Best Of

The KB351986 - E1: WRKFLW: Work Center Not Displaying Messages Correctly Due to Bad Data article is indeed for purging the tables (removing all data) since at step 8th, the next number of workcenter messages is reset to 1.

Please note that JN19220 ESU is delivering a redesigned R01131P which runs faster and no more orphan records are left behind in F01131M (no need to run R01131M afterwards). The same fix is delivered since UN6 onwards.

The overall performance of R01131P is given by the total number of records required to be deleted (it may spend some time to delete data in F01131M) and the overall performance of the database, so please check also if the data is fragmented at database level and if you need to rebuild any statistics.

It is strongly recommended to run R01131P instead of directly manipulate the data in the tables using queries.

However, you can always delete the records from the tables based on the tickler date. Here are the queries for an Oracle Database:

DELETE FROM PRODDTA.F00166 A WHERE EXISTS (SELECT * FROM PRODDTA.F01131 B WHERE B.ZZSERK = A.GTTXKY AND A.GTOBNM='GT01131' AND B.ZZDTI<XXXXXX);

DELETE FROM PRODDTA.F00165 A WHERE EXISTS (SELECT * FROM PRODDTA.F01131 B WHERE B.ZZSERK = A.GDTXKY AND A.GDOBNM='GT01131' AND B.ZZDTI<XXXXXX);

DELETE FROM PRODDTA.F01131T A WHERE EXISTS (SELECT * FROM PRODDTA.F01131 B WHERE B.ZZSERK = A.ZCSERK AND B.ZZDTI<XXXXXX);

DELETE FROM PRODDTA.F01133 A WHERE EXISTS (SELECT * FROM PRODDTA.F01131 B WHERE B.ZZSERK = A.ZTSERK AND B.ZZDTI<XXXXXX);

DELETE FROM PRODDTA.F01131M A WHERE EXISTS (SELECT * FROM PRODDTA.F01131 B WHERE B.ZZSERK = A.ZMPPSRK AND B.ZZDTI<XXXXXX);

DELETE FROM PRODDTA.F01131 WHERE ZZDTI<XXXXXX;

where XXXXXX represents the JDE Julian Date.

Thank you!

Best regards,
Cezar

CezarB-Oracle

1

Re: Table size Huge - records are low

Hello SGU,

another hypothesis is that the table is experiencing things like "DELETE all_or_many_rows" followed by "SQL Loader direct path", inserting data in new extents even if there is a lot of free space in existing extents. If this is repeated, we end up with a table way too large compared to the actual volume of data.

You might have to reorg your table, there are various ways including ONLINE MOVE, but to have an idea of the potential gain I would perform a simple small test with "CTAS" on a sample (example 1 percent):

CREATE TABLE xxttd.one_percent AS SELECT * FROM xxttd.xxttd_fa_invest_reg_tbl SAMPLE( 1 );

this should create a table with approx 840 rows. Look at the size of one_percent and extrapolate, to know if reorg would make a big difference (if yes, reorg the table as a "one shot" operation and try to identify the cause of the "unnecessary growth" to work on it). If the one_percent table is still amazingly large, then you have to investigate in another direction. (don't forget to drop one_percent at the end of the test)

HTH,

Bruno Vroman.

BrunoVroman

1

Solaris 11.4 SRU 90: Scan with pending free

Solaris 11.4 SRU 90 now does resilvering and scrubbing, while freeing space is pending.

Oracle Solaris had reached ZFS pool version 53 so far. With SRU 90, version 54 was added.

For more details, please visit: https://www.c0t0d0s0.org/blog/scanfree.html

Prashant Deo-Oracle

1

Re: Deduction Calculation Using Age Graded Benefit Rate

Hello, The COBOL you are likely looking for is PSPDCEMP, which pulls in all of the personal data for calculations.

Regards - Beth

Beth S - Oracle-Oracle

1

Re: Inquiry regarding access to the old Oracle Support screen (screenshot attached)

Thank you for checking.

I confirmed that when searching by entering the desired keyword, data did not appear in the 'Knowledge' section at first, but the list appeared after a few minutes.

I struggled a bit because no values were displayed in the 'Knowledge' section at first.

thank you!

User_S23EX

1