Re: Replication: How to query current segments allocation relative to "Wal keep segments"?
От | Edson Carlos Ericksson Richter |
---|---|
Тема | Re: Replication: How to query current segments allocation relative to "Wal keep segments"? |
Дата | |
Msg-id | BLU436-SMTP166099D4BDAD8C0C36CBECDCF580@phx.gbl обсуждение исходный текст |
Ответ на | Re: Replication: How to query current segments allocation relative to "Wal keep segments"? (Michael Paquier <michael.paquier@gmail.com>) |
Ответы |
Re: Replication: How to query current segments allocation
relative to "Wal keep segments"?
|
Список | pgsql-general |
On 05-01-2015 02:08, Michael Paquier wrote: > On Sun, Jan 4, 2015 at 1:48 AM, Edson Carlos Ericksson Richter > <edsonrichter@hotmail.com> wrote: >> How to query current segments allocation relative to "Wal keep segments" in >> each master server? > What is your server version? You can have a look at > pg_stat_replication on the master which contains information about the > WAL segments written, flushed and replayed on each slave: > http://www.postgresql.org/docs/devel/static/monitoring-stats.html#PG-STAT-REPLICATION-VIEW In production servers, 9.3.5. In development servers, 9.4.0. I had a quick look at this view before: Example (on 9.3.5): postgres=# select * from pg_stat_replication; pid | usesysid | usename | application_name | client_addr | client_hostname | client_port | backend_start | state | sent_location | write_location | flush_location | replay_location | sync_priority | sync_state ------+----------+----------+------------------+-------------+-----------------+-------------+------------------------------+-----------+---------------+--------------- -+----------------+-----------------+---------------+------------ 2808 | 10 | postgres | walreceiver | 10.68.73.1 | | 36075 | 2015-01-03 20:17:48.53706-02 | streaming | 22/F94D1A90 | 22/F94D1A90 | 22/F94D1A90 | 22/F94D1A90 | 0 | async (1 registro) but in either case (9.3.5, 9.4.0), I get lots of info, but not the count of wal_segments consumed. Would this kind of count being recorded somewhere else? How does the server knows that the wal_segments have been exhausted? Thanks, Edson
В списке pgsql-general по дате отправления: