一下把数据文件全清空了，还好是测试环境

本地文件导入hive。。。。
json
{
"job": {
"content": [
{
"reader": {
"name": "txtfilereader",
"parameter": {
"encoding": "UTF-8",
"fieldDelimiter": "\t",
"path": ["D:/dataxPath"]
}
},
"writer": {
"name": "hdfswriter",
"parameter": {
"column": [{
"name":"a",
"type":"String"
},{
"name":"b",
"type":"String"
}],
"defaultFS": "hdfs://xxxx:xxxx",
"fieldDelimiter": "\t",
"fileName": "xxxx",
"fileType": "text",
"path": "/apps/hive/warehouse/orc_table",
"writeMode": "append"
}
}
}
],
"setting": {
"speed": {
"channel": "2"
}
}
}
}

最后hdfs上数据文件都没了
日志输出：

Review the provided txtfilereader and hdfswriter configuration and attached log first. Reproduce the import against a test HDFS/Hive path, investigate why files under /apps/hive/warehouse/orc_table were removed despite writeMode append, and define a regression test or reproduction that demonstrates the corrected behavior.

Written by the indexing model from the issue text.