[druid-user] Druid 0.20.2 failing jobs due to parsing exception

Hello All,

We have recently upgraded from druid version 0.19 to 0.20.2 after which we are seeing issues while ingestion of data through kafka.

If any of the event has issue with parsing, entire job is been failed by the system.
We have kept “reportParseExceptions” as false, but still we are seeing ingestion getting failed if any incoming record is having parsing issues.

Any suggestions are welcomed.

Regards,
Akhilesh

Hi Nishant,

here is the tuningConfig for the task in question

“tuningConfig”: {
“type”: “kafka”,
“appendableIndexSpec”: {
“type”: “onheap”
},
“maxRowsInMemory”: 50000,
“maxBytesInMemory”: 0,
“maxRowsPerSegment”: 5000000,
“maxTotalRows”: null,
“intermediatePersistPeriod”: “PT10M”,
“basePersistDirectory”: “/apps/eenpm/customized_druid/var/tmp/1578656204152-0”,
“maxPendingPersists”: 0,
“indexSpec”: {
“bitmap”: {
“type”: “concise”
},
“dimensionCompression”: “lz4”,
“metricCompression”: “lz4”,
“longEncoding”: “longs”,
“segmentLoader”: null
},
“indexSpecForIntermediatePersists”: {
“bitmap”: {
“type”: “concise”
},
“dimensionCompression”: “lz4”,
“metricCompression”: “lz4”,
“longEncoding”: “longs”,
“segmentLoader”: null
},
“buildV9Directly”: true,
“reportParseExceptions”: false,
“handoffConditionTimeout”: 0,
“resetOffsetAutomatically”: true,
“segmentWriteOutMediumFactory”: null,
“workerThreads”: null,
“chatThreads”: null,
“chatRetries”: 8,
“httpTimeout”: “PT10S”,
“shutdownTimeout”: “PT80S”,
“offsetFetchPeriod”: “PT30S”,
“intermediateHandoffPeriod”: “P2147483647D”,
“logParseExceptions”: true,
“maxParseExceptions”: 2147483647,
“maxSavedParseExceptions”: 0,
“skipSequenceNumberAvailabilityCheck”: false,
“repartitionTransitionDuration”: “PT120S”
}

Happy Learning !!