Thread: Can one Dump schema without index/constraints?
I'm working with slon and the index portion for at least 3 of my tables take hours to complete and thus with this instance of slony being a wide area replica, sessions time out and slon fails to complete.
So I'm looking at dumping the schema without index information, install that on the slon slave and replicate that way, once replication of the data is done, I can run commands to create the indexes.
So I'm looking at dumping the schema without index information, install that on the slon slave and replicate that way, once replication of the data is done, I can run commands to create the indexes.
But I'm not 100% if there are tools or private scripts written to pull indexes from a schema only dump and then allow for an easy recreation of the indexes at the end of the slon replication process (once all sets are replicated)?
Thanks
Tory
Thanks
Tory
On Sat, Feb 15, 2014 at 6:06 AM, Tory M Blue <tmblue@gmail.com> wrote: > > > I'm working with slon and the index portion for at least 3 of my tables take > hours to complete and thus with this instance of slony being a wide area > replica, sessions time out and slon fails to complete. > > So I'm looking at dumping the schema without index information, install that > on the slon slave and replicate that way, once replication of the data is > done, I can run commands to create the indexes. > > But I'm not 100% if there are tools or private scripts written to pull > indexes from a schema only dump and then allow for an easy recreation of the > indexes at the end of the slon replication process (once all sets are > replicated)? "pg_dump --section" can be used for that. pre-data includes table definitions and everything other than post-data dumps. post-data has contraint, trigger, index and rules. -- Michael
On Fri, Feb 14, 2014 at 5:55 PM, Michael Paquier <michael.paquier@gmail.com> wrote:
On Sat, Feb 15, 2014 at 6:06 AM, Tory M Blue <tmblue@gmail.com> wrote:
>
>
> I'm working with slon and the index portion for at least 3 of my tables take
> hours to complete and thus with this instance of slony being a wide area
> replica, sessions time out and slon fails to complete.
>
> So I'm looking at dumping the schema without index information, install that
> on the slon slave and replicate that way, once replication of the data is
> done, I can run commands to create the indexes.
>
> But I'm not 100% if there are tools or private scripts written to pull
> indexes from a schema only dump and then allow for an easy recreation of the
> indexes at the end of the slon replication process (once all sets are
> replicated)?
"pg_dump --section" can be used for that. pre-data includes table
definitions and everything other than post-data dumps. post-data has
contraint, trigger, index and rules.
--
Michael
Thanks Michael, that looks like the ticket!! Will give this a whirl
Thanks again
Tory
Thanks again
Tory