MQRC 2119: Message not converted

What it means

MQGMO_CONVERT asks the queue manager to convert message data into the CodedCharSetId and Encoding that the application sets in the MQMD before the get. The message carries its own CCSID, encoding and format from when it was put. 2119 means the conversion between the two failed. The call ends with MQCC_WARNING, and the data is returned as it was stored. The MQMD fields are set to describe the data as returned.

The usual reasons are a CCSID the queue manager can't convert between, wrong values in the MQMD when the message was put, or data that isn't valid in the character set it claims to use. For a message made of several parts, such as one with a dead-letter header, some parts may be converted while others aren't. The CCSID and encoding fields in each part still describe that part correctly.

Some more specific codes exist for parts of the same problem. 2111 and 2115 report a bad source or target CCSID, and 2120 says the converted message is too big for the buffer. 2119 is the general case when conversion didn't happen.

Common causes

• The sender put the message with the wrong CodedCharSetId, so the data doesn't match the character set the queue manager thinks it is in.
• The requested conversion isn't supported on this platform, for example an unusual EBCDIC or double-byte CCSID with no conversion table installed.
• The receiving program sets an invalid or unintended CodedCharSetId or Encoding in the MQMD before the get.
• Binary data is marked as MQFMT_STRING, so it holds byte sequences that can't be converted.
• Only part of a multi-part message could be converted.

How to find the problem

1. Browse the message and note its Format, CodedCharSetId and Encoding. On distributed MQ use amqsbcg or dmpmqmsg.
2. Log the CodedCharSetId and Encoding the receiving program sets before MQGET. A common bug is reusing an MQMD from an earlier get, so the values are whatever the last message had.
3. Check the error log of the queue manager, or of the client if the client did the conversion, for messages about conversion around the time of the get.
4. Look at a hex dump of the data to see whether it really matches the CCSID it claims. UTF-8 text marked as an EBCDIC code page is a typical example.

How to fix it

• Fix the sender so the MQMD describes the data correctly. Set CodedCharSetId to the real code page and Format to MQFMT_STRING only for text.
• Reset CodedCharSetId and Encoding in the receiver's MQMD before every get, to the values the program wants.
• If the platform can't convert between the two CCSIDs, convert in the application, or have the sender use a common CCSID such as UTF-8 (1208).
• Handle 2119 in the receiver by checking the returned CCSID and processing or rejecting the message, instead of treating the bytes as local text.

Technical reviewer: Albert Joseph, WND Software.

Last reviewed 2026-10-18 by the WND Software team. Written in our own words. For the full reference, check IBM’s MQ documentation for your version.