Josh Luthman wrote:
I'm not entire sure what you mean when you reference the failover capability. Could you please explain how this works?
I'm basically just a user of the capability, how much detail do you want? It's just the standard failover built into bb.
I'm interested in knowing how the hostname reflects to what IP addresses, hardware running what software specifically, etc.
How the hostname reflects to what IP address? I'm not sure what you mean. There are no tricks here, just the standard dns naming scheme.
I'm not sure what hardware has to do with it, but we're running SLES on HP/Compaq DL servers.
I'm not sure what you mean by "what software" - you mean the OS, or the applications being monitored, or the exact version of bb?
Coming from BB1.9btf I don't know of many expansions between 1.9 and 3.3.
We're on 1.9 here, patched with bbgen to keep it going - AFAIK the bb code has basically languished since it was bought back in 2001 or so, so I'm curious about this version 3.3 that you speak of.
We had some discussion about multiple servers and redundancy just a short while ago: http://www.hswn.dk/hobbiton/2007/10/msg00423.html
Yes, those discussions look mostly like the typical ha requirements, e.g. managing bb failover via external proxies, redirectors etc, which adds a whole new layer of cost and complexity. It would be a hard sell to justify all the new ha infrastructure if we are replacing bb with something newer and better, since bb currently handles that all by itself, with no need of an external ha system.
Joe