Re: Need to fix one more glitch in upgrade to -10.2 - Mailing list pgsql-general

From Rich Shepard
Subject Re: Need to fix one more glitch in upgrade to -10.2
Date
Msg-id alpine.LNX.2.20.1802180757430.10894@salmo.appl-ecosys.com
Whole thread Raw
In response to Re: Need to fix one more glitch in upgrade to -10.2  (Rich Shepard <rshepard@appl-ecosys.com>)
Responses Re: Need to fix one more glitch in upgrade to -10.2  (Adrian Klaver <adrian.klaver@aklaver.com>)
List pgsql-general
On Sat, 17 Feb 2018, Rich Shepard wrote:

> That's what I was thinking, too. I can remove the 10.2 package, rebuild
> and re-install it. Run initdb, then, as postgres, read in the .sql file.
> This is probably the pragmatic thing to do.

   Rather than doing this my reading of the 10.2 initdb pages suggest that I
can remove /var/lib/pgsql/10.2/data/*, leaving an empty data directory owned
by postgres.postgres. Then, as user postgres, I can run initdb with -D
/full/path/to/data/ and start over. Is this appropriate?

   Looking at the backed up .sql file I see roles for databases that were
removed a long time ago. I'll read the roles section of the 10.2 docs and
edit the .sql file to clean those up.

   Since I'm the only user on this host and postgres cluster would 'trust' be
the appropriate method for use in pb_hba.conf; e.g., 
# "local" is for Unix domain socket connections only
local   all             all                                     trust
# IPv4 local connections:
host    all             all             127.0.0.1/32            trust

so I can access my databases from the desktop or portables without a
password?

Regards,

Rich



pgsql-general by date:

Previous
From: Tomas Vondra
Date:
Subject: Re: query performance
Next
From: Adrian Klaver
Date:
Subject: Re: Need to fix one more glitch in upgrade to -10.2