-
Bug
-
Resolution: Done
-
Major
-
1.9.0.Final
-
None
-
False
-
None
-
False
-
Bug report
What Debezium connector do you use and what version?
Postgres 1.9.0.Final
What is the connector configuration?
{
"connector.class": "io.debezium.connector.postgresql.PostgresConnector",
"tasks.max": "1",
"database.hostname": "HOSTNAME.us-east-1.rds.amazonaws.com",
"database.port": "5432",
"database.user": "${USERNAME}",
"database.password": "${PASSWORD}",
"database.dbname" : "DBNAME",
"database.server.name": "audit-log-prod-test",
"decimal.handling.mode": "string",
"plugin.name": "pgoutput",
"slot.name" : "debezium",
"schema.exclude.list": "journal",
"table.exclude.list": "public.audit_log_heartbeat",
"column.exclude.list": "public.user.password",
"heartbeat.action.query": "INSERT INTO audit_log_heartbeat (id, ts) VALUES (1, NOW()) ON CONFLICT(id) DO UPDATE SET ts=EXCLUDED.ts;",
"heartbeat.interval.ms": 60000,
"snapshot.mode": "never",
"topic.creation.default.replication.factor": 3,
"topic.creation.default.partitions": 1
}
What is the captured database version and mode of depoyment?
AWS RDS Postgres
Engine version: 11.7
What behaviour do you expect?
No exceptions. Debezium continues to operate.
What behaviour do you see?
2022-04-19 20:52:51 ERROR Cannot parse time/date value 'infinity', expected format 'y..y-MM-dd[ GG]'
2022-04-19 20:52:51 ERROR Producer failure
...
2022-04-19 20:52:51 ERROR WorkerSourceTask{id=xxxx-0} Task threw an uncaught and unrecoverable exception. Task is being killed and will not recover until manually restarted
org.apache.kafka.connect.errors.ConnectException: An exception occurred in the change event producer. This connector will be stopped.
at io.debezium.pipeline.ErrorHandler.setProducerThrowable(ErrorHandler.java:50)
at io.debezium.connector.postgresql.PostgresStreamingChangeEventSource.execute(PostgresStreamingChangeEventSource.java:172)
at io.debezium.connector.postgresql.PostgresStreamingChangeEventSource.execute(PostgresStreamingChangeEventSource.java:41)
at io.debezium.pipeline.ChangeEventSourceCoordinator.streamEvents(ChangeEventSourceCoordinator.java:174)
at io.debezium.pipeline.ChangeEventSourceCoordinator.executeChangeEventSources(ChangeEventSourceCoordinator.java:141)
at io.debezium.pipeline.ChangeEventSourceCoordinator.lambda$start$0(ChangeEventSourceCoordinator.java:109)
at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515)
at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264)
at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128)
at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628)
at java.base/java.lang.Thread.run(Thread.java:829)
Caused by: org.apache.kafka.connect.errors.ConnectException: Error while processing event at offset {transaction_id=null, lsn_proc=79619438026376, lsn_commit=79619438056848, lsn=79619438026376, txId=315899941, ts_usec=1650401570841182}
at io.debezium.pipeline.EventDispatcher.dispatchDataChangeEvent(EventDispatcher.java:254)
at io.debezium.connector.postgresql.PostgresStreamingChangeEventSource.lambda$processMessages$0(PostgresStreamingChangeEventSource.java:267)
at io.debezium.connector.postgresql.connection.pgoutput.PgOutputMessageDecoder.decodeDelete(PgOutputMessageDecoder.java:498)
at io.debezium.connector.postgresql.connection.pgoutput.PgOutputMessageDecoder.processNotEmptyMessage(PgOutputMessageDecoder.java:197)
at io.debezium.connector.postgresql.connection.AbstractMessageDecoder.processMessage(AbstractMessageDecoder.java:33)
at io.debezium.connector.postgresql.connection.PostgresReplicationConnection$1.deserializeMessages(PostgresReplicationConnection.java:508)
at io.debezium.connector.postgresql.connection.PostgresReplicationConnection$1.readPending(PostgresReplicationConnection.java:500)
at io.debezium.connector.postgresql.PostgresStreamingChangeEventSource.processMessages(PostgresStreamingChangeEventSource.java:207)
at io.debezium.connector.postgresql.PostgresStreamingChangeEventSource.execute(PostgresStreamingChangeEventSource.java:169)
... 9 more
Caused by: org.apache.kafka.connect.errors.ConnectException: java.time.format.DateTimeParseException: Text 'infinity' could not be parsed at index 0
