[pgpool-general: 5755] Re: "Return" node automatically

[Giancarlo Celli]

Thanks a lot!
Can I know what pcp_recovery_node does exactly?
How does it perform recovery?
It not runs recovery_1st_stage.sh....so what does it do?
Thanks a lot.
 Best regards!

Giancarlo Celli


Il 28 Set 2017 02:33, "Bo Peng" <pengbo at sraoss.co.jp> ha scritto:

> Hi,
>
> If you need recovery the backend node first,
> you need to use "pcp_recovery_node".
>
> If the backend node is already started as a standby,
> you can use "pcp_attach_node"to let Pgpool-II know the node.
>
> Therefore you can just add "pcp_attach_node" in your origin script.
>
> On Mon, 25 Sep 2017 15:38:00 +0200
> Giancarlo Celli <giancarlocelli at gmail.com> wrote:
>
> > Hi,
> >
> > I've  used "pcp_recovery_node" to recovery the node as standby in my
> > follow_master script and it's working now!
> > I've a question for you: why use pcp_recovery_node and not
> pcp_attach_node?
> > What is the difference?
> > Thanks a lot!
> > Best regards.
> >
> > Giancarlo
> >
> > 2017-09-22 6:47 GMT+02:00 Bo Peng <pengbo at sraoss.co.jp>:
> >
> > > Hi
> > >
> > > I confirmed your follow_master script.
> > >
> > > I think you should use "pcp_recovery_node" command to
> > > recovery the backend as standby.
> > >
> > > I didn't see your initiate_replication.sh,
> > > but I think because the backend node is already detached
> > > while you use "initiate_replication.sh" to recovery the backend node.
> > >
> > > Even if the backend is started after recovery, Pgpool-II
> > > can not know the node. That's the reason why you need to do "Return"
> > > by pgpoolAdmin.
> > >
> > > Try to use "pcp_recovery_node" to recovery the node as standby in your
> > > follow_master
> > > script.
> > >
> > > I attached my follow_master script.
> > >
> > > On Thu, 14 Sep 2017 17:55:37 +0200
> > > Giancarlo Celli <giancarlocelli at gmail.com> wrote:
> > >
> > > > Hi,
> > > >
> > > > thanks for your help.
> > > > I'm sending you pgpool.conf file and follow_master script.
> > > > Let me know, please.
> > > > Best regards.
> > > > Giancarlo
> > > >
> > > >
> > > > 2017-09-14 14:52 GMT+02:00 Bo Peng <pengbo at sraoss.co.jp>:
> > > >
> > > > > Hi,
> > > > >
> > > > > By me the "follow_master_command" works.
> > > > > After failover, "follow_master_command" called pcp_recovery_node
> > > > > command to recovery the slave node to the new primary node, and
> > > > > "Return" is not needed.
> > > > >
> > > > > Can I have your pgpool.conf and follow_master script configured
> > > > > in the follow_master_command?
> > > > >
> > > > > On Mon, 11 Sep 2017 11:38:40 +0200
> > > > > Giancarlo Celli <giancarlocelli at gmail.com> wrote:
> > > > >
> > > > > > Hi to all.
> > > > > >
> > > > > > I would like to know how to do a "Return" node just gets
> detected by
> > > > > pgpool.
> > > > > > Specifically, I've created a follow_master_command that, after a
> > > > > failover,
> > > > > > performs recovery of all standby nodes to hook them to the new
> > > primary
> > > > > > master. Everything works, but then I have to click "return" on
> > > > > pgpooladmin
> > > > > > to complete the procedure. I need to make the "return"
> automatically.
> > > > > >
> > > > > > Can you help me, please?
> > > > > > Attached is the image of the pgpooladmin who asks for the Return
> > > > >
> > > > >
> > > > > --
> > > > > Bo Peng <pengbo at sraoss.co.jp>
> > > > > SRA OSS, Inc. Japan
> > > > >
> > > > >
> > >
> > >
> > > --
> > > Bo Peng <pengbo at sraoss.co.jp>
> > > SRA OSS, Inc. Japan
> > >
>
>
> --
> Bo Peng <pengbo at sraoss.co.jp>
> SRA OSS, Inc. Japan
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.sraoss.jp/pipermail/pgpool-general/attachments/20170928/c1fd7dc6/attachment.html>