| From: | Mark Kirkwood <mark(dot)kirkwood(at)catalyst(dot)net(dot)nz> |
|---|---|
| To: | Azimuddin Mohammed <azimeiu(at)gmail(dot)com> |
| Cc: | Peter Eisentraut <peter(dot)eisentraut(at)2ndquadrant(dot)com>, pgsql-admin(at)postgresql(dot)org |
| Subject: | Re: Replication mode |
| Date: | 2018-01-11 01:37:07 |
| Message-ID: | 486c6844-6e55-0151-449c-6ccf5dcc3cf6@catalyst.net.nz |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-admin |
On 11/01/18 14:10, Azimuddin Mohammed wrote:
> :-)
> Our fear is using open source tools for HA creates dependency and
> single point of failure(pgpool). Or say pacemaker which our enterprise
> do not recommend.
> We are planning to use postgres to use for our hadoop metastore which
> is presently on Oracle.
>
Ok, I understand. The usual way this single point of failure thing is
handled is to use something like a pair of HAProxy servers with
redundant network switches (i.e make the HA manager itself HA). While
this is a bit fiddly it is reasonably well documented and standard
(experienced system and network admins should have no difficultly
getting it to work).
There are also specialized Postgres related solutions for HA (e.g repmgr
from 2ndquadrant) that might be worth a look as well.
regards
Mark
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Rui DeSousa | 2018-01-11 01:39:00 | Re: Replication mode |
| Previous Message | Azimuddin Mohammed | 2018-01-11 01:10:40 | Re: Replication mode |