pg_upgradecluster 的问题

星期五我使用pg_upgradecluster（在 Debian Buster 上）从 PostgreSQL 13 升级到 14。这很有效，所以今天我想升级另一台服务器（Linux Mint 20.3 Cinnamon）。这一次它不起作用。这是我所做的：

root@farao:~# pg_lsclusters
Ver Cluster Port Status Owner    Data directory              Log file
13  main    5432 online postgres /srv/local/postgresql/13    /var/log/postgresql/postgresql-13-main.log
14  main    5433 online postgres /var/lib/postgresql/14/main /var/log/postgresql/postgresql-14-main.log
root@farao:~# pg_dropcluster --stop 14 main
root@farao:~# pg_upgradecluster -V -v 13 main /srv/local/postgresql/data/14
Use of uninitialized value $newversion in string eq at /usr/bin/pg_upgradecluster line 398.
Use of uninitialized value $_[0] in concatenation (.) or string at /usr/share/perl5/PgCommon.pm line 1049.
Stopping old cluster...
Restarting old cluster with restricted connections...
Notice: extra pg_ctl/postgres options given, bypassing systemctl for start operation
Use of uninitialized value $argv[10] in system at /usr/bin/pg_upgradecluster line 461.
Error: invalid version ''
Error: Could not create target cluster
root@farao:~# pg_lsclusters
Ver Cluster Port Status Owner    Data directory           Log file
13  main    5432 online postgres /srv/local/postgresql/13 /var/log/postgresql/postgresql-13-main.log
root@farao:~# ps -ef | grep postgres
postgres   24351    1306  0 20:57 ?        00:00:00 /usr/lib/postgresql/13/bin/postgres -D /srv/local/postgresql/13 -c config_file=/etc/postgresql/13/main/postgresql.conf -c hba_file=/tmp/pg_hba.UabjBQ.conf
postgres   24353   24351  0 20:57 ?        00:00:00 postgres: 13/main: checkpointer 
postgres   24354   24351  0 20:57 ?        00:00:00 postgres: 13/main: background writer 
postgres   24355   24351  0 20:57 ?        00:00:00 postgres: 13/main: walwriter 
postgres   24356   24351  0 20:57 ?        00:00:00 postgres: 13/main: autovacuum launcher 
postgres   24357   24351  0 20:57 ?        00:00:00 postgres: 13/main: stats collector 
postgres   24358   24351  0 20:57 ?        00:00:00 postgres: 13/main: logical replication launcher 
root       24488   18929  0 20:58 pts/0    00:00:00 grep --color=auto postgres
root@farao:~# cat /tmp/pg_hba.UabjBQ.conf
local all postgres ident
root@farao:~# pg_upgradecluster -V -v 13 main /srv/local/postgresql/data/14
Use of uninitialized value $newversion in string eq at /usr/bin/pg_upgradecluster line 398.
Use of uninitialized value $_[0] in concatenation (.) or string at /usr/share/perl5/PgCommon.pm line 1049.
Stopping old cluster...
Restarting old cluster with restricted connections...
Notice: extra pg_ctl/postgres options given, bypassing systemctl for start operation
Use of uninitialized value $argv[10] in system at /usr/bin/pg_upgradecluster line 461.
Error: invalid version ''
Error: Could not create target cluster
root@farao:~# pg_upgradecluster
Usage: /usr/bin/pg_upgradecluster [OPTIONS] <old version> <cluster name> [<new data directory>]
root@farao:~# 

没有-V参数，结果是一样的。我做错了什么？我需要安装 PERL 模块吗？

备注我只记得我只安装了 postgresql-13 并使用了仍然存在的数据库和配置。PostgreSQL 是否在某处存储有关它创建的集群的信息initdb？不合逻辑，因为pg_lsclusters可以找到它。