cubefs
[BUG]: Full table scan scenario check condition
大表扫描场景里，判断SparkPlanInfo节点是否为扫描节点的逻辑为：node.getNodeName().startsWith("Scan")
但是实际的扫描节点的名称并不是Scan开头，逻辑是否有误？
"children": [{
"children": [],
"metadata": {},
"metrics": [{
"accumulatorId": 21,
"metricType": "sum",
"nodeName": "number of output rows"
}],
"nodeName": "HiveTableScan",
"simpleString": "HiveTableScan [id#6L, plan_type#10], HiveTableRelation dev.data_skew_table_partitioned, org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe, [id#6L, task_pod_name#7, project_id#8L, create_time#9, plan_type#10, deleted#11], [ds#12]"
}],
大表扫描场景里，判断SparkPlanInfo节点是否为扫描节点的逻辑为：node.getNodeName().startsWith("Scan")
但是实际的扫描节点的名称并不是Scan开头，逻辑是否有误？ "children": [{ "children": [], "metadata": {}, "metrics": [{ "accumulatorId": 21, "metricType": "sum", "nodeName": "number of output rows" }], "nodeName": "HiveTableScan", "simpleString": "HiveTableScan [id#6L, plan_type#10], HiveTableRelation
dev.data_skew_table_partitioned, org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe, [id#6L, task_pod_name#7, project_id#8L, create_time#9, plan_type#10, deleted#11], [ds#12]" }],
可以提供完整的一点eventlog？或者dag graph截图？
您好，有结论吗
另外，抱歉，昨天加错解析json包，比较晚看到结果。这个是通过个spark submit提交的任务吗？看解析plan里面的table可能需要小调整