Re: [Proposal] add portaddr like hostaddr

Hi Hackers!

Attached is v3 of the portaddr series. No design changes; this is the
result of a self-review pass over v2, plus a rebase onto current master
(applies cleanly on 957d4eae52e).

Changes since v2:

* The new TAP test is now actually exercised by CI: portaddr was missing
from PG_TEST_EXTRA in the CI workflow, so 007_portaddr.pl was skipped
on every automated run. Same treatment as load_balance got in
7f5b19817ea.

* pg_isready now prints the port it actually probes when portaddr is
given, the same way it already prefers hostaddr over host in its
output. Previously it probed portaddr but printed port.

* pg_upgrade refuses to run with PGPORTADDR set. check_pghost_envvar()
exists precisely to keep a stray libpq environment variable from
silently redirecting pg_upgrade's connections, and PGPORTADDR is a
redirect hazard of exactly that class (the series already scrubs it in
pg_regress and PostgreSQL::Test::Utils for the same reason).

* pg_regress --use-existing now also drops an inherited PGPORTADDR when
--host alone is overridden, not only --port, matching the series' own
rule that a portaddr is stale when either half of the endpoint changes.

* Added the LIBPQ_HAS_PORTADDR feature macro for PQportaddr(), per the
policy of shipping a feature-test macro with new libpq API (6991e774e).

* psql: the \conninfo "Port Address" row is no longer suppressed when a
socket-path host is overridden into a TCP connection by hostaddr (the
one corner where the row was hidden while portaddr was in effect), and
the \connect success banner now reports the port address when it
differs, mirroring what 6e5f8d489a did for hostaddr.

* Documentation touch-ups: PQport() now spells out the
portaddr-without-port case (returns the default port number); the
pgpass section states the default-port fallback for the port field,
matching the adjacent host-field sentence; a \connect wording fix in
the psql reference page.

* Commit message fixes, including removing the "ssh -L 127.0.0.1:0:..."
example: OpenSSH rejects port 0 on -L (it is only valid for -R), which
is exactly why the tools that need a dynamic local port pick a free
one themselves right before opening the tunnel.

