Re: Dirty buffers with suppress_redundant_updates_trigger - Mailing list pgsql-general

From Mike Noordermeer
Subject Re: Dirty buffers with suppress_redundant_updates_trigger
Date
Msg-id CAF0ozqsqXmKc6wd3zChPW_n8y7LnacRcckWYfOM4nxjdbE=6bA@mail.gmail.com
Whole thread Raw
In response to Re: Dirty buffers with suppress_redundant_updates_trigger  (Mike Noordermeer <mike@normi.net>)
Responses Re: Dirty buffers with suppress_redundant_updates_trigger  (Mike Noordermeer <mike@normi.net>)
List pgsql-general
On Mon, 14 Sep 2020 at 06:32, Mike Noordermeer <mike@normi.net> wrote:
> So it seems
> suppress_redundant_updates_trigger() does not entirely avoid writing
> _something_ to the blocks, and I don't know what it is and how to
> avoid it.

Looking at the pg_waldump output, it seems to write something
concerning locking to the WAL, I am seeing lots of the following
records:

rmgr: Heap        len (rec/tot):     54/    54, tx:       6747, lsn:
B/BC7785F0, prev B/BC7785B8, desc: LOCK off 36: xid 6747: flags 0x00
LOCK_ONLY EXCL_LOCK , blkref #0: rel 1663/13408/45636 blk 8332

Still trying to figure out what this means...

Kind regards,

Mike



pgsql-general by date:

Previous
From: Mike Noordermeer
Date:
Subject: Re: Dirty buffers with suppress_redundant_updates_trigger
Next
From: nandha kumar
Date:
Subject: Reg:CHARSET_COVERSION_LATIN_TO_UTF8