apcupsd & Fedora 14

[Jorge Fábregas]

On 01/23/2011 09:52 AM, jjletho67-arodef at yahoo.it wrote:
> Killing the power after the halt script has umounted and remounted read only the 
> root filesystem it's not enough to garuantee the safety of raid volume which are 
> signaled to close by the kernel at the very end of the shutdown process. Killing 
> the power before this could leave raid volume in an unclean status. 

Hi,

I didn't know about this issue with raid volumes.  Is it documented
somewhere (on bugzilla)?  My current UPS (and the previous one) will cut
the power in about 25 seconds after it recieves the killpower signal. I
think most of them will do this (about 15 to 20 seconds).   And on my
system, the elapsed time between when killpower is called and the
machine shuts off is about 4 seconds so I have about 21 extra seconds
there (enough time for any raid to finish I think).


> In the manual you will find very clear explanations on this.
> 
> http://www.apcupsd.com/manual/manual.html#the-shutdown-sequence-and-its-discontents

Thanks. I replaced the code last night with the same suggestion on that
link and it's working right.  I asked here because it's been a while
since I looked at the halt script and thought that perhaps, there was a
new logic there (ups-monitoring-program agnostic logic).

Best regards,
Jorge