Restore Enterprise Data Preparation
Restore the application service metatdata, repositories, and storage locations associated with Enterprise Data Preparation.
You do not need to complete the restore procedures in sequence. However, you should disable the Enterprise Data Preparation Service and the associated application services before performing a restore procedure.
Step 1. Restore the Model Repository Content
Restore the Model repository content from the repository backup file.
Verify that the Model repository is empty. If the Model repository contains content, the restore option is disabled. You must delete the content from the Model repository before restoring the repository content from the backup file.
1. In the Administrator tool, select the select the Services and Nodes view.
2. In the Domain Navigator, select the Model Repository Service associated with the Enterprise Data Preparation Service.
3. If the Model repository contains content, delete the content. From the Actions menu, select Repository Contents > Delete.
4. From the Actions menu, select Repository Contents > Restore.
The Restore Repository Contents dialog box appears.
5. Select the Model repository backup file to restore.
6. Enter the properties required to restore the Model repository.
The following table describes the properties:
Property | Description |
---|
Username | User name of any user in the domain. |
Password | Password of the domain user. |
Security Domain | Security domain to which the domain user belongs. Default is native. |
7. Click OK.
Step 2. Create the Hive and HDFS Resources in the Catalog
Create the Hive and HDFS scanner resources in the Enterprise Data Catalog used by the Enterprise Data Preparation Service.
For each scanner resource, enter the property values you copied during the backup phase into each tab in the Resource view in Catalog Administrator.
For more information about creating resources, see the Informatica Catalog Administrator Guide.
1. In the Administrator tool, select the Services and Nodes view.
2. In the Domain Navigator, select the Enterprise Data Preparation Service.
3. In the Properties view, expand Catalog Service Options.
4. Select the Catalog Service.
5. Click the Catalog Administrator URL.
The URL has the following format:
<Catalog Service host name>:<Catalog Service port>/ldmadmin
6. In Catalog Administrator, click Resource, and then select New > Resource.
The create resource view opens.
7. Create each Hive and HDFS scanner resource.
Enter the property values you copied on each tab in Catalog Administrator.
Step 3. Restore the Enterprise Data Preparation Connections
Import the connection objects for the HDFS, Hive, and Hadoop connections used by the Enterprise Data Preparation Service from the .xml file to which you exported the connection objects.
Use the infacmd isp ImportDomainObjects command to import the connection object for each connection from the .xml file.
For more information about using the infacmd isp ImportDomainObjects command, see the Informatica Command Reference.
Run the following command:
infacmd isp ImportDomainObjects -dn <domain name> -un <user name> -pd <password>
-sdn <security domain> -fp <import file>
The following table describes the infacmd isp ImportDomainObjects command options and arguments you might specify to import the connection objects for the HDFS, Hive, and Hadoop connections:
Option | Argument | Description |
---|
-DomainName -dn | domain_name | Required. Name of the Informatica domain. |
-UserName -un | user_name | Required if the domain uses Native or LDAP authentication. User name to connect to the domain. Optional if the domain uses Kerberos authentication. To run the command with single sign-on, do not set the user name. If you set the user name, the command runs without single sign-on. |
-Password -pd | password | Required if you specify the user name. |
-SecurityDomain -sdn | security_domain | Required if the domain uses LDAP authentication. Optional if the domain uses native authentication or Kerberos authentication. Name of the security domain to which the domain user belongs. If the domain uses native or LDAP authentication, the default is Native. If the domain uses Kerberos authentication, the default is the LDAP security domain created during installation. |
-ImportFilePath -fp | import_file_path | Required. Path and file name of the XML file from which you import the objects. |
The following example imports the connection objects for the HDFS, Hive, and Hadoop connections from the specified .xml file:
infacmd isp ImportDomainObjects -dn InfaDomain -un Administrator -pd password
-fp "/backup/edl_connections.xml"
Step 4. Restore the Data Preparation Repository
Restore the Data Preparation repository, and then use the Administrator tool to configure the Interactive Data Preparation Service to use the repository schema.
1. Restore the MySQL database or Oracle database used as the Data Preparation repository.
For more information, see the MySQL documentation or the Oracle documentation.
