On Sun, Dec 30, 2018 at 11:20 AM Peter Eisentraut <peter.eisentraut@xxxxxxxxxxxxxxx> wrote:
On 29/12/2018 20:04, Chuck Martin wrote:
> I thought I knew how to do this, but I apparently don't. I have to set
> up a new server as a standby for a PG 11.1 server. The main server has a
> lot more resources than the standby. What I want to do is run
> pg_basebackup on the main server with the output going to the data
> directory on the new server. But when I give this command:
>
> pg_basebackup -D "ssh root@10.0.1.16:/mnt/dbraid/data" -P -v -X s
>
>
> it instead writes to my root drive which doesn't have the space, so it
> fails and deletes the partial backup.
What you might be thinking of is the "old" method of doing base backups
before pg_basebackup: Call pg_start_backup() and then do file system
operations (tar, scp, whatever) to move the data files to where you want
them. This is mostly obsolete. You should run pg_basebackup on the
host where you want to set up your standby
Thanks. It’s been a while since I set up replication. Not to mention several Postgres versions. I’ve started pg_basebackup from the standby. It failed once due to an ssh error, but I reloaded sshd and started again. May take a while. It about 750gb.
.
--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
Chuck Martin
Avondale Software
