Skip to Main Content

GoldenGate

Announcement

For appeals, questions and feedback about Oracle Forums, please email oracle-forums-moderators_us@oracle.com. Technical questions should be asked in the appropriate category. Thank you!

How to diagnose a ridiculously old replicat Instantiation CSN?

charlesdschultzMay 22 2019 — edited Jun 7 2019

I keep coming up with really strange “no data found” errors when I try to instantiate tables. I am looking very carefully at the “view report” output, and I am having a lot of trouble understanding what the report is trying to tell me. Especially the “CSN”, which makes no sense to me.

Goldengate 12.3.0.1.4

Database is OSEE 12.2.0.1

Using DBOPTIONS ENABLE_INSTANTIATION_FILTERING in the replicat parameter file.

Currently in my source database, I have:

Minimum SCN from v$log_history: 245629214555 (May 3rd, 2019)

Resetlog SCN (v$log_history): 226853860544

When I exported data from the source, I used FLASHBACK_SCN=246223888605, which corresponds to about 1:35 pm on Tuesday, May 21. No errors were reported during expdp nor impdp.

When I fire up the replicat, it abends with no data found. And I see crazy CSN numbers that seem to have ABSOLUTELY nothing to do with the SCN I am targeting:

<$host> /u01/app/oracle/product/gg_current/dirrpt: grep "Instantiation CSN filtering is enabled on table " RREPLIND0.rpt    

2019-05-21 15:03:23  INFO    OGG-10155  Instantiation CSN filtering is enabled on table SATURN.SFRSTCR at CSN 97,936,581,522.

Question: Can you explain why/how SFRSTCR has an Instantiation CSN of 97,936,581,522?

Some of the other Instantiation CSNs make a little more sense, but only just a little (they are at least from the same year). I can't explain where this really old CSN is coming from. Help?

Comments
Post Details
Added on May 22 2019
12 comments
571 views