Re: [HACKERS] VACUUM and ANALYZE disagreeing on what reltuples means

Alvaro Herrera <alvherre@alvh.no-ip.org>

From: Alvaro Herrera <alvherre@alvh.no-ip.org>

To: Tom Lane <tgl@sss.pgh.pa.us>

Cc: Tomas Vondra <tomas.vondra@2ndquadrant.com>, Haribabu Kommi <kommi.haribabu@gmail.com>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>

Date: 2017-11-20T16:46:18Z

Lists: pgsql-hackers

BTW see bug #14863 which is related to this:
https://postgr.es/m/CAEBTBzu5j_E1K1jb9OKwTZj98MPeM7V81-vadp5adRm=NhJnwA@mail.gmail.com

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Commits

Sync up our various ways of estimating pg_class.reltuples.
- 7c91a0364fcf 11.0 landed
Fix tuple counting in SP-GiST index build.
- 649f1792508f 11.0 landed
- eee190da7eeb 9.5.13 landed
- db35bf507ffd 9.6.9 landed
- bf14575c840f 10.4 landed
- 7f6f8ccd976d 9.4.18 landed
- 46f80803a127 9.3.23 landed
Fix errors in contrib/bloom index build.
- c35b47286960 11.0 landed
- df90401556ce 9.6.9 landed
- 76e2b5ae4151 10.4 landed
When updating reltuples after ANALYZE, just extrapolate from our sample.
- d44ce7b1afdb 9.3.23 landed
- d04900de7d0c 11.0 landed
- c9414e7867f7 9.5.13 landed
- c2c4bc628bbe 9.6.9 landed
- 25a2ba35edbc 9.4.18 landed
- 1bfb5672306d 10.4 landed
Avoid holding AutovacuumScheduleLock while rechecking table statistics.
- 5328b6135756 9.3.23 landed
- 95f08d32dec4 9.4.18 landed
- 231329a17564 9.5.13 landed
- 4b0e717053e3 9.6.9 landed
- 4460964aedaa 10.4 landed
- 38f7831d703b 11.0 landed