Re: Issue upgrading from V11 to V12 on Debian

On Tue, Dec 03, 2019 at 11:35:40AM -0800, Adrian Klaver wrote:
> On 12/3/19 9:51 AM, stan wrote:
> >
> > On Tue, Dec 03, 2019 at 08:58:58AM -0800, Paul Jungwirth wrote:
> > > On 12/3/19 8:46 AM, stan wrote:> So, I have V12 running as the default on
> > > the machine I am testing this on
> > > > now:
> > > >
> > > > Ver Cluster Port Status Owner    Data directory              Log file
> > > > 11  main    5433 down   postgres /var/lib/postgresql/11/main
> > > > /var/log/postgresql/postgresql-11-main.log
> > > > 12  main    5432 online postgres /var/lib/postgresql/12/main
> > > > /var/log/postgresql/postgresql-12-main.log
> > > >
> > > > BUT, I went to the directory where I have the exentsion's source, did a
> > > > make clean ; make ; make install, and the files were still installed in the
> > > > V11 tree. How can I instruct the system to put these in the V12 tree?
> > > >
> > > > The  extension in question is pgemailaddr if this matters, BTW.
> > >
> > > That extension (and every extension I've seen) uses Postgres's normal
> > > extension-building infrastructure, so it runs pg_config to learn where to
> > > put files. If you run pg_config on your system it will probably report
> > > directories belonging to v11. (That's a little surprising because on Ubuntu
> > > systems I've always had it report the latest version.)
> > >
> > > Many other Ubuntu Postgres commands accept a PGCLUSTER envvar to specific
> > > which cluster to use. If you want to add that to your pg_config you could do
> > > it like this: https://stackoverflow.com/a/43403193/122087 Then just set
> > > PGCLUSTER before building. (Make sure you `make clean` first.)
> > >
> > And weirder.
> >
> > So I purged the V1 server again, edited the postgresql.conf file to put it
> > back to port 5432, reinstalled the V12 package, and now:
> >
> > stan@stantest:~/src/pgemailaddr-master$ pg_lsclusters
> > Ver Cluster Port Status Owner    Data directory              Log file
> > 11  main    5432 online postgres /var/lib/postgresql/11/main
> > /var/log/postgresql/postgresql-11-main.log
> > 12  main    5434 online postgres /var/lib/postgresql/12/main
> > /var/log/postgresql/postgresql-12-main.log
> >
> > So, I decided to take a quick look at the production machine, just to make
> > sure this matched up to what was going on over there, and look what I found:
> >
> > stan@ica-db:~$ htop
> > stan@ica-db:~$ pg_lsclusters
> > Ver Cluster Port Status Owner    Data directory              Log file
> > 11  main    5432 online postgres /var/lib/postgresql/11/main
> > /var/log/postgresql/postgresql-11-main.log
> > 12  main    5432 online postgres /var/lib/postgresql/11/main
> > /var/log/postgresql/postgresql-12-main.log
> >
> > This seems impossible to me.
> >
>
> What does ps ax | grep post show?
>

On the production machine, just teh V11 biaries.

Looks like the pg_lsclusters methond may not be totaly relaible.

BGot the test machine upgraded.

Thanks to all the helpful folks that provided expert advice on this.
--
"They that would give up essential liberty for temporary safety deserve
neither liberty nor safety."
                        -- Benjamin Franklin