Re: [ADMIN] Recovery from Transaction wraparound failure

Поиск
Список
Период
Сортировка
От Tom Lane
Тема Re: [ADMIN] Recovery from Transaction wraparound failure
Дата
Msg-id 2262.1486570427@sss.pgh.pa.us
обсуждение исходный текст
Ответ на [ADMIN] Recovery from Transaction wraparound failure  (Ganesh Kannan <ganesh.kannan@weatheranalytics.com>)
Ответы Re: [ADMIN] Recovery from Transaction wraparound failure  (Ganesh Kannan <ganesh.kannan@weatheranalytics.com>)
Список pgsql-admin
Ganesh Kannan <ganesh.kannan@weatheranalytics.com> writes:
> 1) After reading the documentation, I thought I could try to vacuum
> those 20 tables that were never vacuumed individually ( with relfrozenid
> more than 2 billion) in single-user mode, but it is not working (error
> copied below).

That is not an error, it's just a warning:

> backend>  vacuum freeze schema.wd_p51_y2015;
> 2017-02-07 23:25:15 EST [54045]: [8-1] user=,db=,app=,client= WARNING:  database "db1" must be vacuumed within 999999
transactions
> 2017-02-07 23:25:15 EST [54045]: [9-1] user=,db=,app=,client= HINT:  To avoid a database shutdown, execute a
database-wideVACUUM in that database. 

It's telling you that you haven't finished fixing the problem, not that
this command didn't do anything.

> Is there a way to do vacuum of individual tables in
> single-user mode?

You're doing it.  I don't think you really need the "freeze" modifier
though, and that may be forcing more I/O than is needed to get out of
trouble.

            regards, tom lane


В списке pgsql-admin по дате отправления:

Предыдущее
От: Ganesh Kannan
Дата:
Сообщение: Re: [ADMIN] Recovery from Transaction wraparound failure
Следующее
От: "Lazaro Garcia"
Дата:
Сообщение: [ADMIN] Why pgpool TPS is lowest versus postgresql direct connections?