adf upsert

Hi all, 

I'm trying to create sample Upsert data pipeline in ADF (insert missing records and update changed ones). I created sample source and target Postgres table with all data types available, the number of source and target columns slightly differ.

CREATE TABLE "schema1".source_for_upsert_0001 (
    col_serial SERIAL PRIMARY KEY,
    col_smallint SMALLINT,
    col_integer INTEGER,
    col_bigint BIGINT,
    col_decimal DECIMAL(10, 2),
    col_numeric NUMERIC(10, 2),
    col_real REAL,
    col_double DOUBLE PRECISION,
    col_smallserial SMALLSERIAL,
    col_serial_alias SERIAL,
    col_bigserial BIGSERIAL,
    col_money MONEY,
    col_char CHAR(5),
    col_varchar VARCHAR(50),
    col_text TEXT,
    col_bytea BYTEA,
    col_timestamp TIMESTAMP,
    col_timestamptz TIMESTAMPTZ,
    col_date DATE,
    col_time TIME,
    col_timetz TIMETZ,
    col_boolean BOOLEAN,
    col_uuid UUID,
    col_json JSON,
    col_jsonb JSONB,
    col_xml XML,
    col_inet INET,
    col_cidr CIDR,
    col_macaddr MACADDR,
    col_bit BIT(8),
    col_varbit VARBIT(16),
    col_interval INTERVAL,
    col_point POINT,
    col_line LINE,
    col_lseg LSEG,
    col_box BOX,
    col_path PATH,
    col_polygon POLYGON,
    col_circle CIRCLE,
    col_tsquery TSQUERY,
    col_tsvector TSVECTOR
);

INSERT INTO "schema1".source_for_upsert_0001 (
    col_smallint, col_integer, col_bigint, col_decimal, col_numeric, col_real, 
    col_double, col_smallserial, col_serial_alias, col_bigserial, col_money, 
    col_char, col_varchar, col_text, col_bytea, col_timestamp, col_timestamptz, 
    col_date, col_time, col_timetz, col_boolean, col_uuid, col_json, col_jsonb, 
    col_xml, col_inet, col_cidr, col_macaddr, col_bit, col_varbit, col_interval, 
    col_point, col_line, col_lseg, col_box, col_path, col_polygon, col_circle, 
    col_tsquery, col_tsvector
) VALUES (
    1, 100, 1000, 1234.56, 1234.56, 12.34, 
    12345.6789, 1, DEFAULT, DEFAULT, '$1234.56', 
    'A', 'Sample Text', 'This is a text field.', E'\x48656c6c6f', 
    '2024-12-13 12:00:00', '2024-12-13 12:00:00+00', '2024-12-13', 
    '12:00:00', '12:00:00+00', TRUE, 
    '550e8400-e29b-41d4-a716-446655440000', '{"key": "value"}', '{"key": "value"}', 
    '<note><to>User</to><message>Hello!</message></note>', '192.168.1.1', '192.168.0.0/24', 
    '08:00:2b:01:02:03', B'10101010', B'1010101010101010', '1 year 2 months', 
    '(1,1)', '((0,0),(1,1))', '((0,0),(1,1))', '((0,0),(2,2))', 
    '((0,0),(1,1),(2,2),(2,0),(0,0))', '((0,0),(1,1),(2,2),(2,0),(0,0))', '<(1,1),1>', 
    'cat & dog', 'cat:3A dog:2A'
);

CREATE TABLE "schema1".target_for_upsert_0001 (
    col_2_serial SERIAL PRIMARY KEY,
    col_2_smallint SMALLINT,
    col_2_integer INTEGER,
    col_2a_integer INTEGER,
    col_2_bigint BIGINT,
    col_2_decimal DECIMAL(10, 2),
    col_2_numeric NUMERIC(10, 2),
    col_2_real REAL,
    col_2_double DOUBLE PRECISION,
    col_2_smallserial SMALLSERIAL,
    col_2_serial_alias SERIAL,
    col_2_bigserial BIGSERIAL,
    col_2_money MONEY,
    col_2_char CHAR(5),
    col_2_varchar VARCHAR(50),
    col_2_text TEXT,
    col_2_bytea BYTEA,
    col_2_timestamp TIMESTAMP,
    col_2_timestamptz TIMESTAMPTZ,
    col_2_date DATE,
    col_2_time TIME,
    col_2_timetz TIMETZ,
    col_2_boolean BOOLEAN,
    col_2_uuid UUID,
    col_2_json JSON,
    col_2_jsonb JSONB,
    col_2_xml XML,
    col_2_inet INET,
    col_2_cidr CIDR,
    col_2_macaddr MACADDR,
    col_2_bit BIT(8),
    col_2_varbit VARBIT(16),
    col_2_interval INTERVAL,
    col_2_point POINT,
    col_2_line LINE,
    col_2_lseg LSEG,
    col_2_box BOX,
    col_2_path PATH,
    col_2_polygon POLYGON,
    col_2_circle CIRCLE,
    col_2_tsquery TSQUERY,
    col_2_tsvector TSVECTOR
);

I used "data flow".

Upsert source:

Upsert derived column - as we don't have "updated at" timestamp column in source / target, I plan to use md5 of all present row values to compare changes

Upsert Alter Row - Upsert if: isNull(md5_columns)==false()

Upsert Sink:

Debugging: 

Could someone kindly look into it to advise what could be wrong? The last screenshot seems intended to clarify, but it doesn't really help to see the root cause.