Horton

Re: pg_dump, gzwrite, and errno

Alvaro Herrera <alvherre@2ndquadrant.com>

View thread View on postgr.es
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Justin Pryzby <pryzby@telsasoft.com>, pgsql-hackers@postgresql.org, Kunshchikov Vladimir <Vladimir.Kunshchikov@infotecs.ru>
Date: 2020-06-19T21:30:47Z
Lists: pgsql-hackers 
On 2020-Jun-18, Tom Lane wrote:

> Surely it's insufficient as-is, because there is no reason to suppose
> that errno is zero at entry.  You'd need to set errno = 0 first.

Oh, right.

> Also it's fairly customary in our sources to include a comment about
> this machination; so the full ritual is usually more like

Yeah, I had that in my local copy.  Done like that in all the most
obvious places.  But there are more places that are still wrong: I
believe every single place that calls WRITE_ERROR_EXIT is doing the
wrong thing.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Commits

  1. Ensure write failure reports no-disk-space

  2. Fix build on zlib-less environments

  3. Fix pg_dump's errno checking for zlib I/O