Thank you,
BR,
Diego

On 2026-08-13 15:49, Diego wrote:
> Hi Hackers!
>
> tl;dr: I want to add portaddr, like hostaddr, when the ssh tunnels
> uses dynamic ports is a mess.
>        From my last attempt, [1], I undertand it sounds better in my
> mind that the first try ;p
>
>
> Proposal
> --------
>
> Add a libpq connection parameter that says where the server is actually
> reached, so that port can go on identifying it -- exactly what hostaddr
> does for host.  The .pgpass lookup then stays on the logical (host, port)
> pair, and the password file code does not change at all.
>
> I do not have a strong opinion on the name: portaddr is the one that
> mirrors hostaddr, but I am happy to take suggestions.
>
> It is also easy to reason about for security: like hostaddr, it is an
> explicit, opt-in assertion by the user, and nothing changes unless it is
> given.
>
> This is not hypothetical. I ran into it myself while adding SSH tunnel
> support to pgcli (a widely used Postgres CLI): with the tunnel active, an
> explicit-port .pgpass entry never matches, because the lookup happens
> against the random local forwarding port. The user is prompted for a
> password even though the matching entry is right there, and only a
> wildcard port papers over it. Other tools hit the same wall:
>
> - pgAdmin 4: control the SSH tunnel local port for .pgpass matching
> https://github.com/pgadmin-org/pgadmin4/issues/6903
>
> - DBeaver: .pgpass looked up by 127.0.0.1 through an SSH tunnel
> https://github.com/dbeaver/dbeaver/issues/16499
>
> - pgcli: SSH tunnel rewrites the port before the .pgpass lookup (myself)
> https://github.com/dbcli/pgcli/pull/1546
>
>
> Recap, since the approach changed
> ---------------------------------
>
> The first version of this proposal added a parameter that affected only
> the .pgpass lookup.  Christoph pointed out [2] that for a tunnel you open
> by hand, with a fixed local port, you can simply write that port into
> .pgpass.  That is correct, and I withdrew that part of the motivation.
>
> What remains is the case where the local port is not known when the
> password file is written, because the tunnel is not opened by hand.
> Clients that open it for you bind the local end to port 0 and let the
> kernel choose.  pgcli, the case I ran into, does exactly that: it hands
> sshtunnel a local bind address with no port in it,
>
>     "local_bind_address": ("127.0.0.1",),
>
> and only afterwards asks which port it got,
>
>     port = self.ssh_tunnel.local_bind_ports[0]
>
> which it then substitutes into the connection string.  The forwarding
> port is different on every run, and those two lines are precisely what
> breaks the .pgpass lookup.  pgAdmin 4 and DBeaver open their own tunnels
> the same way, which is what the two reports above are about, and outside
> SSH the same shape shows up in kubectl port-forward, where ":5432" means
> "listen on a random local port".
>
> There is no port you can write into .pgpass in advance.
>
> (Plain ssh(1) cannot even express this: OpenSSH rejects port 0 on -L and
> accepts it only for -R, so the wrappers pick a free local port
> themselves and pass it to -L.  Same outcome for .pgpass.)
>
> The model
> ---------
>
> libpq already allows the host that identifies a server to differ from the
> address it is reached at:
>
>     host identifies the server, hostaddr is where we connect,
>     and the password file is searched by host.
>
> v2 does the same for ports, instead of adding a password-file-only knob:
>
>     port identifies the server, portaddr is where we connect,
>     and the password file is searched by port.
>
> The tunnel case is then spelled
>
>     host=db.example.com hostaddr=127.0.0.1 port=5432 portaddr=39907
>
> and the password file entry is the one a direct connection already uses:
>
>     db.example.com:5432:appdb:alice:secret
>
> What I like about this shape is that the password file code does not
> change at all.  passwordFromFile() already receives connhost[i].port, so
