Postgres 13 cluster in Debian Linux server contains 30 databases. Databases contain number of schemas. How to find biggest files which occupy most space in disk ? I tried
select
relname::char(25),
pg_size_pretty(pg_total_relation_size(c.oid))::char(10) as totalsize,
n.nspname::char(12),
case
when c.relkind='i' then 'index'
when c.relkind='t' then 'toast'
when c.relkind='r' then 'table'
when c.relkind='v' then 'view'
when c.relkind='c' then 'composite type'
when c.relkind='S' then 'sequence'
else c.relkind::text
end ::char(14) as "type"
from
pg_class c
left join pg_namespace n on n.oid = c.relnamespace
left join pg_tablespace t on t.oid = c.reltablespace
where
(pg_total_relation_size(c.oid)>>21)>0 order by
pg_total_relation_size(c.oid) desc
But it returns sizes for current database only. How to run in over whole cluster ? Can some plpgsql script used for this. Output should include database name column.
Client application uses psqlODBC to get data so psql or shell scripts should preferably avoided.