On 10/9/06, Michael Richardson <[EMAIL PROTECTED]> wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
>>>>> "sacha" == sacha panasuik <[EMAIL PROTECTED]> writes:
sacha> to worry about integrating a PCI card into the mix. This
sacha> does mean that I am without conference capabilities (I
sacha> haven't managed to get ztdummy compiled and running happily
sacha> but I haven't tried too hard), but for my installation,
ztdummy won't work, I think.
It needs support from the Xen hypervisor.
I know how to make all of this work, but just haven't got the time to
do it... maybe in January.
sacha> One of my next projects is to see if I can integrate some
sacha> high availability into my installation, that is, have two
sacha> computers each running xen be able to keep the filesystems in
sacha> sync and take over a downed virtual machine in the case of a
sacha> failure (or for maintenance): since my asterisk installation
sacha> just needs an IP address, I should be able to bounce it
sacha> around the network without too much trouble.
You need to at least store your /etc/asterisk on some kind of
network store, or at least rsync it and voice mail regularly.
<SNIP>
Michael,
Thanks for your comments, I probably wasn't clear - I'd like to investigate
keeping the entire filesystem for the asterisk virtual machine in sync on
another physical server at the block level with something like DRDB (
http://www.drbd.org/). So that, in the event of a failover (either manually
initiated or with a heartbeat type software, such as heartbeat (
http://linux-ha.org/download)), the entire virtual machine can be started on
the alternate hardware and asterisk can resume doing what it does best. I'm
sure I'm not the best person to think about ways of introducing high
availability to asterisk - I'm sure there are existing implementations out
there - but this is just a research and development interest of mine.
--
sacha
