Re: increase of xact_commit vs txid_current

From: Julien Rouhaud <rjuju123(at)gmail(dot)com>
To: reg_pg_stefanz(at)perfexpert(dot)ch
Cc: "pgsql-generallists(dot)postgresql(dot)org" <pgsql-general(at)lists(dot)postgresql(dot)org>
Subject: Re: increase of xact_commit vs txid_current
Date: 2020-05-07 18:22:00
Message-ID: CAOBaU_ZWR_fp5Xr8K8HhAi654byOtBekpqaZ+mKhx8n2S=bXWA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Thu, May 7, 2020 at 7:01 PM <reg_pg_stefanz(at)perfexpert(dot)ch> wrote:
>
> Hi,
>
> I am confused, the documentation says for pg_stat_database
> xact_commit Number of transactions in this database that have
> been committed
> and somewhere else
> txid_current() get current transaction ID, assigning a new one
> if the current transaction does not have one
>
> I would have naively expected txid_current() be more or less in line
> with xact_commit, or increasing faster as txid_current() should be
> global but xact_commit is per database.
>
> However xact_commit seems to increases faster than txid_current(), what
> am I missing? Are there commits that do not increase the xid number?

Indeed, read-only transactions usually don't consume a transaction id,
to avoid the need to perform a FREEZE too often. Note that pcalling
txid_current() will consume such a transaction id, even in a read only
query.

In response to

Browse pgsql-general by date

  From Date Subject
Next Message github kran 2020-05-07 18:23:04 AutoVacuum and growing transaction XID's
Previous Message Amarendra Konda 2020-05-07 18:07:42 Re: Explain plan changes - IN CLAUSE ( Passing direct values Vs INNER Query )