We use standalone server - iceberg sink.
We also use Reselect Postprocessor.
The version of Debezium used is 3.2.1
We encountered a connector error, after which the connector reboots

"exceptionType": "java.sql.SQLException",
"message": "ORA-01555: snapshot too old: rollback segment number with name \"\" too small\nORA-22924: snapshot too old\n\nhttps://docs.oracle.com/error-help/db/ora-01555/",

exceptionType": "io.debezium.DebeziumException
"message": "Couldn't convert value for column USERATTRS",
"exceptionType": "org.apache.kafka.connect.errors.ConnectException",
"message": "Error while processing event at offset

{commit_scn=36464115523:1:0f001e00e8091500, txId=0f001e00e8091500, snapshot_scn=30897580459, scn=36464087267}

",
*
USERATTRS - is a CLOB field.
In that moment we have a time lag between the source and sink with reselect postprocessor turned on
Our UNDO_RETENTION is not big and reselect took data from database (not from flashback)
ArchiveLogs are exist on that moment.

See https://debezium.zulipchat.com/#narrow/channel/348250-community-oracle/topic/ORA-01555.20.2B.20ORA-22924/near/537856359