| From: | Bruce Momjian <bruce(at)momjian(dot)us> |
|---|---|
| To: | Yaroslav Tykhiy <yar(at)barnet(dot)com(dot)au> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Warm standby: 1 to N |
| Date: | 2009-06-02 18:52:26 |
| Message-ID: | 200906021852.n52IqQV21383@momjian.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Yaroslav Tykhiy wrote:
> Hi All,
>
> Let's consider the following case: WAL segments from a master have
> been shipped to N warm standby servers, and now the master fails.
> Using this or that mechanism, one of the warm standbys takes over and
> becomes the new master. Now the question is what to do with the other
> N-1 warm standbys. By the failure, all N warm standbys were the same
> exact copies of the master. So at least in theory, the N-1 warm
> standbys left can be fed with WAL segments from the new master. Do
> you think it will work in practice? Are there any pitfalls?
I think it should work.
--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com
+ If your life is a hard drive, Christ can be your backup. +
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Louis Lam | 2009-06-02 19:13:15 | catalog view use to find DATABASE, LANGUAGE, TABLESPACE, SCHEMA, SEQUENCE privileges granted to user or role |
| Previous Message | Tom Lane | 2009-06-02 18:47:55 | Re: Change view definition - do not have to drop it |