[jira] [Created] (FLINK-19016) Checksum mismatch when restore from RocksDB

[Shang Yuanchun (Jira)]

Jiayi Liao created FLINK-19016:
----------------------------------

             Summary: Checksum mismatch when restore from RocksDB
                 Key: FLINK-19016
                 URL: https://issues.apache.org/jira/browse/FLINK-19016
             Project: Flink
          Issue Type: Bug
            Reporter: Jiayi Liao


The error stack trace below:

{code:java}
Caused by: org.apache.flink.util.FlinkException: Could not restore keyed state backend for KeyedMapBundleOperator_44cfc1ca74b40bb44eed1f38f72b3ea9_(71/300) from any of the 1 provided restore options.
        at org.apache.flink.streaming.api.operators.BackendRestorerProcedure.createAndRestore(BackendRestorerProcedure.java:135)
        at org.apache.flink.streaming.api.operators.StreamTaskStateInitializerImpl.keyedStatedBackend(StreamTaskStateInitializerImpl.java:307)
        at org.apache.flink.streaming.api.operators.StreamTaskStateInitializerImpl.streamOperatorStateContext(StreamTaskStateInitializerImpl.java:135)
        ... 6 more
Caused by: org.apache.flink.runtime.state.BackendBuildingException: Caught unexpected exception.
        at org.apache.flink.contrib.streaming.state.RocksDBKeyedStateBackendBuilder.build(RocksDBKeyedStateBackendBuilder.java:333)
        at org.apache.flink.contrib.streaming.state.RocksDBStateBackend.createKeyedStateBackend(RocksDBStateBackend.java:580)
        at org.apache.flink.streaming.api.operators.StreamTaskStateInitializerImpl.lambda$keyedStatedBackend$1(StreamTaskStateInitializerImpl.java:291)
        at org.apache.flink.streaming.api.operators.BackendRestorerProcedure.attemptCreateAndRestore(BackendRestorerProcedure.java:142)
        at org.apache.flink.streaming.api.operators.BackendRestorerProcedure.createAndRestore(BackendRestorerProcedure.java:121)
        ... 8 more
Caused by: java.io.IOException: Error while opening RocksDB instance.
        at org.apache.flink.contrib.streaming.state.RocksDBOperationUtils.openDB(RocksDBOperationUtils.java:74)
        at org.apache.flink.contrib.streaming.state.restore.AbstractRocksDBRestoreOperation.openDB(AbstractRocksDBRestoreOperation.java:131)
        at org.apache.flink.contrib.streaming.state.restore.RocksDBIncrementalRestoreOperation.restoreFromLocalState(RocksDBIncrementalRestoreOperation.java:214)
        at org.apache.flink.contrib.streaming.state.restore.RocksDBIncrementalRestoreOperation.restoreFromRemoteState(RocksDBIncrementalRestoreOperation.java:188)
        at org.apache.flink.contrib.streaming.state.restore.RocksDBIncrementalRestoreOperation.restoreWithoutRescaling(RocksDBIncrementalRestoreOperation.java:162)
        at org.apache.flink.contrib.streaming.state.restore.RocksDBIncrementalRestoreOperation.restore(RocksDBIncrementalRestoreOperation.java:148)
        at org.apache.flink.contrib.streaming.state.RocksDBKeyedStateBackendBuilder.build(RocksDBKeyedStateBackendBuilder.java:277)
        ... 12 more
Caused by: org.rocksdb.RocksDBException: checksum mismatch
        at org.rocksdb.RocksDB.open(Native Method)
        at org.rocksdb.RocksDB.open(RocksDB.java:286)
        at org.apache.flink.contrib.streaming.state.RocksDBOperationUtils.openDB(RocksDBOperationUtils.java:66)
        ... 18 more
{code}

The machine goes down because of hardware problem, then the job cannot restart successfully anymore. After digging a little bit, I found that RocksDB in Flink uses sync instead of fsync to synchronized the data with the disk. With sync operation, the RocksDB cannot guarantee that the current in-progress file can be persisted on disk in takeDBNativeCheckpoint.








--
This message was sent by Atlassian Jira
(v8.3.4#803005)