Issue

The TBR-11048 error occurs in the Tibero to Oracle DB Link. This document provides guidance on how to address this issue. 

Note

Oracle Information

• charset : WE8ISO8859P1

 

Tibero Information

• charset : MSWIN949
• TB_NLS_LANG environment variable value: unspecified (default MSWIN949)

Cause

In the existing Oracle environment, a 1-byte character set such as ASCII / WE8DEC was used with a workaround to store Korean characters. Therefore, when converting to a proper character set (such as mswin949), Korean characters do not display correctly. To support this environment, Tibero added the SKIP_CHAR_CONV option to the gateway.

SKIP_CHAR_CONV

SKIP_CHAR_CONV is a parameter that can only be used between Oracle and Tibero. It determines whether to skip the conversion process caused by the gateway character set setting.

• When set to SKIP_CHAR_CONV=N, the gateway first converts the data fetched from Oracle into the character set specified by TB_NLS_LANG. Then Tibero reconverts the data into its own character set, resulting in two conversion processes. During this process, the byte size of the data may increase, which can exceed the column length and cause the TBR-11048 error.

[Result of Oracle - Tibero DB LINK INSERT (SKIP_CHAR_CONV=N)]

Solutions

When SKIP_CHAR_CONV=Y, the gateway passes Oracle data directly to Tibero without intermediate conversion. Tibero then converts the received data into its own character set only once, eliminating unnecessary conversions and preventing the TBR-11048 error.

[Result of Oracle - Tibero DB LINK INSERT (SKIP_CHAR_CONV=Y)]