Hi,
could it be that the database uses an encoding other than UTF-8 (e.g.
SQL_ASCII or LATIN1)? If so, another client may at some point have
written data that cannot be represented in UTF-8. Since your connection
uses UTF-8 as the client encoding, those raw bytes can't be converted,
which would explain why the error only shows up intermittently, i.e.
only when the affected rows are involved.
Am 06.10.2026 um 09:23 schrieb Igor Korot:
> Hi, ALL,
>
> When run my program and connect to the Postgres server with ODBC
> INTERMITTENTLY receive the error:
>
> ERROR: invalid byte sequence for encoding "UTF8" 0xed 0xbf 0xa0
> Does this sound familiar to anyone?
>
> Thank you.
>
> Trying to google I only got meaningful sugesston for when the last byte s 0
>
> I code n C++.
