| From: | Robert Haas <robertmhaas(at)gmail(dot)com> |
|---|---|
| To: | chinnaobi <chinnaobi(at)gmail(dot)com> |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: hot standby PSQL 9.1 Windows 2008 Servers |
| Date: | 2012-06-26 16:11:21 |
| Message-ID: | CA+Tgmoa_rsqwqOyOryq-cGzi5TAJiJJxeJ2=0xOn=AWUxbywNg@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Tue, May 22, 2012 at 12:15 PM, chinnaobi <chinnaobi(at)gmail(dot)com> wrote:
> You mean when the primary which is going to switch its role to standby might
> not have sent all the WAL records to the standby and If it is switched to
> standby it has more WAL records than the standby which is now serves as
> primary. Is it ??
Yes, that is possible. Or the standby might have received all the WAL
records but not be caught up in terms of replaying them.
> It is actually the standby server which has to be restored from archive when
> it is switching to primary right .. Not the primary which is switching to
> standby ??
If you want to promote a standby, you can just do it (pg_ctl promote).
If you have a master that you want to demote to a standby, you've got
to resync it to whatever the current master is. I understand repmgr
has some tooling to help automate that, although I have not played
with it myself. In any event rsync can be a big help in reducing the
resync time.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Alvaro Herrera | 2012-06-26 16:12:27 | Re: [PATCH] lock_timeout and common SIGALRM framework |
| Previous Message | Robert Haas | 2012-06-26 16:05:29 | Re: empty backup_label |