Slony1 PostgreSQL Replication - Safe to delete from pg

Author

Safe to delete from pg_namespace?

Andrew Gold

2005-08-03, 8:24 pm

Hi everyone,

I've used the sample scripts to set up my first attempt at replication.
I screwed up on one of the lines and forgot to comment out the line
where it adds an index to the public.history table. The script quits as
it should but cannot be rerun as it exists because the new namespace
(_slony1, in my example) has been created.

I would like to rerun the script, but I need to clean up the partial
initialization (init cluster). Can I safely delete the entry in
pg_namespace, or are there side effects? Is there another way to clean
up a failed initialization? I am adding a pre-existing table from a live
database, so I need to be sure that removing the entry in pg_namespace
doesn't cause problems, although I suppose I could just create the same
scenario on a test machine ...

The table is a junk table which I created for testing, but the parent
database is valid.
set add table (set id=1, origin=1, id=1, fully qualified name =
'public.spam', comment='test table')

I'd like to know for sure.

Thanks,

Andrew Gold

Jan Wieck

2005-08-03, 8:24 pm

On 8/3/2005 7:23 PM, Andrew Gold wrote:
> Hi everyone,
>
> I've used the sample scripts to set up my first attempt at replication.
> I screwed up on one of the lines and forgot to comment out the line
> where it adds an index to the public.history table. The script quits as
> it should but cannot be rerun as it exists because the new namespace
> (_slony1, in my example) has been created.
>
> I would like to rerun the script, but I need to clean up the partial
> initialization (init cluster). Can I safely delete the entry in
> pg_namespace, or are there side effects? Is there another way to clean
> up a failed initialization? I am adding a pre-existing table from a live
> database, so I need to be sure that removing the entry in pg_namespace
> doesn't cause problems, although I suppose I could just create the same
> scenario on a test machine ...
>
> The table is a junk table which I created for testing, but the parent
> database is valid.
> set add table (set id=1, origin=1, id=1, fully qualified name =
> 'public.spam', comment='test table')
>
> I'd like to know for sure.

There's a special slonik command for that.

http://gborg.postgresql.org/project..._uninstall_node

Jan

>
> Thanks,
>
> Andrew Gold
> ____________________
____________________
_______
> Slony1-general mailing list
> Slony1-general- AuKwsB3Fm+ugFIWk8tvy
RWD2FQJk+8+b@public.gmane.org
> http://gborg.postgresql.org/mailman.../slony1-general

Andrew Gold

2005-08-04, 8:24 pm

Thanks for the answer Jan, I was hoping there was a reasonable solution.
I figured that I couldn't be the first person in this situation.

Andrew

Jan Wieck wrote:

> There's a special slonik command for that.
>
> http://gborg.postgresql.org/project..._uninstall_node
>
>
>
> Jan
>
>
>
>

Sponsored Links