Message-ID: <5d51a9cf-8f2e-a854-424f-001f0a260a8b@enterprisedb.com>
Date: Fri, 19 Nov 2021 11:10:46 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:91.0)
 Gecko/20100101 Thunderbird/91.3.0
Subject: Re: Deficient error handling in pg_dump and pg_basebackup
Content-Language: en-US
To: Tom Lane <tgl@sss.pgh.pa.us>, pgsql-hackers@lists.postgresql.org
References: <1343113.1636489231@sss.pgh.pa.us>
From: Peter Eisentraut <peter.eisentraut@enterprisedb.com>
In-Reply-To: <1343113.1636489231@sss.pgh.pa.us>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: 
 <https://www.postgresql.org/message-id/5d51a9cf-8f2e-a854-424f-001f0a260a8b%40enterprisedb.com>
Precedence: bulk

On 09.11.21 21:20, Tom Lane wrote:
> Why is this different from the half-dozen
> other fsync-error cases in the same file?  Why, if fsync failure
> here is so catastrophic, is it okay to just return a normal failure
> code when tar_close doesn't even get to this point because of some
> earlier error?  At the very least it seems like it'd be preferable
> to do the exit(1) at the caller level.
> 
> The addition of the exit(1) seems to have been intentional in
> 1e2fddfa3, so cc'ing Peter for comment.

That commit addressed the behavior of fsync() failure in pg_receivewal 
and pg_recvlogical, which are long-running daemon processes, so this 
change was analogous to the server-side changes at the time.  I don't 
know what the behavior of fsync() failure in pg_basebackup should be, so 
calls that are only reachable from pg_basebackup are currently being 
handled differently.