DB2
IBM DB2 is IBM's flagship relational database management system, renowned for enterprise-grade performance, scalability, and rock-solid reliability in mission-critical environments.
Supported Versions and Architectures
- Versions: DB2 9.7 through 11.x (extensively tested and validated on DB2 11.x)
- Platforms: Linux, UNIX, Windows, AIX, AS/400
- Architectures: All deployment configurations supported
Supported Data Types
| Category | Data Types |
|---|---|
| String and Text | VARCHAR, CHARACTER, CLOB, LONG VARCHAR |
| Numeric | INTEGER, DECIMAL, BIGINT, SMALLINT, DOUBLE, REAL |
| Date and Time | TIMESTAMP, DATE, TIME |
| Boolean | BOOLEAN |
| Binary Large Object | BLOB |
| XML | XML |
Quick Setup Guide
1. Create Database User
On Linux systems:
sudo useradd xpipes
sudo passwd xpipes
2. Grant Permissions
- Full sync only
- Full + incremental sync
GRANT USAGE ON SCHEMA your_schema TO USER xpipes;
GRANT DBADM TO xpipes;
tip
Incremental sync requires database admin privileges for ReadLog API access.
3. Configure Character Encoding (if needed)
For non-English environments:
# UTF-8 encoding
DB2set DB2codepage=1208
# GBK encoding
DB2set DB2codepage=1386
# Country code
DB2set DB2country=86
Important Considerations
- DDL operations: Run
CALL SYSPROC.ADMIN_CMD('REORG TABLE <schema>.<table>');after DDL changes during incremental sync - Large ReadLogs: Initial incremental sync may be slower with very large transaction logs
- Variable-length fields: May cause apparent operation type mismatches in logs (data accuracy unaffected)
- ReadLog API overhead: Periodic API calls consume database and network resources
Limitations
- Character encoding: Raw log service must match DB2 database encoding
- DDL timing: Requires table reorganization after schema changes
- Log parsing: Large transaction logs can impact initial sync performance