On 14-Jul-2014 12:38 -0500, Steinmetz, Paul wrote:
I found something interesting.
On my production LPAR, PENCOR05, I see:
QDBJXQ0001 *JRNRCV QRECOVERY
However, it was created on my R&D LPAR, Pencor06, I'm confused about
that. It almost matches the date I migrated from P5 to P7, which was
a total system restore, 2 days later, 05/06/12.
Creation date/time . . . . . . . . . : 05/04/12 18:30:41
Created by user . . . . . . . . . . : QSYS
System created on . . . . . . . . . : PENCOR06
And presumably the "Save/Restore information" is all blank(s). If so...
Objects in the QRECOVERY library are solely for the purpose of the OS
[essentially to reflect the current operational\recovery status of the
installed\active system and IPL], so whatever is stored in the "System
created on" attribute of the Object Information Record\Repository (OIR),
is immaterial. The objects are thus never saved as part of B&R, and
should never be [saved and then] restored to the same or another
system\partition. The point at which some of the objects in that
library [as External Objects for which an OIR is maintained] are
/created/ for a release [or IPL or phase\era\whatever] legitimately may
be generated such the OIR may not be consistent with other objects that
are instead /installed/ either with the release or with maintenance.
Aside from possibly improper restore activity... That seemingly
anomalous attribute is likely easily explained by the history of the
storage location for the Network Attributes (see CHGNETA), most probably
as it existed at the time of the D-Mode IPL. Had that storage area
contained the actual [not a pending] name of PENCOR06 instead of the
new\eventual name PENCOR05, then the latter name value likely would have
been assigned sometime later in that D-Mode IPL, overwriting the old
value. That change would occur later than when the /recovery/ object
[in this case, a *JRNRCV object] was required to be established [for the
purposes of the OS]. Note: the Journal Receiver object is created
before the Journal object; the *JRN may [and likely does?] share the
same System Created On attribute as the *JRNRCV.
If the SAVSYS or DSLO being used to install the partition that would
eventually become PENCOR05 had originated on a system named PENCOR06,
then the latent name information from the Network Attributes would be a
probable [and acceptable\legitimate] effect given that NetAtr storage
area was restored or existed\remained-on-disk prior to the creation of
that new *JRNRCV object, and that *JRNRCV object was created [and
assigned the value from that storage location, by the Librarian (LI)
component] before the [re]assignment of the partition name to the new
value of PENCOR05.