> keeping portaddr in a separate field leaves the lookup logical by
> construction.  That is the same reasoning as e3f99e03e2e, which settled
> that the .pgpass host key is host and not hostaddr.
>
> The patches
> -----------
>
> Both patches are attached.  They apply cleanly on master as of
> 957d4eae52e.
>
> 0001  libpq: add portaddr, the port equivalent of hostaddr
>       The parameter, the PGPORTADDR environment variable, documentation,
>       and a TAP test.  It also contains a psql \connect fix, see below.
>
> 0002  libpq: add PQportaddr(), and show the port address in psql \conninfo
>       The accessor mirroring PQhostaddr(), and the psql display. This one
>       is separable: if the list does not want it, 0001 stands on its own.
>
> The new test, src/interfaces/libpq/t/007_portaddr.pl, needs the server to
> listen on TCP, so it is skipped unless PG_TEST_EXTRA lists portaddr:
>
>     make check-world PG_TEST_EXTRA=portaddr
>
> The environment variable works for meson builds as well.  Without it the
> test plans skip_all and nothing else in the suite is affected.
>
> Decisions I made, and where I would like guidance
> -------------------------------------------------
>
> * portaddr applies to TCP only, and is ignored for Unix-domain socket
>   connections, whose socket file name is built from port. hostaddr does
>   not apply to those either.
>
> * The list semantics follow port rather than hostaddr: a single portaddr
>   applies to every host, otherwise the element count must match the host
>   count, and an empty item means "use the corresponding port". hostaddr
>   has to match exactly because it determines the number of hosts;
>   portaddr does not.
>
> * port is still parsed and range-checked even when portaddr overrides it.
>   hostaddr does not validate host, but any string is a plausible host
>   name, whereas a port has a syntax; silently accepting a bad value that
>   then becomes the password file search key seemed worse than erroring
>   out.
>
> * PQport() keeps returning port, mirroring PQhost(), which returns host
>   and not hostaddr.  Its documentation promised "the port actually
>   connected to", so I reworded it, and 0002 adds PQportaddr() for the
>   actual port.
>
> * Connection failure messages report the port actually attempted, since
>   "connection to server at ..., port N failed" is about where we tried to
>   go.
>
> * psql's \connect drops an inherited hostaddr when the host argument
>   changes.  portaddr needs the same treatment, and it has to key on both:
>   a portaddr describes where to reach one particular server, so 0001 drops
>   it when either the host or the port argument changes.  Without that,
>   "\c - - otherhost" would keep tunnelling to the old portaddr on a
>   different machine.  This is in 0001 because leaving it out is a bug, not
>   a missing polish item.
>
> * PGPORTADDR is added to the environment variables cleared by pg_regress
>   and PostgreSQL::Test::Utils, next to PGHOSTADDR, which is scrubbed there
>   for exactly the same reason: it would silently redirect test connections
>   away from the temporary cluster.
>
> * The new TAP test has to make the server listen on TCP, since portaddr
>   does not apply to Unix-domain sockets.  Following ssl, kerberos,
> ldap and
>   load_balance, it runs only when PG_TEST_EXTRA lists portaddr, and
>   regress.sgml documents the new value.  I am not attached to the name, or
>   to having a keyword of its own rather than folding it into an existing
>   one.
>
> The POC #1 first flight
> -----------------------
>
> daf(at)t:postgres$ test=/home/daf/scripts/postgres/portaddr-demo
> daf(at)t:postgres$ BIN_CON="$test/con-el-patch/usr/local/bin"
> daf(at)t:postgres$
> LIB_CON="$test/con-el-patch/usr/local/lib/x86_64-linux-gnu"
> daf(at)t:postgres$ PSQL_CON="$test/con-el-patch/usr/local/bin/psql"
> daf(at)t:postgres$ LIB_SIN="$test/sin-patch/usr/local/lib/x86_64-linux-gnu"
