Re: Streaming Questions and Sizing
От | Adrian Klaver |
---|---|
Тема | Re: Streaming Questions and Sizing |
Дата | |
Msg-id | 5593EB48.5010400@aklaver.com обсуждение исходный текст |
Ответ на | Streaming Questions and Sizing (Andy Erskine <andy.erskine@jds.net.au>) |
Список | pgsql-general |
On 07/01/2015 06:24 AM, Andy Erskine wrote: CCing list > Yes that's configured on the master So the segments would pile up there not on the standby. You do not say what OS you are using, but if it is a Unixen variation, then run: du -h on the master and standby directories to see where the capacity is being used. > > On 1 Jul 2015 11:20 pm, "Adrian Klaver" <adrian.klaver@aklaver.com > <mailto:adrian.klaver@aklaver.com>> wrote: > > On 06/30/2015 09:14 PM, Andy Erskine wrote: > > I have db of 123GB Currently and this is streaming to a secondary DB > which also shows a size of 123GB (Using pgAdmin) > > The db's both reside on a 250GB directorys and on the Master i'm > using > 60% of capacity which seems expected > On the secondary i am using 88% of the disks capacity. I assume > this is > something to do with the WAL segments ? which is currently > wal_keep_segments = 500 > > > That is for the server sending the segments, so I would assume you > are getting this number from the master not the standby? > > > If this is the issue how do i go about tuning this to what is > required ? > > If i make the segments to small then a restore may end up with an un > synchronised database right ? > > thanks. > > > > -- > Adrian Klaver > adrian.klaver@aklaver.com <mailto:adrian.klaver@aklaver.com> > -- Adrian Klaver adrian.klaver@aklaver.com
В списке pgsql-general по дате отправления: