Parameter · Client Connection Defaults / Locale and Formatting
timezone_abbreviations
Selects a file of time zone abbreviations.
- Default:
'' (empty) - Type: string
- Applies: Session (any user)
In plain words
Abbreviations like EST or IST are ambiguous worldwide: IST is Israel, India or Ireland depending on where you are. This setting chooses the list PostgreSQL uses to translate them when reading date/time input.
For DBAs and operators
The value names a file in share/timezonesets/: Default, Australia (EST = Australian Eastern, +10) or India (IST = +05:30); you can add your own. Only input is affected; output abbreviations come from the TimeZone database. View the active list in pg_timezone_abbrevs.
In PostgreSQL 18 an abbreviation used by the session's current TimeZone is recognized first, before this list: with TimeZone = 'Asia/Kolkata', '10:00 IST' means +05:30 even under Default (which maps IST to Israel). Behaviour can therefore differ between sessions with different TimeZone values.
Best practice: have applications send numeric offsets (+05:30) or full zone names (Asia/Kolkata) instead of abbreviations. An unknown abbreviation in timestamp input fails with ERROR: invalid input syntax for type timestamp with time zone: "2025-01-01 10:00 XYZ".
How to change it
Any user can change it for their session with SET.