Flink sink StarRocks報錯整理
阿新 • • 發佈:2021-09-13
問題一
檢視flink日誌報錯如下:
2021-09-12 09:22:04,652 ERROR com.dorisdb.connector.flink.manager.DorisStreamLoadVisitor [] - Stream Load response: {"Status":"Fail","BeginTxnTimeMs":33,"Message":"Failed to commit txn 4102465. Tablet [588483] success replica num 1 is less then quorum replica num 2 while error backends","NumberUnselectedRows":0,"CommitAndPublishTimeMs":0,"Label":"904fee77-a943-49e4-9a36-c70fa39c37f6","LoadBytes":9373,"StreamLoadPutTimeMs":9,"NumberTotalRows":37,"WriteDataTimeMs":104,"TxnId":4102465,"LoadTimeMs":152,"ReadDataTimeMs":0,"NumberLoadedRows":37,"NumberFilteredRows":0}
檢視be warn日誌報錯如下:
W0912 08:15:14.366163 27079 version_graph.cpp:431] fail to find path in version_graph. spec_version: 0-260870 W0912 08:15:14.370337 27079 tablet.cpp:562] status:-214, tablet:588515.1199425153.0b4cf96b8a0289ae-eb9d270dc129feae, missed version for version:[0-260870] W0912 08:15:14.370579 27079 tablet.cpp:1014] 588515.1199425153.0b4cf96b8a0289ae-eb9d270dc129feae has 1 missed version:[260869-260869], W0912 09:05:29.844586 27080 version_graph.cpp:431] fail to find path in version_graph. spec_version: 0-111272 W0912 09:05:29.850328 27080 tablet.cpp:562] status:-214, tablet:508224.512708038.2e46a60a452a1f84-a27f660a5ee865be, missed version for version:[0-111272] W0912 09:05:29.854856 27080 tablet.cpp:1014] 508224.512708038.2e46a60a452a1f84-a27f660a5ee865be has 2 missed version:[111270-111270],[111271-111271], W0912 09:11:39.266932 27080 version_graph.cpp:431] fail to find path in version_graph. spec_version: 0-268418 W0912 09:11:39.267238 27080 tablet.cpp:562] status:-214, tablet:588422.1709206849.5f4f1c24fd6669a2-8ed5989a3359f2a6, missed version for version:[0-268418] W0912 09:11:39.267259 27080 tablet.cpp:1014] 588422.1709206849.5f4f1c24fd6669a2-8ed5989a3359f2a6 has 1 missed version:[268417-268417], W0912 09:14:04.371534 27080 version_graph.cpp:431] fail to find path in version_graph. spec_version: 0-111466 W0912 09:14:04.371690 27080 tablet.cpp:562] status:-214, tablet:508230.512708038.9b4f28dcb864c56c-a796c40711b247ae, missed version for version:[0-111466] W0912 09:14:04.371888 27080 tablet.cpp:1014] 508230.512708038.9b4f28dcb864c56c-a796c40711b247ae has 1 missed version:[111465-111465], W0912 09:22:04.623602 27220 stream_load_executor.cpp:202] commit transaction failed, errmsg=Failed to commit txn 4102464. Tablet [588483] success replica num 1 is less then quorum replica num 2 while error backends id=204bb4f2cf78eddb-1b536f94abf32a9e, job_id=-1, txn_id=4102464, label=dafb789e-b9b2-4907-87a0-9373c40a1326 W0912 09:22:04.624311 27220 stream_load.cpp:113] Fail to handle streaming load, id=204bb4f2cf78eddb-1b536f94abf32a9e errmsg=Failed to commit txn 4102464. Tablet [588483] success replica num 1 is less then quorum replica num 2 while error backends W0912 09:22:05.704756 27225 stream_load_executor.cpp:202] commit transaction failed, errmsg=Failed to commit txn 4102471. Tablet [588483] success replica num 1 is less then quorum replica num 2 while error backends id=2948270eb826fcd3-da7c129b82e34dba, job_id=-1, txn_id=4102471, label=dafb789e-b9b2-4907-87a0-9373c40a1326 W0912 09:22:05.705308 27225 stream_load.cpp:113] Fail to handle streaming load, id=2948270eb826fcd3-da7c129b82e34dba errmsg=Failed to commit txn 4102471. Tablet [588483] success replica num 1 is less then quorum replica num 2 while error backends
與StarRocks支援人員溝通
所以更改了匯入間隔,從原來的5s更改到60s,暫時解決此問題,需後續持續觀察。
問題二
檢視flink日誌如下:
2021-09-06 14:12:13,737 ERROR com.dorisdb.connector.flink.manager.DorisStreamLoadVisitor [] - Stream Load response: {"Status":"Fail","BeginTxnTimeMs":3,"Message":"failed to call frontend service","NumberUnselectedRows":0,"CommitAndPublishTimeMs":0,"Label":"e07e6987-c78f-4ba8-8472-712512b68d5b","LoadBytes":0,"StreamLoadPutTimeMs":0,"NumberTotalRows":0,"WriteDataTimeMs":0,"TxnId":3338839,"LoadTimeMs":0,"ReadDataTimeMs":0,"NumberLoadedRows":0,"NumberFilteredRows":0}
網路問題,檢視節點之間的網路通訊,重新啟動任務即可。