| From: | Guillaume Lelarge <guillaume(at)lelarge(dot)info> |
|---|---|
| To: | "David F(dot) Skoll" <dfs(at)roaringpenguin(dot)com> |
| Cc: | pgsql-admin(at)postgresql(dot)org |
| Subject: | Re: Upgrading to 9.4 to get replication slots (was Re: Streaming replication and WAL archiving) |
| Date: | 2015-02-11 17:02:50 |
| Message-ID: | CAECtzeWOxKcW7-u3DTU15-SC+FbNNij+kijnzf1nAq6xdz9B6g@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-admin |
Le 11 févr. 2015 16:42, "David F. Skoll" <dfs(at)roaringpenguin(dot)com> a écrit :
>
> On Wed, 11 Feb 2015 08:04:43 -0700
> Scott Ribe <scott_ribe(at)elevated-dev(dot)com> wrote:
>
> > I don't remember the beginning of the thread. Did you consider Slony?
> > It's more configuration, but it would reduce downtime to almost 0.
>
> We have looked at Slony in the past, but AFAIK it cannot replicate
> tables that lack a primary key and we have a few of those.
This is a good reason to avoid slony but...
> Also
> AFAIK, slony requires an initial sync of the databases with
> dump/restore. Finally our database is quite busy and very
> write-heavy, so I worry about the performance impact of all the Slony
> triggers.
>
This is definitely wrong. Slony can do the initial sync by itself (ie
without dump/restore).
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Elías David | 2015-02-12 01:27:26 | Re: pg_createcluster failing on Trusty |
| Previous Message | David F. Skoll | 2015-02-11 15:40:06 | Re: Upgrading to 9.4 to get replication slots (was Re: Streaming replication and WAL archiving) |