> daf(at)t:postgres$ PSQL_SIN="$test/sin-patch/usr/local/bin/psql"
> daf(at)t:postgres$ unset PGPASSWORD PGPORTADDR
> daf(at)t:postgres$ export LD_LIBRARY_PATH="$LIB_CON"
> daf(at)t:postgres$ export PGDATA=/tmp/pgd-test-portaddr
> daf(at)t:postgres$ export PGPASSFILE=/tmp/pgpass-test-portaddr
> daf(at)t:postgres$
> daf(at)t:postgres$ # init
> daf(at)t:postgres$
> daf(at)t:postgres$ "$BIN_CON/initdb" -D "$PGDATA" -U postgres -A trust
> --no-sync
> The files belonging to this database system will be owned by user "daf".
> This user must also own the server process.
>
> The database cluster will be initialized with locale "C.UTF-8".
> The default database encoding has accordingly been set to "UTF8".
> The default text search configuration will be set to "english".
>
> Data page checksums are enabled.
>
> creating directory /tmp/pgd-test-portaddr ... ok
> creating subdirectories ... ok
> selecting dynamic shared memory implementation ... posix
> selecting default "max_connections" ... 100
> selecting default "shared_buffers" ... 128MB
> selecting default time zone ... America/Argentina/Buenos_Aires
> creating configuration files ... ok
> running bootstrap script ... ok
> performing post-bootstrap initialization ... ok
>
> Sync to disk skipped.
> The data directory might become corrupt if the operating system crashes.
>
> Success. You can now start the database server using:
>
> /home/daf/scripts/postgres/portaddr-demo/con-el-patch/usr/local/bin/pg_ctl
> -D /tmp/pgd-test-portaddr -l logfile start
>
> daf(at)t:postgres$ printf 'local all all trust\nhost all all 127.0.0.1/32
> scram-sha-256\n' > "$PGDATA/pg_hba.conf"
> daf(at)t:postgres$ cat "$PGDATA/pg_hba.conf"
> local all all trust
> host all all 127.0.0.1/32 scram-sha-256
> daf(at)t:postgres$ "$BIN_CON/pg_ctl" -D "$PGDATA" -o "-p 5441 -c
> listen_addresses=127.0.0.1" -w start
> waiting for server to start....2026-08-13 14:23:31.316 -03 [1018117]
> LOG:  starting PostgreSQL 20devel on x86_64-linux, compiled by
> gcc-12.2.0, 64-bit
> 2026-08-13 14:23:31.316 -03 [1018117] LOG:  listening on IPv4 address
> "127.0.0.1", port 5441
> 2026-08-13 14:23:31.319 -03 [1018117] LOG:  listening on Unix socket
> "/tmp/.s.PGSQL.5441"
> 2026-08-13 14:23:31.325 -03 [1018122] LOG:  database system was shut
> down at 2026-08-13 14:22:49 -03
> 2026-08-13 14:23:31.330 -03 [1018117] LOG:  database system is ready
> to accept connections
>  done
> server started
> daf(at)t:postgres$ "$BIN_CON/psql" -X -p 5441 -U postgres -h /tmp -d
> postgres -c "create role tuser login password 'sekret';"
> CREATE ROLE
> daf(at)t:postgres$
> daf(at)t:postgres$ echo "127.0.0.1:5440:postgres:tuser:sekret" >
> "$PGPASSFILE"
> daf(at)t:postgres$ chmod 600 "$PGPASSFILE"
> daf(at)t:postgres$
>
> daf(at)t:postgres$
> daf(at)t:postgres$ cat "$PGPASSFILE"
> 127.0.0.1:5440:postgres:tuser:sekret
> daf(at)t:postgres$
> daf(at)t:postgres$ # 1) Original without patch, FAIL -> bug: .pgpass not
> match
> LD_LIBRARY_PATH="$LIB_SIN" "$PSQL_SIN" -X -w "host=127.0.0.1 port=5441
> user=tuser dbname=postgres" -tAc "select 'AUTENTICO-OK'"
> psql: error: connection to server at "127.0.0.1", port 5441 failed:
> fe_sendauth: no password supplied
> daf(at)t:postgres$
> daf(at)t:postgres$ # 2) without patch + portaddr -> FAIL
> LD_LIBRARY_PATH="$LIB_SIN" "$PSQL_SIN" -X -w "host=127.0.0.1 port=5440
> portaddr=5441 user=tuser dbname=postgres" -tAc "select 'AUTENTICO-OK'"
> psql: error: invalid connection option "portaddr"
> daf(at)t:postgres$
> daf(at)t:postgres$ # 3) WITH patch + portaddr -> AUTENTICA (same line as
> #2, different libpq)
> LD_LIBRARY_PATH="$LIB_CON" "$PSQL_CON" -X -w "host=127.0.0.1 port=5440
> portaddr=5441 user=tuser dbname=postgres" -tAc "select 'AUTENTICO-OK'"
> AUTENTICO-OK
> daf(at)t:postgres$
> daf(at)t:postgres$ # 4) WITH patch, no portaddr -> FAIL, default not changed
> LD_LIBRARY_PATH="$LIB_CON" "$PSQL_CON" -X -w "host=127.0.0.1 port=5441
> user=tuser dbname=postgres" -tAc "select 'AUTENTICO-OK'"
