On Nov 27, 2007 12:56 PM, Carlos Rodrigues [EMAIL PROTECTED] wrote:
1. most medium/large datacenters have a small number of large UPSes
with generator backups, so NUT is rather pointless;
2. the datacenter has a big UPS and no generator backup, and people
find their 60 minute runtime enough
On Nov 27, 2007 6:01 AM, Carlos Rodrigues [EMAIL PROTECTED] wrote:
Sorry about the change in the subject, I was goink to ask how hard
would it be to implement some kind of auto-discovery/centralized
configuration for slave upsmon's but after thinking a bit I though it
would be too much work
A big issue with auto-discovery is trust. You wouldn't want someone to
be able to randomly shut down machines on your network just by
starting a rogue copy of NUT.
Another problem is when you discover multiple UPS'es, which one you're
connected to? In a desktop system with only one UPS
On Nov 27, 2007 3:59 PM, Arjen de Korte [EMAIL PROTECTED] wrote:
When you have only one UPS in your network, this might work. Otherwise,
one UPS going offline might cause all clients to shutdown. This is
probably not very desireable.
What I had in mind was more of a centrally managed
Carlos Rodrigues wrote:
When you have only one UPS in your network, this might work. Otherwise,
one UPS going offline might cause all clients to shutdown. This is
probably not very desireable.
What I had in mind was more of a centrally managed configuration, than
no configuration. For
5 matches
Mail list logo
