All Products
Search
Document Center

Data Transmission Service:Collect database information

Last Updated:Mar 28, 2026

Advanced Database & Application Migration (ADAM) collects structural data from your Oracle database to generate feasibility and compatibility reports before migration. Two collection methods are available: online collection and collector-based collection.

Choose a collection method

MethodWhen to use
Online collectionThe source database is accessible from the cloud. ADAM collects data directly after you add ADAM server IP addresses to the database whitelist.
Collector-based collectionThe source database cannot be accessed over the Internet or from the cloud. You download and run the ADAM Database Collector on a separate device in the same network as the database.

Online collection supports two connectivity options:

  • With a public IP address: The source database is accessible over the Internet.

  • Without a public IP address: Connect through Database Gateway (DG). DG enables remote access to databases in a private network, including databases in your on-premises data center or on other cloud platforms without a public IP address. For more information, see What is Database Gateway?.

ADAM currently supports DG instances in the China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hangzhou), and China (Shanghai) regions.

Online collection

Prerequisites

Before you begin, make sure you have:

Set up the collection account

Log in to the source Oracle database and run the following SQL statements to create a dedicated collection account.

Oracle 10g, 11g, or 12c (non-CDB mode — local user)

create user eoa_user identified by eoaPASSW0RD default tablespace users;
grant connect,select_catalog_role to eoa_user;

Oracle 12c (CDB mode — common user)

Connect to the CDB, then run:

create user c##eoa_user identified by "eoaPASSW0RD" default tablespace users;
grant connect,select_catalog_role to c##eoa_user container=all;
alter user c##eoa_user set container_data=all container=current;

Create and run a collection task

  1. Log on to the Data Management (DMS) 5.0 console.

  2. In the top menu bar, choose Data + AI > Heterogeneous Database Migration (ADAM) > Database Evaluation.

  3. On the Database Information Collection tab, click Online Database Collection.

  4. Click Create Collection Task.

  5. Configure the collection account as described in Set up the collection account, then click Next.

  6. In the Create Collection Task panel, configure the following parameters.

    ParameterDescription
    Collection Task NameEnter a name for the collection task.
    Source Database TypeOnly ORACLE is supported.
    Source Database ProfileSelect the profile for the source database.
    List Of Schemas To IncludeSpecify the schemas to collect. Separate multiple names with commas (,).
    List Of Schemas To ExcludeSpecify the schemas to exclude. Separate multiple names with commas (,).
    Collect SQL StatementsSelect whether to include SQL statements in the collection.
  7. Click Test Connectivity. If the connection succeeds, click Start Collection.

  8. After the task completes, select the task and click Next: Create Profile to generate a database profile.

Collector-based collection

Prerequisites

Before you begin, make sure you have:

  • A separate device that meets the minimum requirements listed in Collector device requirements

  • SYSDBA-level access to the source Oracle database to create a collection account

Collector device requirements

Run the ADAM Database Collector on a separate device — not on the same machine as the source database. The device must meet the following requirements:

RequirementMinimum
NetworkMust be able to connect to the source database
CPU2 cores
Memory8 GB
Disk100 GB free space

Download the collector

  1. Log on to the Data Management (DMS) 5.0 console.

  2. In the top menu bar, choose Data + AI > Heterogeneous Database Migration (ADAM) > Database Evaluation.

  3. On the Database Information Collection tab, click Download Collector.

  4. Download the ADAM client version that matches the operating system of your collector device, then decompress the package.

Set up the collection account

If you already have an account with the required permissions, skip this step.

Log in to the source Oracle database using an account with SYSDBA permissions and run the following SQL statements.

Oracle 10g, 11g, or 12c (non-CDB mode — local user)

create user eoa_user identified by eoaPASSW0RD default tablespace users;
grant connect,select_catalog_role to eoa_user;

Oracle 12c (CDB mode — common user)

Connect to the CDB, then run:

create user c##eoa_user identified by "eoaPASSW0RD" default tablespace users;
grant connect,select_catalog_role to c##eoa_user container=all;
alter user c##eoa_user set container_data=all container=current;

Run the collector

  1. On the collector device, run the collection command for your Oracle version.

    Use the .bat script on Windows and the .sh script on Linux.
    • Oracle 10g

      collect_10g[.sh|.bat] -h -u -p -d <service_name>
    • Oracle 11g R1

      collect_11gR1[.sh|.bat] -h -u -p -d <service_name>
    • Oracle 11g R2

      collect_11gR2[.sh|.bat] -h -u -p -d <service_name>
    • Oracle 12c, 18c, or 19c

      collect_12c[.sh|.bat] -h <host> -u <username> -p <password> -P <port> -d <service_name> -s <sid>

      To collect from a specific pluggable database (PDB) in Oracle 12c, use the collect_12c script. The operation is similar to collecting from Oracle 11g.

    ParameterDescriptionExample
    -hHostname or IP address of the source database192.168.1.100
    -uUsername for the collection accounteoa_user
    -pPassword for the collection accounteoaPASSW0RD
    -PPort of the source database1521
    -dService name of the source database. For Oracle 12c, use the service name of the target PDB.orcl
    -sSID (instance name) of the source databaseorcl
  2. After the collection completes, the system generates a data packet and prints the output path. The output looks similar to:

    [***] ******************************************************************
    [***] * Collect Successfully!
    [***] *
    [***] * Complete the file packaging, the package result path is:
    [***] * ~rainmeter/out/data.zip
    [***] ******************************************************************

Clean up the collection account (optional)

After migration, drop the temporary collection account to remove unnecessary access.

Connect to the database using an account with SYSDBA permissions and run the appropriate statement:

  • Oracle 10g, 11g, 12c, 18c, or 19c (non-CDB mode)

    drop user eoa_user cascade;
  • Oracle 12c, 18c, or 19c (CDB mode)

    drop user c##eoa_user cascade;

Performance considerations

  • Run the collector during off-peak hours for best performance.

  • The collector has minimal CPU impact. If the collection includes getddl operations on database objects, CPU load may increase slightly at the start. The degree of increase depends on the source database specifications and current load.

  • Memory usage from the collector on the source database is negligible.