×
The internal search function is temporarily non-functional. The current search engine is no longer viable and we are researching alternatives.
As a stop gap measure, we are using Google's custom search engine service.
If you know of an easy to use, open source, search engine ... please contact support@midrange.com.
FWiW the USROPT(*EXPERT) which enables that /compact view/ presented
by DSPOBJAUT for the Object and Data rights, is good for a visual, but
naming each authority value separately as /special values/ is another
option. The special values are on the GRTOBJAUT and RVKOBJAUT commands
for the AUT() parameter.
Those values listed in the presented DSPOBJAUT compact view are:
*OBJOPR, *OBJMGT, and *OBJALTER. Searching STRJRNPF or QDFTJRN with
*OBJMGT will find the following post in which Carsten suggests *OBJALTER
is required [kwd sql7905 from that post]:
http://archive.midrange.com/midrange-l/200511/msg01125.html
So apparently that is an additional object right that is required
beyond what is documented in the Redbooks document your referenced;
"Journaling At Object Creation On DB2 For iSeries" noted, but no link.
A a copy of a Technote by that name, suffixed with TIPS0604, is found at
both:
http://www.redbooks.ibm.com/abstracts/tips0604.html?Open
http://www.mgd.dk/websider/artikler/tips0604.pdf
And if so, it that seems like an appropriate use of a Reader Comment
submission, if an InfoCenter reference can even be found that states
only *OBJOPR and *OBJMGT are required, or a link from InfoCenter to the
.pdf for it not even mentioning required authority:
https://publib.boulder.ibm.com/infocenter/systems/topic/rzaki/rzaki.pdf
However it seems this may be changed in v6r1 [errr IBM i OpSys 6.1];
i did not actually read the article.
http://www.google.com/search?num=50&hl=en&lr=&as_qdr=all&q=qdftjrn+objmgt+&btnG=Search
http://systeminetwork.com/article/start-journaling-libraries-ios-61-fixing-problems-qdftjrn
And although the article implies *OBJALTER is /absurdly/ high level
of authority, that is a daft claim since that authority offers no other
capability to the *JRN object. IMO the *OBJMGT requirement could more
rightly be claimed as excessive, for what that right offers; but even
that is moot, since neither MOVOBJ nor RNMOBJ are even supported for the
*JRN object type... just as there is no /alter/ function applicable to a
*JRN.
Note that the *OBJOPR, although it enables redirecting to what Data
rights are available, IIRC, without that object right, there is no
ability even to obtain a pointer to the object.
Regards, Chuck
Jeff Crosby wrote:
Thanks for the idea, I didn't even think of QUADJRN!
QAUDJRN showed it was indeed the *JRN object causing the authority
failure. I did some EDTOBJAUT experimenting on the *JRN object and
found the following to be the minimum *PUBLIC authority (so far) I
could get away with:
-----Object------ ------Data-------
O M E A R R A U D E
X X X
(This lines up for me prior to sending the email, but just in case it
doesn't upon arrival, the 3 X's are all under the "-----Object------"
portion of the screen, under the "O", "M", and "A", respectively.
The help text says:
Object - Help
The authorities that the user has specifically to the objects. An "X"
in the column indicates that the user has the specified authority to
the object named in the row. The specific object authorities are:
O - Object operational authority provides authority to look at the
object's attributes and to use the object as specified by the data
authorities that the user has to the object.
<<SNIP>>
I'm going to try removing from under the "O" and "M" individually and
see what happens. If I read it right, I can remove it from under
the "O" 'cause *PUBLIC doesn't *have* any data rights.
midrange.com