> psql: error: connection to server at "127.0.0.1", port 5441 failed:
> fe_sendauth: no password supplied
> daf(at)t:postgres$
> daf(at)t:postgres$ # 5) WITH patch, env var instead of the parameter ->
> AUTENTICA
> PGPORTADDR=5441 LD_LIBRARY_PATH="$LIB_CON" "$PSQL_CON" -X -w
> "host=127.0.0.1 port=5440 user=tuser dbname=postgres" -tAc "select
> 'AUTENTICO-OK'"
> AUTENTICO-OK
> daf(at)t:postgres$ # 6) WITH patch, what the client sees -> Server Port
> 5440 / Port Address 5441
> LD_LIBRARY_PATH="$LIB_CON" "$PSQL_CON" -X -w "host=127.0.0.1 port=5440
> portaddr=5441 user=tuser dbname=postgres" -c "\conninfo"
>       Connection Information
>       Parameter       |   Value
> ----------------------+-----------
>  Database             | postgres
>  Client User          | tuser
>  Host                 | 127.0.0.1
>  Server Port          | 5440
>  Port Address         | 5441
>  Options              |
>  Protocol Version     | 3.2
>  Password Used        | true
>  GSSAPI Authenticated | false
>  Backend PID          | 1018919
>  SSL Connection       | false
>  Superuser            | off
>  Hot Standby          | off
> (13 rows)
>
> daf(at)t:postgres$
> daf(at)t:postgres$ 2026-08-13 14:28:31.428 -03 [1018120] LOG: checkpoint
> starting: time
> 2026-08-13 14:28:36.176 -03 [1018120] LOG:  checkpoint complete: time:
> wrote 47 buffers (0.3%), wrote 3 SLRU buffers; 0 WAL file(s) added, 0
> removed, 0 recycled; write=4.713 s, sync=0.024 s, total=4.748 s; sync
> files=15, longest=0.018 s, average=0.002 s; distance=345 kB,
> estimate=345 kB; lsn=0/017D4910, redo lsn=0/017D4878
>
> "$BIN_CON/pg_ctl" -D "$PGDATA" -m immediate stop
> waiting for server to shut down....2026-08-13 14:29:17.551 -03
> [1018117] LOG:  received immediate shutdown request
> 2026-08-13 14:29:17.563 -03 [1018117] LOG:  database system is shut down
>  done
> server stopped
>
> The POC #2 withreal case
> ------------------------
>
>
> daf(at)t:postgres$ # ---- setup
> daf(at)t:postgres$ test=/home/daf/scripts/postgres/portaddr-demo
> daf(at)t:postgres$
> LIB_CON="$test/con-el-patch/usr/local/lib/x86_64-linux-gnu"
> daf(at)t:postgres$ PSQL_CON="$test/con-el-patch/usr/local/bin/psql"
> daf(at)t:postgres$ LIB_SIN="$test/sin-patch/usr/local/lib/x86_64-linux-gnu"
> daf(at)t:postgres$ PSQL_SIN="$test/sin-patch/usr/local/bin/psql"
> daf(at)t:postgres$ SSH_HOST=beta
> daf(at)t:postgres$ REAL_HOST=beta.xxxx.yyyy.zzz
> daf(at)t:postgres$ REAL_PORT=55432
> daf(at)t:postgres$ LOCAL_PORT=5533
> daf(at)t:postgres$ SSH_SOCK=/tmp/portaddr-beta-ssh.sock
> daf(at)t:postgres$
> daf(at)t:postgres$ unset PGPASSWORD PGPORTADDR
> daf(at)t:postgres$ export PGPASSFILE=/tmp/pgpass-demo-portaddr-beta
> daf(at)t:postgres$
> daf(at)t:postgres$ ssh -M -S "$SSH_SOCK" -f -N "$SSH_HOST"
> daf(at)t:postgres$ ssh -S "$SSH_SOCK" -O forward -L
> $LOCAL_PORT:localhost:$REAL_PORT "$SSH_HOST"
> daf(at)t:postgres$ ss -ltn | grep $LOCAL_PORT
> LISTEN 0      128        127.0.0.1:5533       0.0.0.0:*
> daf(at)t:postgres$
> daf(at)t:postgres$ cat "$PGPASSFILE"
> 127.0.0.1:55432:*:daf:sekret
> daf(at)t:postgres$ chmod 600 "$PGPASSFILE"
> daf(at)t:postgres$
> daf(at)t:postgres$
> daf(at)t:postgres$ # 1) Original without patch, FAIL -> bug: .pgpass no
> matchea el puerto del tunel
> LD_LIBRARY_PATH="$LIB_SIN" "$PSQL_SIN" -X -w "host=127.0.0.1
> port=$LOCAL_PORT user=daf dbname=daf" -tAc "select 'AUTENTICO-OK'"
> psql: error: connection to server at "127.0.0.1", port 5533 failed:
> fe_sendauth: no password supplied
> daf(at)t:postgres$
> daf(at)t:postgres$ # 2) without patch + portaddr -> FAIL
> LD_LIBRARY_PATH="$LIB_SIN" "$PSQL_SIN" -X -w "host=127.0.0.1
> port=$REAL_PORT portaddr=$LOCAL_PORT user=daf dbname=daf" -tAc "select
> 'AUTENTICO-OK'"
> psql: error: invalid connection option "portaddr"
> daf(at)t:postgres$
> daf(at)t:postgres$ # 3) WITH patch + portaddr -> AUTENTICA (same like #2,
> other libpq)
> LD_LIBRARY_PATH="$LIB_CON" "$PSQL_CON" -X -w "host=127.0.0.1
> port=$REAL_PORT portaddr=$LOCAL_PORT user=daf dbname=daf" -tAc "select
