be modified by maintenance that should signal me that I need to write IPLTEXT to my IPL volume?

I typically just see if SYS1.NUCLEUS gets modified and write IPLTEXT regardless, but I wasn't sure if there was a 100% reliable way to know in case the HOLDDATA neglects to mention it.

Steven Perva
  • 78
  • 7

2 Answers2

4

Hopefully no one outside of IBM is updating the IPLText and when they do a HOLDDATA is the way they should communicate it.

Now if you have IPLTEXT for other than z/OS then you'll have to check with those (i.e. the stand alone editor ZZSA found on www.cbttape.org in File 538).

Lionel B Dyck
  • 552
  • 2
  • 9
  • I don't know how secureboot (secureipl) will play into this. – Lionel B Dyck Jun 28 '23 at 14:09
  • 1
    see https://www.ibm.com/docs/en/zos/2.5.0?topic=zos-what-is-validated-boot – Lionel B Dyck Jun 28 '23 at 14:11
  • 3
    For Validated Boot, you really need to plan on re-signing everything every time you install service. Strictly speaking, you don't have to, but it's easiest to just include signing into the process and then not worry about it going forward. – Kevin McKenzie Jun 28 '23 at 14:16
3

There's a wait state 019 that I only saw once in my career when a sysprog accidentally pointed their ICKDSF IPLTEXT JCL to SYS1.SAMPLIB on the previous OS level instead of the new OS level. So other than that mistake, I don't think the system will complain at IPL time about other possible updates, which I think is what you are addressing.

What about checking SYS1.SAMPLIB(IPLRECS) and SYS1.SAMPLIB(IEAIPL00) for updates, which (unless things have changed since I last wrote out IPLTEXT) are the source for the IPL records on the res pack?

Tom Brennan
  • 156
  • 1
  • 3