Re: stat() vs ERROR_DELETE_PENDING, round N + 1

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Thomas Munro <thomas(dot)munro(at)gmail(dot)com>
Cc: pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>, Juan José Santamaría Flecha <juanjo(dot)santamaria(at)gmail(dot)com>, Michael Paquier <michael(at)paquier(dot)xyz>
Subject: Re: stat() vs ERROR_DELETE_PENDING, round N + 1
Date: 2021-09-02 10:31:51
Message-ID: 2241078.1630578711@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Thomas Munro <thomas(dot)munro(at)gmail(dot)com> writes:
> A disruptive solution that works in my tests: we could reuse the
> global barrier proposed in CF #2962. If you see EACCES, ask every
> backend to close all vfds at their next CFI() and wait for them all to
> finish, and then retry. If you get EACCES again it really means
> EACCES, but you'll very probably get ENOENT.

That seems quite horrid :-(. But if it works, doesn't that mean that
somewhere we are opening a problematic file without the correct
sharing flags?

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Eisentraut 2021-09-02 10:36:51 Re: public schema default ACL
Previous Message Thomas Munro 2021-09-02 10:28:09 Re: stat() vs ERROR_DELETE_PENDING, round N + 1