data loss after power failure

[Micha Silver]

shimi wrote:
>
>
> On Sat, Aug 15, 2009 at 7:26 AM, Amos Shapira <amos.shapira at gmail.com 
> <mailto:amos.shapira at gmail.com>> wrote:
>
>     2009/8/14 Micha Silver <micha at arava.co.il <mailto:micha at arava.co.il>>:
>     > shimi wrote:
>
>
>     Well, that's how UPS's are supposed to be used.
>
>     In addition to Wake-On-Lan mentioned by Shimi, there are also:
>     1. Boot on AC power - i.e. if the AC power comes back then start
>     reboot. That's a BIOS setting.
>
>
> Which assumes a power loss after the automatic shutdown. According to 
> Murphy law, and to my experience, power tends to return 1 minute 
> before the battery gets completely drained, just for me to wonder "why 
> oh why did I shut down my whole server room, it would have survived". 
> Of course that if I wouldn't have shut down everything, the outage 
> would be 1 minute longer. :)
That's exactly my experience. Murphy's law is more reliable than gravity...
>
> So still need WoL :) (to be initiated by the machine sampling for the 
> UPS remaining battery after the battery managed to re-charge for 
> additional few minutes of backup power - enough for a second power up 
> and shutdown of everything assuming a second outage...)
We'll try to go for a solution like this.
Thanks to both Amos and Shimi for the suggestions.

>
> -- Shimi
>