Re: Setting pd_lower in GIN metapage

Masahiko Sawada <sawada.mshk@gmail.com>

From: Masahiko Sawada <sawada.mshk@gmail.com>
To: Michael Paquier <michael.paquier@gmail.com>
Cc: Amit Langote <Langote_Amit_f8@lab.ntt.co.jp>, Amit Kapila <amit.kapila16@gmail.com>, Tom Lane <tgl@sss.pgh.pa.us>, Pg Hackers <pgsql-hackers@postgresql.org>
Date: 2017-06-23T07:58:00Z
Lists: pgsql-hackers
On Fri, Jun 23, 2017 at 3:31 PM, Michael Paquier
<michael.paquier@gmail.com> wrote:
> On Fri, Jun 23, 2017 at 3:17 PM, Amit Langote
> <Langote_Amit_f8@lab.ntt.co.jp> wrote:
>> Initially, I had naively set wal_consistency_check = all before running
>> make installcheck and then had to wait for a long time to confirm that WAL
>> generated by the gin test indeed caused consistency check failure on the
>> standby with the v1 patch.
>
> wal_consistency_check = gin would have saved you a lot of I/O.
>
>> But I can see Sawada-san's point that there should be some way for
>> developers writing code that better had gone through WAL consistency
>> checking facility to do it without much hassle.  But then again, it may
>> not be that frequent to need that.

Yeah, it should be optional. I imagined providing such an option of
pg_regress or TAP test for the developers.

Regards,

--
Masahiko Sawada
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
NTT Open Source Software Center


Commits

  1. Flag index metapages as standard-format in xlog.c calls.

  2. Set the metapage's pd_lower correctly in brin, gin, and spgist indexes.