Horton

Re: [PATCH] Unremovable tuple monitoring

Robert Haas <robertmhaas@gmail.com>

View thread
From: Robert Haas <robertmhaas@gmail.com>
To: Yeb Havinga <yebhavinga@gmail.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Alvaro Herrera <alvherre@commandprompt.com>, Royce Ausburn <royce.ml@inomial.com>, Pg Hackers <pgsql-hackers@postgresql.org>, Kevin Grittner <kevin.grittner@wicourts.gov>
Date: 2011-11-16T13:03:42Z
Lists: pgsql-hackers 
On Wed, Nov 16, 2011 at 3:31 AM, Yeb Havinga <yebhavinga@gmail.com> wrote:
> Apparently pg_stat* counts the recently_dead tuple under n_live_tup, else 2
> is a wrong number, where pgstattuple counts recently_dead under
> dead_tuple_count. This could be a source of confusion. If there is any
> serious work considered here, IMHO at least the numbers of the two different
> sources of tuple counters should match in terminology and actual values.

+1.

> Maybe also if pgstattuple were to include the distinction unremovable dead
> tuples vs dead tuples, a log line by vacuum encountering unremovable dead
> tuples could refer to pgstattuple for statistics.

Not sure about the log line, but allowing pgstattuple to distinguish
between recently-dead and quite-thoroughly-dead seems useful.

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