Re: Bug with pg_ctl -w/wait and config-only directories - Mailing list pgsql-hackers

From Peter Eisentraut
Subject Re: Bug with pg_ctl -w/wait and config-only directories
Date
Msg-id 1317714227.11446.10.camel@fsopti579.F-Secure.com
Whole thread Raw
In response to Re: Bug with pg_ctl -w/wait and config-only directories  (Bruce Momjian <bruce@momjian.us>)
Responses Re: Bug with pg_ctl -w/wait and config-only directories
List pgsql-hackers
On mån, 2011-10-03 at 18:44 -0400, Bruce Momjian wrote:
> Agreed.  You could argue that pg_ctl 9.1 is much better than anything
> anyone would be able to craft in a script.

And what facts support that argument?

Anyway, this comes back to your favorite argument upthread.  pg_ctl has
had occasional problems in the past.  So people have created
alternatives that are customized for their particular use case.  And
some of those init scripts and the like support versions back from 8.1
or 8.2 up to last week.  So it will take 2, 3, or 5 years until they'd
even consider abandoning their frameworks for, say, a pg_ctl-based
solution.  And even then there will individual debates on whether it
would be worth doing.

So, the bottom line is, until now there has been no widespread "forced"
use of pg_ctl, outside of Windows and pg_upgrade.  PostgreSQL 9.0 was
not packaged in any released version of Debian or Ubuntu, so I guess not
a lot of people tried pg_upgrade in those configurations.  Then again,
9.1 isn't packaged in a released OS version either, of course, so I have
no idea why this is coming up exactly now.



pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: Bug with pg_ctl -w/wait and config-only directories
Next
From: Dimitri Fontaine
Date:
Subject: Re: Should we get rid of custom_variable_classes altogether?