at io.debezium.connector.postgresql.connection.wal2json.DateTimeFormat$ISODateTimeFormat.format(DateTimeFormat.java:154)
at io.debezium.connector.postgresql.connection.wal2json.DateTimeFormat$ISODateTimeFormat.date(DateTimeFormat.java:135)
at io.debezium.connector.postgresql.connection.AbstractColumnValue.asLocalDate(AbstractColumnValue.java:44)
at io.debezium.connector.postgresql.connection.pgoutput.PgOutputReplicationMessage.getValue(PgOutputReplicationMessage.java:97)
at io.debezium.connector.postgresql.connection.pgoutput.PgOutputMessageDecoder$1.getValue(PgOutputMessageDecoder.java:733)
at io.debezium.connector.postgresql.PostgresChangeRecordEmitter.columnValues(PostgresChangeRecordEmitter.java:184)
at io.debezium.connector.postgresql.PostgresChangeRecordEmitter.getOldColumnValues(PostgresChangeRecordEmitter.java:114)
at io.debezium.relational.RelationalChangeRecordEmitter.emitDeleteRecord(RelationalChangeRecordEmitter.java:133)
at io.debezium.relational.RelationalChangeRecordEmitter.emitChangeRecords(RelationalChangeRecordEmitter.java:56)
at io.debezium.connector.postgresql.PostgresChangeRecordEmitter.emitChangeRecords(PostgresChangeRecordEmitter.java:96)
at io.debezium.pipeline.EventDispatcher.dispatchDataChangeEvent(EventDispatcher.java:217)
... 17 more
Caused by: java.time.format.DateTimeParseException: Text 'infinity' could not be parsed at index 0
at java.base/java.time.format.DateTimeFormatter.parseResolved0(DateTimeFormatter.java:2046)
at java.base/java.time.format.DateTimeFormatter.parse(DateTimeFormatter.java:1948)
at java.base/java.time.LocalDate.parse(LocalDate.java:428)
at io.debezium.connector.postgresql.connection.wal2json.DateTimeFormat$ISODateTimeFormat.lambda$date$0(DateTimeFormat.java:135) 2022-04-19 20:52:51 INFO Connection gracefully closed
2022-04-19 20:52:51 INFO Connection gracefully closed
2022-04-19 20:52:51 INFO Finished streaming
...
2022-04-19 20:52:51 INFO [Producer clientId=connector-producer-XXXX-0] Closing the Kafka producer with timeoutMillis = 30000 ms.
2022-04-19 20:52:51 INFO Metrics scheduler closed
2022-04-19 20:52:51 INFO Closing reporter org.apache.kafka.common.metrics.JmxReporter
As can be seen from the stacktrace, the execution path is entirely within Debezium's codebase. We only collected 3 samples thus far. All of them involved `DELETE` operations. You can see references to `emitDeleteRecord(RelationalChangeRecordEmitter.java:133)` and `decodeDelete(PgOutputMessageDecoder.java:498)` in the stacktrace.
Do you see the same behaviour using the latest released Debezium version?
Yes. 1.9.0.Final is the latest at this time.
Do you have the connector logs, ideally from start till finish?
(You might be asked later to provide DEBUG/TRACE level log)
Yes. We can attempt to collect new samples.
How to reproduce the issue using our tutorial deployment?
<Your answer>
Additional context
Debezium is deployed in a streaming-only (no snapshot) mode. The exception was encountered during a pilot test run of monitoring an actual production database (for stress testing + to collect a lot of samples and scenarios). It hasn't been easy for us to replicate the conditions locally.
Almost every table has `REPLICA IDENTITY FULL`
Debezium metrics are attempted to be collected via Jolokia
Seems related to DBZ-2614, DBZ-2450, DBZ-2844
Seems related to this post in the wild.
What is also confusing is that the docs states:
PostgreSQL supports using +/-infinite values in TIMESTAMP columns. These special values are converted to timestamps with value 9223372036825200000 in case of positive infinity or -9223372036832400000 in case of negative infinity. This behaviour mimics the standard behaviour of PostgreSQL JDBC driver - see org.postgresql.PGStatement interface for reference.
We have not yet tried to re-collect samples with 1.8.1.Final. But given the handling of `infinity` values has been an issue in the past (like 1.5.0) we aren't optimistic that that its merely a new phenomenon that downgrading will help with.
Summary
We believe this is a significant problem. A Postgres-allowed value can effectively kill Debezium.
