Nullbytes in JSON-Data (in the field extra) can cause troubles:

psycopg2.errors.UntranslatableCharacter: unsupported Unicode escape sequence
DETAIL:  \\u0000 cannot be converted to text.
CONTEXT:  JSON data, line 1: ....0\"}, \"malware\": \"Conficker\", \"payload\": {\"text\":...
PL/pgSQL function directives_from_extra(bigint,json) line 5 during statement block local variable initialization
SQL statement \"SELECT directives_from_extra(NEW.id, NEW.extra)\"
PL/pgSQL function events_insert_directives_for_row() line 3 at PERFORM"

A problematic extra-value can look like:

'{"payload": {"text": "M41\u0012)3U>\bxӾ6\u0000\u0013M6M6M4M4]4y]4ӭ4"}}'

Source: Microsoft CTIP

While the json type in postgres allows (encoded) Nullbytes in principle, some systems have a problem with that. I have not found out yet, what causes some Postgres-Systems to throw an error here, while others work.
Tested with PostgreSQL 14.4. Works fine on e.g. openSUSE Tumbleweed, but not on Ubuntu 20.04 with Postgres-APT-Repo. But the latter is a target plattform of IntelMQ, so IntelMQ should be compatible with it.

The only solution I see is to escape the Nullbytes in the SQL Output for Postgres (SQLite does not have an issue with these characters).
Working on it.