Re: pg_basebackup, pg_receivexlog and data durability (was: silent data loss with ext4 / all current versions)

On Fri, Sep 2, 2016 at 2:20 AM, Peter Eisentraut
<peter(dot)eisentraut(at)2ndquadrant(dot)com> wrote:
> On 5/13/16 2:39 AM, Michael Paquier wrote:
>> So, attached are two patches that apply on HEAD to address the problem
>> of pg_basebackup that does not sync the data it writes. As
>> pg_basebackup cannot use directly initdb -S because, as a client-side
>> utility, it may be installed while initdb is not (see Fedora and
>> RHEL), I have refactored the code so as the routines in initdb.c doing
>> the fsync of PGDATA and other fsync stuff are in src/fe_utils/, and
>> this is 0001.
>
> Why fe_utils? initdb is not a front-end program.

Thinking about that, you are right. Let's move it to src/common,
frontend-only though.

>> Patch 0002 is a set of fixes for pg_basebackup:
>> - In plain mode, fsync_pgdata is used so as all the tablespaces are
>> fsync'd at once. This takes care as well of the case where pg_xlog is
>> a symlink.
>> - In tar mode (no stdout), each tar file is synced individually, and
>> the base directory is synced once at the end.
>> In both cases, failures are not considered fatal.
>
> Maybe there should be --nosync options like initdb has?

What do others think about that? I could implement that on top of 0002
with some extra options. But to be honest that looks to be just some
extra sugar for what is basically a bug fix... And I am feeling that
providing such a switch to users would be a way for one to shoot
himself badly, particularly for pg_receivexlog where a crash can cause
segments to go missing.
--
Michael