Horton

Re: pgstattuple "unexpected zero page" for gist and hash indexes

Dilip Kumar <dilipbalaut@gmail.com>

View thread
From: Dilip Kumar <dilipbalaut@gmail.com>
To: Nitin Motiani <nitinmotiani@google.com>
Cc: Michael Paquier <michael@paquier.xyz>, pgsql-hackers@postgresql.org
Date: 2025-10-01T15:23:01Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. pgstattuple: Improve reports generated for indexes (hash, gist, btree)

On Wed, Oct 1, 2025 at 8:32 PM Nitin Motiani <nitinmotiani@google.com> wrote:
>
> Apologies, I accidentally sent my previous reply only to Michael
> instead of hitting 'reply all'. Adding the contents of those messages
> in the quoted text.
>
> On Wed, Oct 1, 2025 at 4:45 PM Michael Paquier <michael@paquier.xyz> wrote:
> >
> > On Wed, Oct 01, 2025 at 04:17:49PM +0530, Nitin Motiani wrote:
> > > Thanks Michael. We can keep the simple change we have in v2 without
> > > reporting any corruption. But perhaps we should check for the opaque
> > > size mismatch for btree (as it's already done for gist and hash) to
> > > keep the code consistent for all three. We can avoid any reporting or
> > > further analysis but we can skip the other operations in the case of
> > > size mismatch. What are your thoughts on that?
> >
> > You mean an check on BTPageOpaqueData with a new else branch in
> > pgstat_btree_page()?  Yep, let's do that as well.
> > --
>
> Thanks Michael. I'm attaching v3 with this change.

I think this looks good to me now, lets see what Michael has to say on
this, after that we can mark ready for committer.

-- 
Regards,
Dilip Kumar
Google