@Vijay Kumar​ 

1. Check your schema of the table using DESCRIBE TABLE and the dataframe you loaded using df.printSchema() and compare which column has a mismatch. You need to see if your custom schema is matching the table schema.
2. MergeSchema will support only the schemas that it can typecast.
3. If you want to completely recreate the table schema with the ones of dataframes, you can use overwriteSchema option.

Please share your code so that we can look and help you.