-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Joshua Penix wrote:
> This is exactly how VMotion works as well.  It generally uses a  
> separate gigabit network between the nodes.

Cool. We use a separate gigabit network between nodes for domain
migrations also.

> If you just go ripping out cables, you'll fall back to VMware's HA  
> functionality which will *notice* that the cables have been ripped  
> and then bring any dead virtual machines back up on other hardware,  
> but the state obviously isn't preserved since it just evaporated from  
> the RAM of the machine with no cables. :)

Gotcha. This is what I am implementing also. Planning on using heartbeat
etc to monitor nodes and bring up the domains that were on them when
they died if heartbeat detects a fault.


- --
Tracy R Reed
http://ultraviolet.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iD8DBQFFHeod9PIYKZYVAq0RApxYAJ9T8LfCdJvWPMHPF7guvtzVgzvD/gCfSVLi
W86nug61xuSXV/yWkjtiCzI=
=bvLB
-----END PGP SIGNATURE-----


-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to