postmaster/postgres merge for testing

Peter Eisentraut <peter_e@gmx.net> writes:

(It's a bit weird in that src/backend already contains a postmaster
directory so you can't build a "postmaster" file there. So in the
build tree it's called postmaster_. Feel free to make better
suggestions.)

backend, maybe? Or just keep calling it postgres at that point.

regards, tom lane

Peter Eisentraut <peter_e@gmx.net> writes:

Attached is a patch that merges postmaster and postgres into just a
postmaster command.

I had some second thoughts about this, specifically about which
direction do we really want to go in. With this patch, it no longer
really matters what the executable file is named, right? We were both
implicitly assuming that the name should end up being "postmaster",
but I think there's a good case to be made that the right thing to do
is to migrate in the direction of having just one executable named
"postgres". We've seen complaints before that having a daemon named
"postmaster" confuses newbies into thinking it's got something to do
with mail. And it's already the case that the child processes all call
themselves "postgres", which will become even more confusing if there is
no longer any executable named "postgres".

If we went in this direction we'd have to keep the installed
postmaster->postgres symlink for awhile to avoid breaking existing
start scripts, but it could be deprecated and then removed in a few
releases.

Thoughts?

regards, tom lane

On Mon, 2006-01-23 at 11:13 -0500, Tom Lane wrote:

Peter Eisentraut <peter_e@gmx.net> writes:

Attached is a patch that merges postmaster and postgres into just a
postmaster command.

I had some second thoughts about this, specifically about which
direction do we really want to go in. With this patch, it no longer
really matters what the executable file is named, right? We were both
implicitly assuming that the name should end up being "postmaster",
but I think there's a good case to be made that the right thing to do
is to migrate in the direction of having just one executable named
"postgres". We've seen complaints before that having a daemon named
"postmaster" confuses newbies into thinking it's got something to do
with mail. And it's already the case that the child processes all call
themselves "postgres", which will become even more confusing if there is
no longer any executable named "postgres".

If we went in this direction we'd have to keep the installed
postmaster->postgres symlink for awhile to avoid breaking existing
start scripts, but it could be deprecated and then removed in a few
releases.

Thoughts?

This is clearly better, IMNSHO. I did wonder about postgresqld or
postgresd or some such - many server programs end in "d" or ".d" to
distinguish them from client programs. But probably just "postgres" is
best.

cheers

andrew

On Mon, Jan 23, 2006 at 11:51:36AM -0500, Andrew Dunstan wrote:

This is clearly better, IMNSHO. I did wonder about postgresqld or
postgresd or some such - many server programs end in "d" or ".d" to
distinguish them from client programs. But probably just "postgres" is
best.

Or postgresql if we want to be consistent...
</nitpick>
--
Jim C. Nasby, Sr. Engineering Consultant jnasby@pervasive.com
Pervasive Software http://pervasive.com work: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461

On 1/23/06, Tom Lane <tgl@sss.pgh.pa.us> wrote:

Peter Eisentraut <peter_e@gmx.net> writes:

Attached is a patch that merges postmaster and postgres into just a
postmaster command.

I had some second thoughts about this, specifically about which
direction do we really want to go in. With this patch, it no longer
really matters what the executable file is named, right? We were both
implicitly assuming that the name should end up being "postmaster",
but I think there's a good case to be made that the right thing to do
is to migrate in the direction of having just one executable named
"postgres". We've seen complaints before that having a daemon named
"postmaster" confuses newbies into thinking it's got something to do
with mail. And it's already the case that the child processes all call
themselves "postgres", which will become even more confusing if there is
no longer any executable named "postgres".

+1 for 'postgres'.

--
marko

pgd?

Gustavo.

Tom Lane wrote:

Peter Eisentraut <peter_e@gmx.net> writes:

Attached is a patch that merges postmaster and postgres into just a
postmaster command.

I had some second thoughts about this, specifically about which
direction do we really want to go in. With this patch, it no longer
really matters what the executable file is named, right? We were both
implicitly assuming that the name should end up being "postmaster",
but I think there's a good case to be made that the right thing to do
is to migrate in the direction of having just one executable named
"postgres". We've seen complaints before that having a daemon named
"postmaster" confuses newbies into thinking it's got something to do
with mail. And it's already the case that the child processes all call
themselves "postgres", which will become even more confusing if there is
no longer any executable named "postgres".

If we went in this direction we'd have to keep the installed
postmaster->postgres symlink for awhile to avoid breaking existing
start scripts, but it could be deprecated and then removed in a few
releases.

Thoughts?

+1 postgres (having the executable name matching the default os
superuser and database accounts seems logical).

On Tue, 2006-01-24 at 05:17, Gustavo Tonini wrote:

pgd?

or taking a page out of apache's book, databased ?

Robert Treat
--
Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL