CentOS - Jan 2006 - *tangent* High Availability using 2 sites -- yep, "propogation."

Hello Les,

Thanks for that info.  I'm playing with this now and although the
'failover' process seems rather slow, it does seem to be doing what I
need. I setup a subdomain entry to point to 4 IP's, only one if which
is actually working, and indeed, when IE get's a non-active IP, it
eventually goes to the next one until it finally finds the actual live
IP.  Once it gets the live one, the site works as expected (of
course).

I'm glad I stumbled onto this thread... for my client's current need,
this will do the trick nicely.

-- 
Best regards,
 Mickael
            mailto:mikelists at silverservers.com

Thursday, January 5, 2006, 12:05:19 PM, you wrote:
> On Thu, 2006-01-05 at 13:33, Mickael Maddison wrote:
>> Forgive me for this sidestep, but you are saying that Windows/IE
>> actually ignores bad IP addresses if a site lists multiple IP's in
a
>> DNS lookup?  I tried this approach for some redundancy a couple years
>> ago and it didn't seem to work as you suggest.  If it has indeed
>> changed to work that way, this will help one of my clients immensely.
> The versions of IE I've tried have been very good at this although
> it doesn't work for other stock windows apps. You may need
> to make sure that ICMP 'unreachable' packets are not firewalled
> so down hosts are detected quickly.   You can test it easily by
> setting up a dummy DNS name with a bunch of A records where some
> point to working servers and some don't.

On Thu, 2006-01-05 at 16:58, Mickael Maddison wrote:
> Hello Les,
> 
> Thanks for that info.  I'm playing with this now and although the
> 'failover' process seems rather slow, it does seem to be doing what
I
> need. I setup a subdomain entry to point to 4 IP's, only one if which
> is actually working, and indeed, when IE get's a non-active IP, it
> eventually goes to the next one until it finally finds the actual live
> IP.  Once it gets the live one, the site works as expected (of
> course).
> 
> I'm glad I stumbled onto this thread... for my client's current
need,
> this will do the trick nicely.
If you are getting the standard ICMP responses for failure it
should find the working address pretty quickly. You can test
that by (from a linux box):
telnet ip_address 80
You should get a 'connection refused' error quickly on a target
that is up but does not have a web server running, or a 'no route
to host' if the machine is completely down at that address.  If
you just have a long delay waiting for the connection you have
a firewall dropping the ICMPs - which may or may not be under
your control.

If you know a location is going to be down it will help to
remove that address from DNS but as the other comments in this
thread point out, you can't count on the changes to reach the
app quickly.

--
  Les Mikesell
   lesmikesell at gmail.com

Mickael Maddison <centos at silverservers.com>
wrote:
> Hello Les,
> Thanks for that info.  I'm playing with this now and
> although the 'failover' process seems rather slow, it does
> seem to be doing what I need. I setup a subdomain entry to
> point to 4 IP's, only one if which is actually working, and
> indeed, when IE get's a non-active IP, it eventually goes
to 
> the next one until it finally finds the actual live
> IP.  Once it gets the live one, the site works as expected
> (of course).
> I'm glad I stumbled onto this thread... for my client's
> current need, this will do the trick nicely.
For a corporate network, this does the job.  You've got
access to the authority, and you're working at wire-speeds.

For the Internet, it doesn't work nearly as well.  The logic
is going to be far more arbitrary based on provider services
between the end user and the end server.

That has been my continued viewpoint.  Again, I've written
some rather interesting GPOs for the Windows resolver and MS
IE to deal with this.



-- 
Bryan J. Smith     Professional, Technical Annoyance                     
b.j.smith at ieee.org      http://thebs413.blogspot.com
----------------------------------------------------
*** Speed doesn't kill, difference in speed does ***