Horton

Re: Problem while setting the fpw with SIGHUP

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>
Cc: amit.kapila16@gmail.com, robertmhaas@gmail.com, hlinnaka@iki.fi, dilipbalaut@gmail.com, pgsql-hackers@postgresql.org
Date: 2018-09-18T06:39:37Z
Lists: pgsql-hackers 
On Tue, Sep 18, 2018 at 01:06:09PM +0900, Kyotaro HORIGUCHI wrote:
> I was wrong here. It was handled in HandleStartupProcInterrupts
> called from StartupXLOG. So, it should be just removed from the
> set. Sorry for the bogus patch.

Thanks for confirming.

Still, it looks like a waste to abuse on SIGINT just to forcibly wake up
the checkpointer and request from it a checkpoint...  And you could just
have used a new parameter for the checkpointer appended with
CHECKPOINT_FORCE.  I think that my approach of just making the set of
events purely ordered will save from any kind of race conditions, while
I suspect that what you propose here does not close all the holes.
--
Michael

Commits

  1. Fix assertion failure when updating full_page_writes for checkpointer.

  2. Attach FPI to the first record after full_page_writes is turned on.

  3. Revamp the WAL record format.

  4. Move the backup-block logic from XLogInsert to a new file, xloginsert.c.