Re: Need to fix one more glitch in upgrade to -10.2

On Sat, 17 Feb 2018, Adrian Klaver wrote:

> Did pg_upgrade spit out any warnings/errors?

Adrian,

   Yes. The uid and gid were mis-matched and, because of that, the
data/directory and all its files were owned by group user, not group
postgres.

> In your previous post you showed:
> # /etc/rc.postgresql reload
>
> yet below shows:
>
> /etc/rc.d/rc.postgresql

   No, the first left off the rc.d/ directory.

> Cut and paste error or not?
>
> So are env variables set correctly?
>
> Your ps ax output showed a Postgres instance running:
>
> postgres -D /var/lib/pgsql/10.2/data

   Well, I had 'killall postgres' and the server shut down. Don't know why
that was displayed.

> Do you know what port it is using and try to connect to it?

   The default: /tmp/.s.PGSQL.5432. Everything's shut down now. So I try:

[root@salmo /etc/rc.d]# killall postgres
[root@salmo /etc/rc.d]# ./rc.postgresql start
Could not find 'postgres' binary. Maybe PostgreSQL is not installed properly?

   Yet,

# ll /usr/bin/postgres 
lrwxrwxrwx 1 root root 35 Feb 17 09:30 /usr/bin/postgres -> ../lib/postgresql/10.2/bin/postgres*

Hence, my confustion.

Rich