2. In the Administrator tool, select the Services and Nodes view.
3. In the Domain Navigator, select the Interactive Data Preparation Service.
4. In the Properties view, expand Data Preparation Repository Options.
5. Click the Edit icon.
6. Enter the properties required to connect to the Data Preparation repository.
The following table describes the repository properties:
Property | Description |
---|
Database Type | Type of database to use for the Data Preparation repository. |
Host Name | Host name of the machine that hosts the Data Preparation repository database. This property appears only if your Database Type is MySQL. |
Database Port Number | Port number for the database. This property appears only if the Database Type is MySQL. |
Connection String | Connection used to access the Oracle database. Use the following connection string: jdbc:informatica:oracle://<host name>:<port>;ServiceName=<database name> This field appears if you select Oracle as your Database Type. |
Secure JDBC Parameters | If the database is secured, information such as TrustStore and TrustStorePassword can be included in this field. It is saved in an encrypted format. Parameters usually configured include the following: EncryptionMethod=<encryption method>;HostNameInCertificate=<hostname>;TrustStore=<truststore file with its location>;TrustStorePassword=<truststore password>;KeyStore==<keystore file with its location>;KeyStorePassword=<keystore password>;ValidateServerCertificate=<true /false> This field appears if you select Oracle as your Database Type. |
Database User Name | Database user account to use to connect to the Data Preparation repository. |
Database User Password | Password for the Data Preparation repository database user account. |
Modify Database User Password | Select this checkbox to modify the database user password. |
Schema Name | Schema or database name of the Data Preparation repository database. This field appears if you select MySQL as your Database Type. |
Step 5. Restore the Interactive Data Preparation Service Durable Storage
Import the Interactive Data Preparation Service durable storage metadata into a durable storage location in the HDFS. After you import the durable storage metadata, use the Administrator tool to configure the Interactive Data Preparation Service to use the durable storage location.
You can use the Hadoop DistCp tool to import the storage metadata from the backup location to the HDFS location.
1. On HDFS, run the following command:
hadoop distcp <backup location> hdfs://<durable storage location>
The following table describes the DistCp command arguments:
Argument | Description |
---|
Durable Storage Location | The HDFS storage location to which to copy the durable storage metadata. |
Backup Location | The path to the backup directory from which to copy the durable storage metadata. |
2. In the Administrator tool, select the Services and Nodes view.
3. In the Domain Navigator, select the Interactive Data Preparation Service.
4. In the Properties view, expand Data Preparation Storage Options, and then click the Edit icon.
5. Enter the following properties:
Property | Description |
---|
Local Storage Location | Directory for data preparation file storage on the node where the Interactive Data Preparation Service runs. |
HDFS Connection | Connection to the durable storage location in HDFS. |
HDFS Storage Location | Durable storage location in HDFS. |
Hadoop Distribution Directory | Location of the Hadoop distribution on HDFS data nodes. |
Step 6. Recycle the Application Services
Start the application services used by Enterprise Data Preparation.
Start the application services in the following order:
- •Model Repository Service
- •Data Integration Service
- •Content Management Service
- •Interactive Data Preparation Service
- •Enterprise Data Preparation Service
1. In the Administrator tool, select the Services and Nodes view.
2. In the Domain Navigator of the Administrator tool, select an application service.
3. Click the Recycle the Service button.
4. Click OK.
Step 7. Restore the Rules Metadata
If rules are configured for the Interactive Data Preparation Service, restore the rules metadata to the Model repository used by the Interactive Data Preparation Service.
Use the infacmd tools ImportObjects command to import the rules metadata, the dictionary content, and the NER and Classifier model files into a project in the Model repository associated with the Interactive Data Preparation Service.
The Model Repository Service that manages the Model repository used by the Interactive Data Preparation Service must be running when you run the command.
For more information about using the infacmd tools ImportObjects command, see the Informatica Command Reference.
Run the following command:
infacmd tools ImportObjects -dn <domain name> -un <user name> -pd <password>
-sdn <security domain> -tn <target project name> -rs <Model Repository Service name>
-fp <import file> -sp <source project name> -tf <target folder name> -oo <options>
The following table describes the infacmd tools ImportObjects options and arguments:
Option | Argument | Description |
---|
-DomainName -dn | domain_name | Name of the Informatica domain. |
-UserName -un | user_name | User name to connect to the domain. Optional if the domain uses Kerberos authentication. |
-Password -pd | password | Password for the user name. |
-SecurityDomain -sdn | security_domain | Required if the domain uses LDAP authentication. Optional if the domain uses native authentication or Kerberos authentication. Name of the security domain to which the domain user belongs. If the domain uses native or LDAP authentication, the default is Native. If the domain uses Kerberos authentication, the default is the LDAP security domain created during installation. |
-TargetProject -tp | target_project | Name of the project in the Model repository into which you want to import the objects. The project must exist in the Model repository before you import the objects. The option is ignored if you use an import control file. |
-RepositoryService - rs | service_name | Model Repository Service name. |
-ImportFilePath -fp | file_path | Path and file name of the XML file from which to import the rules metadata. You can specify an absolute path or a relative path to the file name. |
-SourceProject -sp | project_name | Source project name in the file to import. |
-TargetFolder -tf | folder_name | Target folder within the project into which you want to import the rules objects. If you do not specify a target folder, the objects are imported into the target project. The folder must exist in the repository before you import the objects. |
-OtherOptions -oo | options | Additional options to import the dictionary content and the NER and Classifier model files from the specified zip file. Enter options using the following format: rtm:<option_name>=<value>,<option_name>=<value> Required option names include: - - codePage. Code page of the reference data.
- - refDataFile. Path and file name of the zip file from where you want to import the reference table data.
|
The following example imports rules metadata from a dq_rules_metadata.xml file into a folder named dq_rules within a project named NEW_PROJECT. The example also imports the dictionary content and the NER and Classifier models from a dq_rules_artifacts.zip file into the project.
infacmd tools importObjects -dn InfaDomain -un Administrator -pd password -sdn native
-rs MRS_dev3 -tp NEW_PROJECT -tf dq_rules -fp "/export/dq_rules_metadata.xml" -ow true
-oo "rtm:codePage=UTF-8,refDataFile="/output/dq_rules_artifacts.zip""