On 10/3/07, Rogelio Serrano <[EMAIL PROTECTED]> wrote: .. > The key to fast failovers then is fast detection of dead nodes.
Yes! and this is hard to do reliably with TCP due to all the retries and what-not. That's why LVS wants to use a serial cable or a raw Ethernet for sideband signaling the heartbeat. The other half of reliable failovers is efficient FENCING. Linux-HA's solution to this.. leaves much to be desired (Shoot the other node in the head). _________________________________________________ Philippine Linux Users' Group (PLUG) Mailing List [email protected] (#PLUG @ irc.free.net.ph) Read the Guidelines: http://linux.org.ph/lists Searchable Archives: http://archives.free.net.ph
