Re: [JDBC] Upgrading from PG 8.3.3 to 9.3.4 - FATAL: invalid value for parameter "TimeZone": "PST" - Mailing list pgsql-general

From Tom Lane
Subject Re: [JDBC] Upgrading from PG 8.3.3 to 9.3.4 - FATAL: invalid value for parameter "TimeZone": "PST"
Date
Msg-id 14863.1401639945@sss.pgh.pa.us
Whole thread Raw
In response to Upgrading from PG 8.3.3 to 9.3.4 - FATAL: invalid value for parameter "TimeZone": "PST"  (David Wall <d.wall@computer.org>)
List pgsql-general
David Wall <d.wall@computer.org> writes:
> The exception occurs when JDBC tries to connect to PG:

> 2014-05-31 22:14:34,351 ERROR (eSignForms) SQLException:
> ConnectionPool.makeConnection(esf) to URL:
> jdbc:postgresql://localhost.localdomain:25432/zingr:
> 2014-05-31 22:14:34,352 ERROR (eSignForms)   Message:   FATAL: invalid
> value for parameter "TimeZone": "PST"
> org.postgresql.util.PSQLException: FATAL: invalid value for parameter
> "TimeZone": "PST"

Hm.  libpq will try to set TimeZone at connection time if it sees a value
for the environment variable "PGTZ"; so, if you were using libpq, this
would be explainable by different environment settings in different cases.
There is probably some comparable behavior in the JDBC driver, but I don't
know exactly what.

One hole in this type of theory is that it doesn't explain a behavioral
difference between 8.3.x and 9.3.x; unless maybe the JDBC driver's
behavior changed in this regard since then.  libpq has done the PGTZ thing
for a very long time.

            regards, tom lane


pgsql-general by date:

Previous
From: Adrian Klaver
Date:
Subject: Re: Upgrading from PG 8.3.3 to 9.3.4 - FATAL: invalid value for parameter "TimeZone": "PST"
Next
From: David Wall
Date:
Subject: Re: Upgrading from PG 8.3.3 to 9.3.4 - FATAL: invalid value for parameter "TimeZone": "PST"