Resolution:
This Error message was introduced in SmartSockets 6.8 . Currently it is specific to messages a client receives with more than 256k of data in any header extension field. An example would be a client which constructs a message with named fields, where the total size of all the name field strings is greater than 256k. This error message may encompass other situations in the future as well.

NOTE:  This FAQ refers to SmartSockets 6.8.0 CFIX 3 and higher only

When RTserver receives a message that it believes is corrupted it reports
         "TAL-SS-00234-E Corrupt message received from conn </conn_name>".
The question is what would be causing the corrupted message and/or what in the message is corrupted? Unfortunately as of SmartSockets 6.8 CFIX_V3 RTserver doesn't provide that detail when the error is generated.

Prior to 6.8 CFIX_V2 a corrupted message could cause an RTprocess to either spin or core. What was changed was that if an RTprocess (in this case RTserver) reports a corrupted message it throws the message on the floor and disconnects from the RTprocess it received the message from.

{SVC.SERVER./_foo@tcp:0.0.0.0:5102:157} Failure in reading from SPC 'tcp:0.0.0.0:PORT:PID' {message corrupt}

The reason the connection is dropped is because the connection is no longer trusted. Whichever RTserver initiated the connection will then attempt to reconnect to the other server.

This message can also mean the RTserver has received a message with more than 256k of data in any header extension field. An example would be a client which constructs a message with named fields, where the total size of all the name strings is greater than 256k.

As of SmartSockets 6.8.0 CFIX_V4 when an RTserver reports a corrupted message detail will be provided on what in the message was corrupted (header/payload) and possibly some header information (sender) if available.

What does it mean if I see the message: TAL-SS-00234-E Corrupt message received from conn /ss_unique_subject?