Re: XMIN semantic at peril ? - Mailing list pgsql-general

From Tom Lane
Subject Re: XMIN semantic at peril ?
Date
Msg-id 2464.1192124890@sss.pgh.pa.us
Whole thread Raw
In response to Re: XMIN semantic at peril ?  (Erik Jones <erik@myemma.com>)
Responses Re: XMIN semantic at peril ?  (Karsten Hilbert <Karsten.Hilbert@gmx.net>)
List pgsql-general
Erik Jones <erik@myemma.com> writes:
> If the query is always based on a primary key + XMIN, and since
> vacuum is the only thing that sets FrozenTransactionId, would it be
> unsane to change the update to

> - update row with "... where pk=... and XMIN IN (old_xmin_from_read,
> FrozenTransactionId)

I wouldn't risk it, because that *would* fail in the case Karsten
mentioned of an update sequence taking so long that a concurrent
update got frozen.

            regards, tom lane

pgsql-general by date:

Previous
From: "Merlin Moncure"
Date:
Subject: Re: preferred way to use PG_GETARG_BYTEA_P in SPI
Next
From: Tom Lane
Date:
Subject: Re: preferred way to use PG_GETARG_BYTEA_P in SPI