Re: pg_last_xact_replay_timestamp meaning

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: Bruce Momjian <bruce(at)momjian(dot)us>
Cc: PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: pg_last_xact_replay_timestamp meaning
Date: 2011-03-19 01:35:42
Message-ID: AANLkTikJTEZ4n5sN=941jrSSwwf16Z3_-c5JWuOO1sqO@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Fri, Mar 18, 2011 at 7:23 PM, Bruce Momjian <bruce(at)momjian(dot)us> wrote:
> I just applied a doc patch for pg_last_xact_replay_timestamp, and the
> text now says:
>
>       <entry>Get timestamp of last transaction replayed during recovery.
>        This is the time at which the commit or abort WAL record for that
>        transaction was generated on the primary.
>        If no transactions have been replayed during recovery, this function
>        returns NULL.  Otherwise, if recovery is still in progress this will
>        increase monotonically.  If recovery has completed then this value will
>        remain static at the value of the last transaction applied during that
>        recovery.  When the server has been started normally without recovery
>        the function returns NULL.
>
> Is this really the last commit/abort record or the last WAL record?
> What should it be?  Is the name of this function correct?  Do we care
> only about commit/abort records? Why?

Commit and abort records have a timestamp. Other WAL records don't.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrew Dunstan 2011-03-19 02:19:09 Re: pgsql: Document the all-balls IPv6 address.
Previous Message Robert Haas 2011-03-19 01:20:43 Re: pg_ctl restart - behaviour based on wrong instance