Re: Configuration include directory - Mailing list pgsql-hackers

From Greg Smith
Subject Re: Configuration include directory
Date
Msg-id 4EE7C9C2.4050501@2ndQuadrant.com
Whole thread Raw
In response to Re: Configuration include directory  (Noah Misch <noah@leadboat.com>)
List pgsql-hackers
On 12/13/2011 01:28 PM, Noah Misch wrote:
>> !<para>
>> !       Another possibility for this same sort of organization is to create a
>> !       configuration file directory and put this information into files there.
>> !       Other programs such as<productname>Apache</productname>  use a
>> !<filename>conf.d</>  directory for this purpose.  And using numbered names
>>      
> This specific use of conf.d is a distribution-driven pattern; the upstream
> Apache HTTP Server distribution never suggests it directly...
> ...
>
> Overall, I'd probably just remove these comparisons to other projects.
>    

I hadn't realized that distinction; will have to look into that some 
more.  Thanks again for the thorough review scrubbings, I can see I have 
another night of getting cozy with mmgr/README ahead.  I've gotten more 
than a fair share of feedback time for this CF, I'm going to close this 
patch for now, keep working on it for a bit more, and re-submit later.

My hope with this new section is that readers will realize the 
flexibility and options possible with the include and include_dir 
commands, and inspire PostgreSQL users to adopt familiar conventions 
from other programs if they'd like to.  I've made no secret of the fact 
that I don't like the way most people are led toward inefficiently 
managing their postgresql.conf files, that I feel the default 
configurations both encourages bad practices and makes configuration 
tool authoring a mess.  I would really like to suggest some possible 
alternatives here and get people to consider them, see if any gain 
adoption.  I thought that mentioning the examples are inspired by common 
setups of other programs, ones that people are likely to be familiar 
with, enhanced that message.  That's not unprecedented; 
doc/src/sgml/client-auth.sgml draws a similar comparison with Apache in 
regards to how parts of the pg_hba.conf are configured.  No argument 
here that I need to clean that section up still if I'm going to make 
this argument though.  I didn't expect to throw out 85 new lines of docs 
and get them perfect the first time.

-- 
Greg Smith   2ndQuadrant US    greg@2ndQuadrant.com   Baltimore, MD
PostgreSQL Training, Services, and 24x7 Support  www.2ndQuadrant.us



pgsql-hackers by date:

Previous
From: Alvaro Herrera
Date:
Subject: Re: foreign key locks, 2nd attempt
Next
From: Stephen Frost
Date:
Subject: Re: LibreOffice driver 2: MIT Kerberos vs Microsoft Kerberos