on 05-06-2011 6:56 PM
Hi
Wondered if anybody had experienced similar - I've been running RGZPFM on the GLPCA file with RPLDLTRCD and SHRUPD. What's happening whilst the re-org is running is that the profitability reports for period 1 can deliver different results even though the period 1 data is now static.
If I end the RGZPFM , the period 1 report is always correct !!! The inconsitent report has happened on more than one occasion because it is taking some days to finish the re-org which I've cancelled because of this issue. I cannot consisently produce an inconsistent report but it was definitely happening when the re-org was running and was o.k. when I cancelled.
The SQL trace on GLPCA doesn't have any ordering on it - but this should not affect the results ?
Any advice welcome.
Regards
Dave
Hi David,
I fully understand, that this occurrence is pretty bad ...
but: In my eyes is it "still a correct behaviour" ;-(
Why?
Because of the primary key constraint, a row cannot be in the table twice at the same time. So, the RGZTOOL HAS to remove the row at the end and then "a millisecond later" will add it more in the front of the table. For sure, this will happen in one commit boundary .. but: I'm sure, your report does NOT work with a higher commit level, but just reads with "uc = uncommited read" ;-(
So, this is unfortunately, a "normal behaviour", where I never thought on by now ...
Sorry for that,
Volker
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
88 | |
10 | |
10 | |
9 | |
7 | |
7 | |
6 | |
5 | |
4 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.