Re: [WIP] pg_ping utility
Peter Eisentraut <peter_e@gmx.net>
From: Peter Eisentraut <peter_e@gmx.net>
To: Bruce Momjian <bruce@momjian.us>
Cc: Michael Paquier <michael.paquier@gmail.com>, Phil Sorber <phil@omniti.com>, PostgreSQL-development Hackers <pgsql-hackers@postgresql.org>
Date: 2012-11-27T04:32:15Z
Lists: pgsql-hackers
On Mon, 2012-11-26 at 13:14 -0500, Bruce Momjian wrote: > I would normally agree with this analysis, but pg_ctl -w certainly > need this ping functionality, so it kind of makes sense that others > might need it too. Sure, PQping is useful for this very specific use case of seeing whether the server has finished starting up. If someone came with a plausible use case for a startup script that couldn't use pg_ctl but wanted ping functionality available in a shell script, then pg_ping could be provided. But that would also determine what options to provide. For example, we might not need repeated ping in that case. But I think people see PQping and will see pg_ping as a monitoring facility, and I think that's a mistake.