Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Exit program only PTF not applied via SAVLICPGM / RSTLICPGM

48 views
Skip to first unread message

yl_mra

unread,
Oct 26, 2009, 9:18:24 AM10/26/09
to
Repost from IBM System Programming Group after no response.

I have a product with five associated PTFs. One of the PTFs is an exit
program only PTF. All PTFs except the exit program only PTF appear to
work properly when the product is saved using SAVLICPGM and restored
using RSTLICPGM. The PTF grants special authority to a user profile
and works correctly if removed and re-applied following the RSTLICPGM.
For test purposes, all of the PTFs were temporarily applied prior to
running SAVLICPGM.

A V5R3 system was used to create the PTF to a target V5R1 system.
SAVLICPGM and RSTLICPGM was run on both V5R3 and V5R4 systems with the
same results. All PTFs work correctly when applied outside of the
SAVLICPGM and RSTLICPGM process. There are no error messages
associated with this exit only PTF after SAVLICPGM or RSTLICPGM.

I have searched the IBM web site, the IBM i Information Center and the
web for relevant information and found none. The web search included
IBM i related groups.

Information regarding this problem would be appreciated.
Mike Amos

iseriesflorida

unread,
Oct 26, 2009, 10:26:15 AM10/26/09
to

What are the requirements of the 5 PTF's, do they have co-req PTF's
etc. Could be that the requirements are met in one situaiton and not
the other.

What is the status of it on V5R4, V5R3 and then V5R1.

Dr.Ugo Gagliardelli

unread,
Oct 28, 2009, 8:50:54 AM10/28/09
to
il 26/10/2009 14.18, Scrive yl_mra 42992096:
As far as I know a PTF exit runs only at PTF application or remotion
events, so savlicpgm and rstlicpgm are not events that have the ptf exit
run.
If you need some special actions at rstlicpgm time, you should include
it into the postoperation exit program of the lic.product itself.

--
Dr.Ugo Gagliardelli,Modena,ItalyCertifiedUindoscrasherA�ejoAlcoolInside
Spaccamaroni andate a cagare/Spammers not welcome/Spammers v�o � merda
Spamers iros a la mierda/Spamers allez vous faire foutre/Spammers loop
schijten/Spammers macht Euch vom Acker/Spamerzy wypierdalac'

CRPence

unread,
Oct 29, 2009, 1:20:55 PM10/29/09
to
Dr.Ugo Gagliardelli wrote:
> il 26/10/2009 14.18, Scrive yl_mra 42992096:
>> Repost from IBM System Programming Group after no response.
>>
>> I have a product with five associated PTFs. One of the PTFs is
>> an exit program only PTF. All PTFs except the exit program only
>> PTF appear to work properly when the product is saved using
>> SAVLICPGM and restored using RSTLICPGM. The PTF grants special
>> authority to a user profile and works correctly if removed and
>> re-applied following the RSTLICPGM. For test purposes, all of
>> the PTFs were temporarily applied prior to running SAVLICPGM.
>>
>> <<SNIP>>

>>
>> Information regarding this problem would be appreciated.
>>
> As far as I know a PTF exit runs only at PTF application or
> removal events, so savlicpgm and rstlicpgm are not events that

> have the ptf exit run.
>
> If you need some special actions at rstlicpgm time, you should
> include it into the postoperation exit program of the lic.
> product itself.
>

The results are not a /problem/ because what Ugo describes
reflects the proper operation. The proper resolution to the
scenario described by the OP [which can be inferred from, but is not
explicitly stated, by Ugo] is to make two PTFs, where the first PTF
is the change to the installation exit program, and the second PTF
is the exit-program-only PTF which calls out the other [the change
to the installation program] as a pre-requisite. In that manner any
SAVLICPGM after the PTF event is handled will get re-applied by the
installation; i.e. the corrective action is now part of the restore.

IBM will by default purposely /withhold/ PTFs with such
corrective actions [those PTFs that perform an action on the system
where the PTF is applied] from PTF cumulative tapes. However those
PTFs which were designed properly to ensure the corrective was made
part of the install need not be withheld, and the developer can
override the default action to have the PTF included in the
cumulatives. The flaw in dependence on the /withhold/ method is
that there is nothing preventing anyone from ordering and then
permanently applying the corrective PTF which can then never be
applied again, such that any SAVLICPGM [or DSLO saves for the OS]
will then be in error for the corrective action having never been
applied. What IBM does for that is create a KB document for the
scripted actions to be performed to correct the installation.
Hopefully the flaw in their process makes obvious the proper method
of handling the scenario; to deliver the PTF with a pre-requisite
change to the installation process.

Regards, Chuck

yl_mra

unread,
Dec 15, 2009, 4:58:53 PM12/15/09
to
Thank you to those posting to my question. I may now understand the
problem in that the PTFs should have applied to the base exit program
rather than to the earlier PTFs that were superseded. Unfortunately,
since this problem applies to a very old product release at near end
of life, I cannot correct the base problem. However, with newer
releases I can and will keep this information in mind.

There is also the issue in that the system does the supersede chain
unless specifically designated and I allowed the system to build the
supersede chain. In the future, I expect to test with the SAVLICPGM/
RSTLICPGM save restore method to help insure that the problem is
caught before the PTF release.

Regards,
Mike Amos

0 new messages