On 26 Aug 2013 11:30, Tom E Stieger wrote:
Yeah, now I have to go find any of MY old code that may still be
using sysdummy1!
Huge difference between SQL _code_ referencing SYSDUMMY1 and a VIEW
referencing that TABLE. Albeit given the poor [IMO defective] design of
any post-IPL PTF processing to effect DROP SYSIBM.SYSDUMMY1 then there
is still a possibility that any code running after the DROP and before
the re-create completes, still could be impacted. While the overall
design\implementation makes that unlikely, it is not impossible. AFaIK
the amount of pending work in the job wherein that activity runs is
unpredictable and could delay until long after the user portion of an
IPL starts; if the system jobs have implemented a high-priority enqueued
work path, then the potential is greatly reduced, but I just read an
APAR with MCH5801 as a symptom, from which I infer there is still no
such priority path for the enqueued SYSIBM PTF activity.
The SYSDUMMY1 should have been designed like the [effectively
identical in purpose] database file QSQPTABL in QSYS2, such that the
file would almost never be deleted except in an actual /upgrade install/
[or when the required attributes or its data are incorrect]. The code
could easily check if the TABLE already exists and inferred by some
verifications to be functional, and only DROP the file *if* there is
something wrong with the object or its data. Why anyone would code the
DROP as unconditional, is unfathomable to me.
Regardless, using a user file or the row values-clause would
eliminate any dependence on that system database file. However the
/implementation/ of a row values-clause could be a query cursor over the
system-table QSQPTABL in QSYS2. If some bone-headed PTF activity
against that QSQPTABL file were to mimic the daft effect exhibited
against the SYSDUMMY1 TABLE, then the same potential impacts would
exist, though with an even more negative consequence because even some
row value-clause invocations could fail; i.e. those implemented using a
cursor on the QSQPTABL.
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact
[javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.