×
The internal search function is temporarily non-functional. The current search engine is no longer viable and we are researching alternatives.
As a stop gap measure, we are using Google's custom search engine service.
If you know of an easy to use, open source, search engine ... please contact support@midrange.com.
Hi Rob -
On Fri, 27 Jul 2012 15:54:20 -0400, rob@xxxxxxxxx wrote:
I also don't want to go from saying:
"We only have one sametime server. If that is down we're toast."
to saying:
"We only have one intelligent IP sprayer. If that is down we're toast."
IOW, I don't want to just trickle up the single point of failure.
One thought is round-robin DNS. Of course this requires both the DNS
server and the client to be coded appropriately. It also requires
that each server have it's own internet-routable IP address. (The IP
address selection is made by the client in this scenario.)
Assuming that the DNS servers can handle it, you put in multiple A
records for the same host name with different IP addresses. These
different IP addresses don't have to have any relationship to each
other, so they can be for completely different locations, thereby
helping to eliminate your single-point-of-failure. (And of course you
should already have multiple DNS servers at different locations.)
When a request is made for that host name, the DNS server returns all
of the IP addresses. It rotates the order (hence "round-robin") which
provides some crude load balancing.
If the client is coded well, if it can't connect to the port with the
first IP address, it will try the second and so forth.
Ken
Opinions expressed are my own and do not necessarily represent the views
of my employer or anyone in their right mind.
midrange.com
