On 12-06-19 02:57 AM, Raghav wrote:
> Hi Steve,
>
> Firstly, Thanks for all the help to my other thread I raised and helping
> me in moving forward with my testing.
> Am playing a bit with Slony these days on high end replication solutions.
>
> Am using Slony-I 2.1.0, I followed the documentation for adding the node
> to the existing replication, however when I try to subscribe the new
> node its going into waiting. Am doing in localhost of same cluster with
> multiple databases.
>

> 1. Create database
> create database slave2;
>
> 2. Restore the master structure to slave2;
>
> pg_dump -h localhost -s -p 5432 -d master | psql slave
>
> 3. Drop _myrep schema which copied by pg_dump to slave ( Note: This is
> not documented, I suspect when we copy masterdb schema structure u also
> get the _slonyschema as well. Which throw error as _slonyschema
> existing. Do we really need master _slonyschema to new node ?)

No you do not need to dump the slonyschema, you can exclude that from 
your dump.


>
> psql -d slave -c "drop schema _myrep cascade;"
>
> 4. below is the script to add new node:
>
> #!/bin/bash
>
> slonik <<_eof_
> cluster name = myrep;
> node 1 admin conninfo='host=127.0.0.1 dbname=master user=postgres
> port=5432';
> node 2 admin conninfo='host=127.0.0.1 dbname=slave1 user=postgres
> port=5432';
> node 3 admin conninfo='host=127.0.0.1 dbname=slave2 user=postgres
> port=5432';
>
> store node (id = 3, event node = 1, comment = 'Slave2 Node For The
> Primary postgres');
>
> #Setting Store Paths ...
> echo 'Stored all nodes in the slony catalogs';
> store path(server = 1, client = 3, conninfo='host=127.0.0.1
> dbname=master user=postgres port=5432');
> store path(server = 3, client = 1, conninfo='host=127.0.0.1
> dbname=slave2 user=postgres port=5432');
> _eof_
>
> This goes fine..
>
> 5. When I do subscribe its going into waiting stage.
>
> Subscribe script:
>
> #!/bin/bash
> # Subscribe node
>
> slonik <<_eof_
> cluster name = myrep;
> node 1 admin conninfo='host=127.0.0.1 dbname=master user=postgres
> port=5432';
> node 2 admin conninfo='host=127.0.0.1 dbname=slave user=postgres port=5432';
> node 3 admin conninfo='host=127.0.0.1 dbname=slave2 user=postgres
> port=5432';
>
> try { subscribe set (id = 1, provider = 1 , receiver = 3, forward = yes,
> omit copy = false); } on error { exit 1; } echo 'Subscribed nodes to set 1';
> _eof_
>
> output of the subscribe node;
> -bash-4.1$ ./new_node_subscribe-set.sh
> waiting for events  (3,5000000001) only at (3,0) to be confirmed on node 1
> waiting for events  (3,5000000001) only at (3,0) to be confirmed on node 1
> waiting for events  (3,5000000001) only at (3,0) to be confirmed on node 1
>
> I found the link which pointing to some bug. Correct me, how to achive this.

Did you start the slon daemon for the new node?  If not, starting the 
slon daemon for the new node should allow things to replicate.


>
> http://www.slony.info/bugzilla/show_bug.cgi?id=223
>
>
> --
> Regards
> Raghav
> Blog: htt://raghavt.blogspot.com/ <http://raghavt.blogspot.com/>
>
>
>
> _______________________________________________
> Slony1-general mailing list
> Slony1-general at lists.slony.info
> http://lists.slony.info/mailman/listinfo/slony1-general