| From: | Stephen Frost <sfrost(at)snowman(dot)net> |
|---|---|
| To: | Magnus Hagander <magnus(at)hagander(dot)net> |
| Cc: | Robert Haas <robertmhaas(at)gmail(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Scanning pg_tablespace from walsender |
| Date: | 2011-01-03 16:29:46 |
| Message-ID: | 20110103162946.GK4933@tamriel.snowman.net |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
* Magnus Hagander (magnus(at)hagander(dot)net) wrote:
> It's already a basic requirement that they need to be the same -
> replicating over a CREATE TABLESPACE is going to fail otherwise..
> Being able to deal with that in a nice way would be good, of course,
> but we don't have that today.
If CREATE TABLESPACE replication also looked at the flag I was
proposing, it could work. :) Of course, the admin wouldn't be able
to move the directory/change the symlink to where they actually want
it to be w/o taking the replication server down, but I'm not sure
that's a show-stopper...
It's certainly not the cleanest/nicest approach, don't get me wrong, but
I really hate the idea of forcing people to have an identical filesystem
layout on the slave that they have on the master.
Stephen
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Heikki Linnakangas | 2011-01-03 16:35:18 | Re: Re: new patch of MERGE (merge_204) & a question about duplicated ctid |
| Previous Message | Simon Riggs | 2011-01-03 16:29:19 | Re: Re: new patch of MERGE (merge_204) & a question about duplicated ctid |