Thread: Postgres warm standby with delay
Management wants to have our postgres database server backed-up to a warm-standby server with an 8-hour delay.
The purpose of the delay is to prevent automatic replication of sql mistakes, for example, if someone deletes the wrong data or drops a critical table.
The warm-standby would still have that data for 8 hours to allow time for recovery.
In addition, they want a way to “stop” the replication on-demand and to provide a way to catch-up the warm standby to current.
I’m still trying to find a way to even configure a warm standby server.
If anyone has guides or examples of how to setup a warm standby that would be great.
If anyone knows how to introduce a “lag/delay” into the replication that would also be great.
Currently, im assuming it has something to do with shipping logs from the primary to the secondary server, but I just don’t know how any of this works yet.
Any pointers to useful examples would be appreciated.
Ron
On 3 May 2018 21:20:08 CEST, Ron Watkins <rwatki@gmail.com> wrote: >Management wants to have our postgres database server backed-up to a >warm-standby server with an 8-hour delay. > You are looking for: recovery_min_apply_delay Regards, Andreas -- 2ndQuadrant - The PostgreSQL Support Company
On 3 May 2018 21:20:08 CEST, Ron Watkins <rwatki@gmail.com> wrote:
>Management wants to have our postgres database server backed-up to a
>warm-standby server with an 8-hour delay.
>
You are looking for:
recovery_min_apply_delay
Regards, Andreas
--
2ndQuadrant - The PostgreSQL Support Company
Amit Sharma