> 'AUTENTICO-OK'"
> AUTENTICO-OK
> daf(at)t:postgres$ # 4) WITH patch, no portaddr -> FAIL, same default
> LD_LIBRARY_PATH="$LIB_CON" "$PSQL_CON" -X -w "host=127.0.0.1
> port=$LOCAL_PORT user=daf dbname=daf" -tAc "select 'AUTENTICO-OK'"
> psql: error: connection to server at "127.0.0.1", port 5533 failed:
> fe_sendauth: no password supplied
> daf(at)t:postgres$
> daf(at)t:postgres$ # 5) WITH patch, env var en vez del parametro -> AUTENTICA
> PGPORTADDR=$LOCAL_PORT LD_LIBRARY_PATH="$LIB_CON" "$PSQL_CON" -X -w
> "host=127.0.0.1 port=$REAL_PORT user=daf dbname=daf" -tAc "select
> 'AUTENTICO-OK'"
> AUTENTICO-OK
> daf(at)t:postgres$
> daf(at)t:postgres$ # 6) WITH patch, lo que ve el cliente -> Server Port
> 55432 / Port Address 5533 / Password Used true
> LD_LIBRARY_PATH="$LIB_CON" "$PSQL_CON" -X -w "host=127.0.0.1
> port=$REAL_PORT portaddr=$LOCAL_PORT user=daf dbname=daf" -c "\conninfo"
>       Connection Information
>       Parameter       |   Value
> ----------------------+-----------
>  Database             | daf
>  Client User          | daf
>  Host                 | 127.0.0.1
>  Server Port          | 55432
>  Port Address         | 5533
>  Options              |
>  Protocol Version     | 3.0
>  Password Used        | true
>  GSSAPI Authenticated | false
>  Backend PID          | 290199
>  SSL Connection       | false
>  Superuser            | on
>  Hot Standby          | off
> (13 rows)
>
> daf(at)t:postgres$
> daf(at)t:postgres$
> # ---------------------------------------------------------------
> # 7) THE MAIN CASE: the full symmetry, using your REAL ~/.pgpass untouched
> #
> #   host     = beta.xxxx.yyyy.zzz   hostaddr = 127.0.0.1
> #   port     = 55432                      portaddr = 5533
> #
> # The LOGICAL pair (host, port) is how you know the server, and what
> the .pgpass
> # lookup uses; the PHYSICAL pair (hostaddr, portaddr) is where the
> socket really
> # goes. Without portaddr, hostaddr alone is not enough: you can lie
> about the
> # host, but the port still gives you away.
>
> daf(at)t:postgres$ unset PGPASSFILE
> daf(at)t:postgres$
> daf(at)t:postgres$ # without patch: hostaddr has been there for years,
> but the lookup still uses the tunnel port -> FAILS
> LD_LIBRARY_PATH="$LIB_SIN" "$PSQL_SIN" -X -w "host=$REAL_HOST
> hostaddr=127.0.0.1 port=$LOCAL_PORT user=daf dbname=daf" -tAc "select
> 'AUTENTICO-OK'"
> psql: error: connection to server at "127.0.0.1", port 5533 failed:
> fe_sendauth: no password supplied
> daf(at)t:postgres$
> daf(at)t:postgres$ # WITH patch: (host,port) for the lookup +
> (hostaddr,portaddr) for the socket -> AUTHENTICATES
> LD_LIBRARY_PATH="$LIB_CON" "$PSQL_CON" -X -w "host=$REAL_HOST
> hostaddr=127.0.0.1 port=$REAL_PORT portaddr=$LOCAL_PORT user=daf
> dbname=daf" -tAc "select 'AUTENTICO-OK'"
> AUTENTICO-OK
> daf(at)t:postgres$
> daf(at)t:postgres$ # \conninfo shows the 4 rows together: Host / Host
> Address / Server Port / Port Address
> LD_LIBRARY_PATH="$LIB_CON" "$PSQL_CON" -X -w "host=$REAL_HOST
> hostaddr=127.0.0.1 port=$REAL_PORT portaddr=$LOCAL_PORT user=daf
> dbname=daf" -c "\conninfo"
>              Connection Information
>       Parameter       |          Value
> ----------------------+--------------------------
>  Database             | daf
>  Client User          | daf
>  Host                 | beta.xxxx.yyyy.zzz
>  Host Address         | 127.0.0.1
>  Server Port          | 55432
>  Port Address         | 5533
>  Options              |
>  Protocol Version     | 3.0
>  Password Used        | true
>  GSSAPI Authenticated | false
>  Backend PID          | 290265
>  SSL Connection       | false
>  Superuser            | on
>  Hot Standby          | off
> (14 rows)
>
>
> And that's all folks.
> Please, feel free to send feedback.
>
>
> [1]
> https://www.postgresql.org/message-id/flat/001a6f1d-4adb-42b2-8bf6-44154ed0ab97%40gmail.com
> [2] https://postgr.es/m/ak6FwxXWcBTvvpPo@msg.df7cb.de
>
>
>
> Thank you all,
> BR,
> Diego