![](/img/trans.png)
[英]multiple sessions are created when inserting data into SQL table using ADF Pipeline through Copy Activity
[英]Copy From OnPrem SQL server to DocumentDB using custom activity in ADF Pipeline
我正在尝试从本地 sql 服务器中的 SQL 表复制数据,并使用 Azure 数据工厂管道中的自定义活动将其上传到文档数据库。 谁能告诉我如何使用 IDotNetActivity 或任何其他接口或类来完成它。
实际上,自定义活动目前无法访问本地数据。
类似的问题在这里: On-Prem SQL connection throwing SqlException in Datafactory custom activity
解决方案是将本地数据复制到云端。 然后针对云存储运行自定义活动。 wBob 分享了上面的好样本。
如果必须在一项活动中完成,则可以设置 vNet 和 ExpressRoute 以将 Azure 公共云与本地环境连接起来。
我让它与传统的 Azure 数据工厂 (ADF) 任务一起工作。 不需要自定义任务。 我不会让事情变得比他们需要的更复杂,特别是对于这些难以调试的组件。
以下示例显示:
本地 SQL Server 类型的链接服务:
{
"name": "OnPremLinkedService",
"properties": {
"type": "OnPremisesSqlServer",
"description": "",
"typeProperties": {
"connectionString": "Data Source=<servername - required for credential encryption>;Initial Catalog=<databasename - required for credential encryption>;Integrated Security=False;User ID=<username>;Password=<password>;",
"gatewayName": "<Name of the gateway that the Data Factory service should use to connect to the on-premises SQL Server database - required for credential encryption>",
"userName": "<Specify user name if you are using Windows Authentication>",
"password": "<Specify password for the user account>"
}
}
}
DocumentDB 类型的链接服务:
{
"name": "DocumentDbLinkedService",
"properties": {
"type": "DocumentDb",
"typeProperties": {
"connectionString": "AccountEndpoint=<EndpointUrl>;AccountKey=<AccessKey>;Database=<Database>"
}
}
}
SqlServerTable 类型的输入数据集:
{
"name": "SQLServerDataset",
"properties": {
"structure": [
{
"name": "Id",
"type": "Int32"
},
{
"name": "FirstName",
"type": "String"
},
{
"name": "MiddleName",
"type": "String"
},
{
"name": "LastName",
"type": "String"
}
],
"published": false,
"type": "SqlServerTable",
"linkedServiceName": "OnPremLinkedService",
"typeProperties": {
"tableName": "dbo.Users"
},
"availability": {
"frequency": "Day",
"interval": 1
},
"external": true,
"policy": {}
}
}
DocumentDbCollection 类型的输出数据集:
{
"name": "PersonDocumentDbTableOut",
"properties": {
"structure": [
{
"name": "Id",
"type": "Int32"
},
{
"name": "Name.First",
"type": "String"
},
{
"name": "Name.Middle",
"type": "String"
},
{
"name": "Name.Last",
"type": "String"
}
],
"published": false,
"type": "DocumentDbCollection",
"linkedServiceName": "DocumentDbLinkedService",
"typeProperties": {
"collectionName": "Person"
},
"availability": {
"frequency": "Day",
"interval": 1
}
}
}
使用 SqlSource 和 DocumentDbCollectionSink 复制活动的管道:
{
"name": "PipelineTemplate 3",
"properties": {
"description": "On prem to DocDb test",
"activities": [
{
"type": "Copy",
"typeProperties": {
"source": {
"type": "SqlSource"
},
"sink": {
"type": "DocumentDbCollectionSink",
"writeBatchSize": 2,
"writeBatchTimeout": "00:00:00"
},
"translator": {
"type": "TabularTranslator",
"columnMappings": "id: id, FirstName: Name.First, MiddleName: Name.Middle, LastName: Name.Last"
}
},
"inputs": [
{
"name": "SQLServerDataset"
}
],
"outputs": [
{
"name": "PersonDocumentDbTableOut"
}
],
"policy": {
"timeout": "1.00:00:00",
"concurrency": 1,
"retry": 3
},
"scheduler": {
"frequency": "Day",
"interval": 1
},
"name": "CopyActivityTemplate"
}
],
"start": "2016-10-05T00:00:00Z",
"end": "2016-10-05T00:00:00Z",
"isPaused": false,
"hubName": "adfdocdb2_hub",
"pipelineMode": "Scheduled"
}
}
我能够解决这个问题。 解决方案是在自定义活动中编写代码,使用以下代码将数据从本地 SQL Server 复制到 DocumentDB:
public async Task CopyDataFromTo(string source)
{
try
{
DataTable dtSource = new DataTable();
string EndpointUrl = "https://yourendpoint.documents.azure.com:443/";
string AuthorizationKey = "*****";
SecureString authKey = new SecureString();
foreach(char c in AuthorizationKey.ToCharArray())
{
authKey.AppendChar(c);
}
SqlDataAdapter adapSource = new SqlDataAdapter("Select * From YourTable", source);
adapSource.Fill(dtSource);
foreach (DataRow Dr in dtSource.Rows)
{
dynamic docFirst = new
{
UserID = Int32.Parse(Dr["ColumnOne"].ToString()),
UserAlias = Dr["ColumnTwo"].ToString()
};
using (var client = new DocumentClient(new Uri(EndpointUrl), authKey))
{
Document newDocument = await client.CreateDocumentAsync(UriFactory.CreateDocumentCollectionUri("DatabaseName", "CollectionName"), docFirst);
};
}
}
catch (Exception Ex)
{
throw Ex;
}
}
谢谢查尔斯。 事实证明你是对的。 我实施的解决方案是:
第 1 部分:
实施了数据工厂管道以将数据从本地数据库移动到暂存 DocumentDB 集合。
第 2 部分:
使用自定义活动将来自 documentdb 中不同集合(暂存)的数据组合起来,以创建具有所需输出数据的新 documentdb 集合。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.