On 10/26/2006 7:53 PM, Michael wrote:
> Hi,
>
> I just completed a simple 2 node switchover. Now I’d like to switch
> ‘back’ to my original configuration and I would like to confirm if what
> I’m doing is correct.
>
> My original file was the cookie cutter…
>
>
>
> slonik << _/EOF/_
>
> cluster name = mycluster;
>
> node 1 admin conninfo = ‘dbname=mydb_1 host=##.##.##.##
> port=5432 user=postgres’;
>
> node 2 admin conninfo = ‘dbname=mydb_2 host=##.##.##.##
> port=5432 user=postgres’;
>
> lock set (id = 1, origin = 1);
>
> wait for event (origin = 1, confirmed = 2);
>
> move set (id = 1, old origin = 1, new origin = 2);
>
> wait for event (origin = 1, confirmed = 2);
>
> _/EOF/_
>
>
>
> Now to get things back to where they were before I did the switchover, I
> assume that I should execute…
>
>
>
>
>
> slonik << _/EOF/_
>
> cluster name = mycluster;
>
> node 1 admin conninfo = ‘dbname=mydb_1 host=##.##.##.##
> port=5432 user=postgres’;
>
> node 2 admin conninfo = ‘dbname=mydb_2 host=##.##.##.##
> port=5432 user=postgres’;
>
> lock set (id = 2, origin = 2);
>
> wait for event (origin = 2, confirmed = 1);
>
> move set (id = 2, old origin = 2, new origin = 1);
>
> wait for event (origin = 2, confirmed = 1);
>
> _/EOF/_
No.
None of the ID's actually change their meaning when switching. Set 1 is
still set 1, node 1 is still node 1 and so on. All that has changed is
the origin (home) of set 1 (in your case).
So to switch back, you would do
lock set (id = 1, origin = 2);
wait for event (origin = 2, confirmed = 1);
move set (id = 1, old origin = 2, new origin = 1);
wait for event (origin = 2, confirmed = 1);
Jan
--
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#================================================== [EMAIL PROTECTED] #
_______________________________________________
Slony1-general mailing list
[email protected]
http://gborg.postgresql.org/mailman/listinfo/slony1-general
