Hi Clemens,

We fixed this issue with the following configuration:
state.storage.fs.path-length-limit: 200

Have you tried this property?

-- Alex

On Tue, Mar 31, 2026 at 10:05 PM Clemens S <[email protected]>
wrote:

> I submit Flink SQL jobs on native Kubernetes operator in Application Mode
> (Flink 2.2.0, Kubernetes operator 1.4.0)
>
> I regularly run into these issues (Stack trace below) and I cannot figure
> out how to solve them apart from using short and cryptic app names.
> As you can see I already try to keep the base dir as short as possible.
> I already tried not setting the state.backend.rocksdb.log.dir &
> state.backend.rocksdb.localdir at all (and letting the system choose),
> choosing something invalid (e.g. /dev/null) like the configuration
> documentation suggested,
> reducing rocksdb log level (unfortunately unable to completely turn it
> off) and investigated. I found FLINK-31743
> <https://issues.apache.org/jira/browse/FLINK-31743> that seems to address
> the issue but since I am on a later version, only does so incompletely.
>
> Is there a way of reliably fixing this rather than trying to come up with
> shorter and shorter app names?
>
>
> [...]
> java.io.IOException: Error while opening RocksDB instance.
> at org.apache.flink.state.rocksdb.RocksDBOperationUtils.openDB(
> RocksDBOperationUtils.java:101)
> at org.apache.flink.state.rocksdb.restore.RestoredDBInstance
> .restoreTempDBInstanceFromLocalState(RestoredDBInstance.java:121)
> at org.apache.flink.state.rocksdb.restore.
> RocksDBIncrementalRestoreOperation.copyToBaseDBUsingTempDBs(
> RocksDBIncrementalRestoreOperation.java:788)
> at org.apache.flink.state.rocksdb.restore.
> RocksDBIncrementalRestoreOperation
> .mergeStateHandlesWithCopyFromTemporaryInstance(
> RocksDBIncrementalRestoreOperation.java:628)
> at org.apache.flink.state.rocksdb.restore.
> RocksDBIncrementalRestoreOperation.restoreFromMultipleStateHandles(
> RocksDBIncrementalRestoreOperation.java:446)
> at org.apache.flink.state.rocksdb.restore.
> RocksDBIncrementalRestoreOperation.restoreFromLocalState(
> RocksDBIncrementalRestoreOperation.java:326)
> at org.apache.flink.state.rocksdb.restore.
> RocksDBIncrementalRestoreOperation.lambda$restore$1(
> RocksDBIncrementalRestoreOperation.java:253)
> at org.apache.flink.state.rocksdb.restore.
> RocksDBIncrementalRestoreOperation.runAndReportDuration(
> RocksDBIncrementalRestoreOperation.java:893)
> at org.apache.flink.state.rocksdb.restore.
> RocksDBIncrementalRestoreOperation.restore(
> RocksDBIncrementalRestoreOperation.java:252)
> at org.apache.flink.state.rocksdb.RocksDBKeyedStateBackendBuilder.build(
> RocksDBKeyedStateBackendBuilder.java:390)
> ... 19 more
> Caused by: org.rocksdb.RocksDBException: While open a file for appending:
> /tmp/rdb/tmp_tm_hydra-sql-adr-assoc-device-and-login-features-taskmanager-
> 1-10
> _tmp_job_41471278f6601d1a7ab05da6958d83f7_op_KeyedProcessOperator_d4d5e8c74c3d05d8a9a53a9c312a6161__1_5__uuid_aadf2786-a3dd-
> 4fa9-acaa-59d560e05ce3_b5ea62d0-713f-46c4-bd4e-a4526f117f33_LOG: File
> name too long
> at org.rocksdb.RocksDB.open(Native Method)
> at org.rocksdb.RocksDB.open(RocksDB.java:315)
> at org.apache.flink.state.rocksdb.RocksDBOperationUtils.openDB(
> RocksDBOperationUtils.java:89)
>

Reply via email to