Re: pg_system_identifier()

bricklen <bricklen(at)gmail(dot)com> writes:
> On Thu, Aug 22, 2013 at 6:42 AM, Andres Freund <andres(at)2ndquadrant(dot)com>wrote:
>> FWIW I've wished for that function repeatedly. Mostly just to make sure
>> I am actually connected to the same "network" of replicas and not some
>> other.
>> It's also useful if you're providing support for a limited number of
>> machines and you want some form of identifying a node.

> There's a "hostname" function at PGXN which serves some use-cases:
> http://pgxn.org/dist/hostname/

I have a very vague recollection that we might've intentionally decided
not to expose the system identifier at the SQL level. This could be all
wet, but it'd be worth trolling the archives to see if there was such a
conversation and if so whether the arguments still have merit.

See also recent discussion about changing how the identifier is computed
--- it'd be a good idea to fix that before we expose the identifier to
users, if we decide to do so.

regards, tom lane