Solution 1:

Documentation says

AWS DMS converts BLOBs, CLOBs, and NCLOBs to a VARCHAR on the target Amazon Redshift instance. Amazon Redshift doesn't support VARCHAR data types larger than 64 KB, so you can't store traditional LOBs on Amazon Redshift.

Furthermore, you can store up to 65535 bytes into a "CHAR" datatype column:

Name Storage Range
VARCHAR, CHARACTER VARYING, or NVARCHAR 4 bytes + total bytes for characters, where each character can be 1 to 4 bytes 65535 bytes (64K -1)

Therefore, it looks as if you'll have to split that CLOB into 3 columns, e.g.

TEXT_1 VARCHAR(65000),
TEXT_2 VARCHAR(65000),
TEXT_3 VARCHAR(65000)

and put substrings into them.