No reason besides we were new to slony and had no idea what we were doing
at the beginning and now it will be a hassle to change for everyone that
has this.

On Tue, Nov 3, 2015 at 8:50 AM, Jan Wieck <jan at wi3ck.info> wrote:

> On 10/29/2015 09:49 PM, Sung Hsin Lei wrote:
>
>> Actually, node 2 is correct. I have 2 different clusters. Each cluster
>> has node 1 and node 2. Each cluster replicates to 1 db.
>>
>
> What is the reason for this setup? If the two slaves are subscribed to
> different sets of tables, then the same can be achieved by creating
> multiple sets and subscribing the slaves to individual combinations of them.
>
>
> Regards, Jan
>
>
>
>> On Thu, Oct 29, 2015 at 8:39 PM, Scott Marlowe <scott.marlowe at gmail.com
>> <mailto:scott.marlowe at gmail.com>> wrote:
>>
>>     On Thu, Oct 29, 2015 at 6:06 PM, Sung Hsin Lei <sungh.lei at gmail.com
>>     <mailto:sungh.lei at gmail.com>> wrote:
>>     > Hello,
>>     >
>>     > I successfully update the main db and one replicated db with the
>> following:
>>     >
>>     > cluster name = slony_rep1
>>     >
>>     > node 1 admin conninfo = 'dbname = MyDB host = localhost user =
>> slony1
>>     > password = Ejhfg33EdddsufhErR76 port = 6234';
>>     > node 2 admin conninfo = 'dbname = MyDB host = 86.88.5.4 user =
>> slony1
>>     > password = Ejhfg33EdddsufhErR76 port = 6234';
>>     >
>>     > EXECUTE SCRIPT
>>     > (
>>     >     SQL = 'ALTER TABLE operators RENAME COLUMN firstname TO
>> lastname;',
>>     >     EVENT NODE = 1
>>     > );
>>     >
>>     >
>>     >
>>     >
>>     > However, I have 1 main db to 2 replicated db. The second replicated
>> db has
>>     > the following cluster name and node information:
>>     >
>>     > cluster name = slony_rep2
>>     >
>>     > node 1 admin conninfo = 'dbname = MyDB host = localhost user =
>> slony2
>>     > password = Ejhfg33EdddsufhErR76 port = 6234';
>>     > node 2 admin conninfo = 'dbname = MyDB host = 86.88.5.17 user =
>> slony2
>>     > password = Ejhfg33EdddsufhErR76 port = 6234';
>>
>>     Shouldn't that be node 3 not node 2?
>>
>>     > Do I need to run slonik twice with different cluster and node indo?
>> That
>>     > does not seem right. After running the first time, the main db and
>> the first
>>     > replicated db will be consistent but not the second replicated db.
>> Also, if
>>     > I run it a second time, wouldn't the main db already be updated
>> hence the
>>     > sql statements used for the original update will surely fail?
>>
>>     You can just run it on all three at once.
>>
>>     node 1 admin conninfo = 'dbname = MyDB host = localhost user = slony1
>>     password = Ejhfg33EdddsufhErR76 port = 6234';
>>     node 2 admin conninfo = 'dbname = MyDB host = 86.88.5.4 user = slony1
>>     password = Ejhfg33EdddsufhErR76 port = 6234';
>>     node 3 admin conninfo = 'dbname = MyDB host = 86.88.5.17 user = slony2
>>     password = Ejhfg33EdddsufhErR76 port = 6234';
>>
>>     EXECUTE SCRIPT
>>     (
>>          SQL = 'ALTER TABLE operators RENAME COLUMN firstname TO
>> lastname;',
>>          EVENT NODE = 1
>>     );
>>
>>
>>
>>
>> _______________________________________________
>> Slony1-general mailing list
>> Slony1-general at lists.slony.info
>> http://lists.slony.info/mailman/listinfo/slony1-general
>>
>>
>
> --
> Jan Wieck
> Senior Software Engineer
> http://slony.info
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.slony.info/pipermail/slony1-general/attachments/20151105/1e9c0216/attachment.htm