[
https://issues.apache.org/jira/browse/IMPALA-12854?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Quanlong Huang updated IMPALA-12854:
------------------------------------
Component/s: Backend
Epic Link: IMPALA-12877
> Exceptions in CatalogOp might be logged without the stacktrace
> --------------------------------------------------------------
>
> Key: IMPALA-12854
> URL: https://issues.apache.org/jira/browse/IMPALA-12854
> Project: IMPALA
> Issue Type: Bug
> Components: Backend
> Reporter: Quanlong Huang
> Assignee: Quanlong Huang
> Priority: Critical
>
> We've seen a NullPointerException thrown from a REFRESH operation but the
> stacktrace is missing in the logs:
> {noformat}
> I0227 04:41:52.127296 313 JniUtil.java:166]
> f74a5d26635f02a8:85f249ab00000000] resetMetadata request: REFRESH TABLE
> db1.tbl1 PARTITIONS issued by user1
> I0227 04:41:52.127852 313 CatalogServiceCatalog.java:3219]
> f74a5d26635f02a8:85f249ab00000000] Refreshing partition metadata: db1.tbl1
> ingest_date=2024-02-27 (REFRESH TABLE db1.tbl1 PARTITIONS issued by user1)
> I0227 04:41:52.143682 313 HdfsTable.java:2943]
> f74a5d26635f02a8:85f249ab00000000] Setting the latest refresh event id to
> 41237662 for the reloaded 1 partitions
> I0227 04:41:52.143796 313 HdfsTable.java:2946]
> f74a5d26635f02a8:85f249ab00000000] for table db1.tbl1, file metadataOps:
> FORCE_LOAD, refreshing file metadata for 1 out of 1 partitions to reload in
> reloadPartitions()
> I0227 04:41:52.161365 313 HdfsTable.java:821]
> f74a5d26635f02a8:85f249ab00000000] Loaded file and block metadata for
> db1.tbl1 partitions: ingest_date=2024-02-27. Time taken: 17.338ms
> I0227 04:41:52.161485 313 CatalogServiceCatalog.java:3252]
> f74a5d26635f02a8:85f249ab00000000] Refreshed partition metadata: db1.tbl1
> ingest_date=2024-02-27
> W0227 04:41:52.259997 313 InFlightEvents.java:118]
> f74a5d26635f02a8:85f249ab00000000] Number of DDL events to be stored is at
> its max capacity 100. Ignoring add request for version 123034.
> W0227 04:41:52.260052 313 Table.java:1024]
> f74a5d26635f02a8:85f249ab00000000] Could not add 123034 version to the table
> db1.tbl1. This could cause unnecessary refresh of the table when the event is
> received by the Events processor.
> I0227 04:41:52.260093 313 CatalogServiceCatalog.java:1184]
> f74a5d26635f02a8:85f249ab00000000] Added catalog version 123034 in table's
> db1.tbl1 in-flight events
> E0227 04:41:52.260159 313 JniUtil.java:183]
> f74a5d26635f02a8:85f249ab00000000] Error in REFRESH TABLE db1.tbl1 PARTITIONS
> issued by user1. Time spent: 134ms
> I0227 04:41:52.260222 313 jni-util.cc:288]
> f74a5d26635f02a8:85f249ab00000000] java.lang.NullPointerException
> I0227 04:41:52.260236 313 status.cc:129] f74a5d26635f02a8:85f249ab00000000]
> NullPointerException: null
> @ 0x1026d84
> @ 0x1b10f74
> @ 0xfe0c7b
> @ 0xf7bf95
> @ 0xf3b94d
> @ 0xf2de93
> @ 0xf35c5b
> @ 0x143f1ba
> @ 0x1431c0a
> @ 0x1bf3c99
> @ 0x1bf4f31
> @ 0x24b9607
> @ 0x7fcd7316c1ca
> @ 0x7fcd6fef4e73
> E0227 04:41:52.260288 313 catalog-server.cc:236]
> f74a5d26635f02a8:85f249ab00000000] NullPointerException: null{noformat}
> We should add the exception in the error logs in JniCatalogOp.execOp(). CC
> [~prozsa]
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
