The following errors are observed in the event_log and *ocap* log respectively (where *ocap* log denotes the Capture log) when Capture is stopped due to error in a TDE environment:
Notice 2017-01-16 15:52:11.560107 26348 1462896480 Capture: The database supplemental logging state = Enabled, primary_key = Enabled, unique_key = Enabled (capturing from sourceSID) [module oct]
Error 2017-01-16 15:52:11.866981 26348 1250637568 Capture: Unable to decrypt redo log. Make sure that wallet is configured correctly. (capturing from sourceSID) [module oct]
Error 2017-01-16 15:52:11.868163 26348 1250637568 Capture stopped: Internal error encountered; cannot continue (capturing from sourceSID)
Info 2017-01-16 15:52:12.882188 25981 473339744 Capture exited with code=1, pid = 26348 (capturing from sourceSID)
ocap 2017-01-16 15:52:11.866745 26348 1250637568 Failed to get TDE info: error = -9000003
ocap 2017-01-16 15:52:11.866812 26348 1250637568 OlogparGetNextRecord(139750896514816): cannot get tde info: logpar_seqno=3843 olog_seqno=3843 olog_seek_seqno=3843 num_slave=0
ocap 2017-01-16 15:52:11.866936 26348 1462896480 Parsing Oracle RedoLog in 10G format (starting seqno=3843)cdbt1350
As a workaround, SharePlex was shutdown, ora_setup was run once again to confirm the TDE support, Shareplex was restarted and sp_wallet run, and Capture was restarted but it ran into the same error once again.
One reason for seeing this error is when the root password for TDE has been changed while SharePlex is running.
If the password for wallet has been changed at the database level, close the wallet and reopen it as below at the database level:
SQL>alter system set encryption wallet close identified by "<password>";
SQL>alter system set encryption wallet open identified by "<password>";
Then run sp_wallet prior to restarting Capture.