Re: how to fail over slave to master

[Pepe TD Vo <pepevo@xxxxxxxxx>]

In verison 12, postgresql.auto.conf to replace recovery.conf

do I  need to run 
1. pg_ctl promote -D Pg_data (in slave?)

2. pg_rewind --target -pg_data=master_server (aliase hostname of master) --source-server="port=5433 user=replica dbname=postgres" --progress

3. change postgresql.cond parameter, primary_slot_name ='main'

do I need to run pg_rewind for each dbname? 

I don't understand why pg_rewind set the port=5433 not 5432?

Bach-Nga

No one in this world is pure and perfect.  If you avoid people for their mistakes you will be alone. So judge less, love, and forgive more.

To call him a dog hardly seems to do him justice though in as much as he had four legs, a tail, and barked, I admit he was, to all outward appearances. But to those who knew him well, he was a perfect gentleman (Hermione Gingold)

**Live simply **Love generously **Care deeply **Speak kindly.

*** Genuinely rich *** Faithful talent *** Sharing success

On Monday, December 28, 2020, 02:10:14 PM EST, Scott Ribe <scott_ribe@xxxxxxxxxxxxxxxx> wrote:

> On Dec 28, 2020, at 11:54 AM, Pepe TD Vo <pepevo@xxxxxxxxx> wrote:

>
> Version 12 and can you show me step by step what needs to do?  I found a document in version 10 to promote salve to master with pg_ctl and create recovery.conf.  Recovery.conf is not existed in version 12 but in version 1.

1) The parameters that used to be set in recovery.conf are instead set in postgresql.conf in 12 & above.

2) They are set on your replica; otherwise it would not come up as a replica. They are not needed in a master.