In this DocumentPurpose
Last Review Date
Instructions for the Reader
Troubleshooting Details
Applies to:Oracle GoldenGate - Version: 4.0.0 to 10.4.0.70 - Release: 4.0.0 to 10.4.0
Information in this document applies to any platform.
PurposeOccasionally, Replicat will show a status of running but a report will show no records have been updated. There are 3 common causes for this. Last Review DateJune 14, 2010 Instructions for the Reader
A Troubleshooting Guide is provided to assist in debugging a specific issue. When possible, diagnostic tools are included in the document to assist in troubleshooting.
Troubleshooting DetailsThe 3 causes are:
1) The wrong trail is being read
2) The wrong trail file is being read
3) The table name format is incorrectly specified
Each of these cases will show a status of running. A review of the report will show that no records have been replicated.
Case 1: The wrong trail is being read
Do an info detail of the extract or look at the rmttrail statement in the extract params
Do an info detail of the replicat or review the replicat report
The replicat trail file being read must be the same name, machine and location of the trail the extract is writing.
Common errors are specifying the wrong target machine or the wrong subdirectory for the trail
Erroneous extract param files and add exttrail or add replicat statements are often the cause
This is a setup error and needs to be reconfigured
Case 2: The wrong trail file is being read
Do an info detail of the replicat or review the replicat report
Examine the trail files (ls -l) on the target.
Does the trail file being read exist?
Is the replicat at the end of a trail file not advancing to the next available file?
The solution is to do an > alter rep <repname>, extseqno <NNNN>, extrba 0 to cause the replicat to advance to the next file and restart the replicat
This is commonly caused by doing an etrollover on the source. This generally requires doing an alter on all downstream processes. This is an administrative error.
Case 3: The table name format is incorrectly specified
The schema.table name in the replicat param file must match the schema.table name in the trail.
Do a logdump on a trail file and do a count command with detail turned on
Or use logdump to look directly at a row to be applied with ghdr turned on
The table.schema in the replicat must match that in the trail file. On NSK, this includes matching any system name occurrence or absence.
Additionally, changes may be made on the source side that may change the table specification
This is commonly a setup error that is easily corrected. Adjust the replicat specification, repoint the replicat to the beginning of data and rerun.