betty_sun0 <ambetty at sina.com> writes:
> and after submit : Slony1-general Subscription results Your
> subscription request has been received, and will soon be acted
> upon. Depending on the configuration of this mailing list, your
> subscription request may have to be first confirmed by you via
> email, or approved by the list moderator. If confirmation is
> required, you will soon get a confirmation email which contains
> further instructions.
>
> but up to the present,I have not receive confirm email.

As it turns out...

 a) No further confirmation is required;
 b) You won't be getting a confirmation email;
 c) You ARE on the mailing list, so what you wanted has been
    accomplished; nothing more is needed.

A suggestion: Please trim out the old material in the email that isn't
relevant to your present message.  Your email contains copies of
replies to replies to replies, and is getting to look like a big long
chain letter, which is a bit annoying.

> I meet new question and this question is urgent,if you know please tell me.Thank you.
>
> I read slony1-1.0.5\doc\howto\slony-I-failover.txt more than three times over.
> but I not understand Failover chapter .
> question1: 
> when time should use Failover chapter's slonik command and step?

There are two slonik commands in that chapter; it isn't clear which
one you intend.

  1.  There's a MOVE SET sample, used if you want to cleanly move the
      origin from one node to another.

  2.  There's a FAILOVER sample, used if the origin node has broken so
      that you cannot use MOVE SET to move the origin to the other node.

      A later step in that process is to DROP NODE on the failed node.

You use "MOVE SET" if you want to request that the origin be moved
from one node to another, and everything is running well.

You use "FAILOVER" if the origin node has failed, due to something
like a hardware failure on that database server.

> question2: how to stop slon process (start with : slon golf
> "dbname=db2 user=posta host=10.10.11.12 port=9543" &).  except for
> use ps command find out pid and then kill -9 pid command. whether or
> not have slon command carry out stop function.

There is no "slon command" to carry out this sort of function.  Nor is
there a slonik command to carry out this sort of function.

In version 1.1, there is an option to write the PID out to a file; if
you do that, then you can look in the PID file for the PID, and kill
that process.  That may be more suitable than searching for processes
using ps.

> question3: if had exec (slonik command : drop node (id = 1, event
> node = 2);) at node1. slon process was stop and cluster was
> delete,now at node1 start slon process (slonik command : slon golf
> "dbname=db2 user=posta host=10.10.11.12 port=9543" &)
> show error info :  
> CONFIG main: slon version 1.0.5 starting up
> ERROR  cannot get sl_local_node_id - ERROR:  schema "_golf" does not exist
> FATAL  main: Node is not initialized properly
>
> how to recover node1's cluster ? because I only set up two node slon
> replicate(one Master another is Slave).

There is no way to "recover" the cluster after it is deleted.

You must recreate the node from scratch, via the slonik command STORE
NODE.
-- 
"cbbrowne","@","ca.afilias.info"
<http://dev6.int.libertyrms.com/>
Christopher Browne
(416) 673-4124 (land)