From: | Jeff Hoffmann <jeff(at)propertykey(dot)com> |
---|---|
To: | Tatsuo Ishii <t-ishii(at)sra(dot)co(dot)jp> |
Cc: | oleg(at)sai(dot)msu(dot)su, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: pgpool question |
Date: | 2005-03-10 03:14:46 |
Message-ID: | 47f7c1a9504d7b3f5188bc8fa01353ed@propertykey.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Mar 9, 2005, at 7:25 PM, Tatsuo Ishii wrote:
>
> That's an intended behavior. Or at least a side effect of failover
> design. If we allow unlimited switching between the master and the
> secondary, pgpool could repeat switching forever if we have unliable
> network or hardware.
I didn't really think of it that way, I had just expected it to toggle
back and forth for some reason. At first I thought it was just me, but
apparently Oleg got the same impression as I did. After you explained
it, though, it makes sense why someone would want it to work that way.
> However it would be easy to modify pgpool to allow automatic switch
> back (with a risk of unwanted repeating switching, of course). Is
> this what you want?
How about making it a switch at run-time? Like "--cycle" for the
automatic fail-over toggling. It seems that there are valid reasons
for both options. What makes the most sense to me would be to make the
"-s" switch always be able to switch to the server specified in the
command line or toggle between the two if you don't specify either
master or secondary. That way an administrator can always have control
or which server is being used & then either leave the automatic
behavior as is or create a cycle switch in case the user preferred that
behavior.
--
Jeff Hoffmann
jeff(at)propertykey(dot)com
From | Date | Subject | |
---|---|---|---|
Next Message | Kevin Brown | 2005-03-10 03:30:08 | Re: We are not following the spec for HAVING without GROUP BY |
Previous Message | Tom Lane | 2005-03-10 02:21:36 | We are not following the spec for HAVING without GROUP BY |