Diego Vargas | 26 Jun 2012 22:48
Picon

Restart Nodes

Hi All,

I want to restart two nodes after a switchover.
The thing is that I make switchover and then I switch the IP of both nodes.
And when that occurs, both nodes stop listening each other.
So, what I thought is restart the nodes, so then they can start listening again.
This is part of the code:

        node 1 admin conninfo = 'dbname=$db_one_name port=$db_one_port host=$db_one_host user=$user';
        node 2 admin conninfo = 'dbname=$db_two_name port=$db_two_port host=$db_two_host user=$user';
        echo 'Restarting first node';
        restart node 1;
        wait for event (origin=1, confirmed=2, wait on=1);
        echo 'Node restarted';
        echo 'Restarting second node';
        restart node 2;
        wait for event (origin=2, confirmed=1, wait on=2);
        echo 'Node restarted';

When it's executed, all seems to be okay, but when I try to see the lag time this is what I get:

postgres=# select st_origin, st_received, st_lag_time from _slony_replication.sl_status;
 st_origin | st_received |   st_lag_time  
-----------+-------------+-----------------
         2 |           1 | 00:08:20.927686
(1 row)

And 8 minutes passed since the IP were switched.

Am I doing something wrong?

Sorry for my bad english.
Thanks,
Diego

_______________________________________________
Slony1-general mailing list
Slony1-general@...
http://lists.slony.info/mailman/listinfo/slony1-general
Steve Singer | 26 Jun 2012 22:55

Re: Restart Nodes

On 12-06-26 04:48 PM, Diego Vargas wrote:
> Hi All,
>
> I want to restart two nodes after a switchover.
> The thing is that I make switchover and then I switch the IP of both nodes.
> And when that occurs, both nodes stop listening each other.
> So, what I thought is restart the nodes, so then they can start
> listening again.

You need to issue STORE PATH (
http://www.slony.info/documentation/2.1/stmtstorepath.html) commands in 
that slonik script to store the new IP addresses of your nodes.

> This is part of the code:
>
>          node 1 admin conninfo = 'dbname=$db_one_name port=$db_one_port
> host=$db_one_host user=$user';
>          node 2 admin conninfo = 'dbname=$db_two_name port=$db_two_port
> host=$db_two_host user=$user';
>          echo 'Restarting first node';
>          restart node 1;
>          wait for event (origin=1, confirmed=2, wait on=1);
>          echo 'Node restarted';
>          echo 'Restarting second node';
>          restart node 2;
>          wait for event (origin=2, confirmed=1, wait on=2);
>          echo 'Node restarted';
>
> When it's executed, all seems to be okay, but when I try to see the lag
> time this is what I get:
>
> postgres=# select st_origin, st_received, st_lag_time from
> _slony_replication.sl_status;
>   st_origin | st_received |   st_lag_time
> -----------+-------------+-----------------
>           2 |           1 | 00:08:20.927686
> (1 row)
>
> And 8 minutes passed since the IP were switched.
>
> Am I doing something wrong?
>
> Sorry for my bad english.
> Thanks,
> Diego
>
>
>
> _______________________________________________
> Slony1-general mailing list
> Slony1-general@...
> http://lists.slony.info/mailman/listinfo/slony1-general

Gmane