Re: Possible performance regression with pg_dump of a large number of relations

Поиск
Список
Период
Сортировка
От Jeff Janes
Тема Re: Possible performance regression with pg_dump of a large number of relations
Дата
Msg-id CAMkU=1xq6cfvFzoHhudhkmqVw7xTx=gpZtYWqZUpOaN7nuRbPw@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Possible performance regression with pg_dump of a large number of relations  (Jeff Janes <jeff.janes@gmail.com>)
Ответы Re: Possible performance regression with pg_dump of a large numberof relations
Список pgsql-hackers
Дерево обсуждения
On Fri, Jan 12, 2018 at 8:01 AM, Jeff Janes <jeff.janes@gmail.com> wrote:
That commit covered a few different things, and I don't what improvement it mentions is the one that motivated this, but the key change was to add this query:

EXISTS (SELECT 1 FROM pg_attribute at LEFT JOIN pg_init_privs pip ON(c.oid = pip.objoid AND pip.classoid = (SELECT oid FROM pg_class WHERE relname = 'pg_class') AND pip.objsubid = at.attnum)WHERE at.attrelid = c.oid AND at.attnum>0 and ((SELECT count(acl) FROM (SELECT unnest(coalesce(at.attacl,acldefault('c',c.relowner))) AS acl EXCEPT SELECT unnest(coalesce(pip.initprivs,acldefault('c',c.relowner)))) as foo) >1 OR (SELECT count(acl) FROM (SELECT unnest(coalesce(pip.initprivs,acldefault('c',c.relowner))) AS acl EXCEPT SELECT unnest(coalesce(at.attacl,acldefault('c',c.relowner)))) as foo) >0))AS changed_acl

Sorry, that query reflects some munging I did to it.  The real part added to the query is:

EXISTS (SELECT 1 FROM pg_attribute at LEFT JOIN pg_init_privs pip ON(c.oid = pip.objoid AND pip.classoid = (SELECT oid FROM pg_class WHERE relname = 'pg_class') AND pip.objsubid = at.attnum)WHERE at.attrelid = c.oid AND ((SELECT array_agg(acl) FROM (SELECT unnest(coalesce(at.attacl,acldefault('c',c.relowner))) AS acl EXCEPT SELECT unnest(coalesce(pip.initprivs,acldefault('c',c.relowner)))) as foo) IS NOT NULL OR (SELECT array_agg(acl) FROM (SELECT unnest(coalesce(pip.initprivs,acldefault('c',c.relowner))) AS acl EXCEPT SELECT unnest(coalesce(at.attacl,acldefault('c',c.relowner)))) as foo) IS NOT NULL OR NULL IS NOT NULL OR NULL IS NOT NULL))AS changed_ac
 
Cheers,

Jeff

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Teodor Sigaev
Дата:
Сообщение: Re: pgbench - add \if support
Следующее
От: Peter Eisentraut
Дата:
Сообщение: Re: PATCH: logical_work_mem and logical streaming of largein-progress transactions