Nut upsuser - Jul 2016 - Best practice to shutdown hosts which has not NUT via upssched

Thank you for answer Stuart
>?????, 13 ???? 2016, 17:54 +03:00 ?? Stuart Gathman <stuart at
gathman.org>:
>
>On 07/13/2016 10:02 AM, Stuart Gathman wrote:
>> On 07/13/2016 04:17 AM, Dmitri Stepanov wrote:
>>>    I need to shutdown a number of hosts which has not NUT from one
which has it.
>> Let's make sure you are not missing an obvious point.  By "has
not NUT",
>> you might mean not running upsd.  The best practice is to run upsmon on
>> the other hosts.  This does not require running NUT proper (server). 
Is
>> there some reason why you can't run upsmon on the other hosts? 
>> Typically, upsd and drivers are packaged as something like
"nut-server",
>> whereas upsmon is packaged as "nut-client".
>If you truly can't run upsmon, then you can speak the upsd protocol
>directly over tcp.  For an example non-nut implementation, look at the
>"check_ups" plugin for nagios.  It queries a NUT server for basic
>information, including whether it is on battery.
>
Some of those hosts haven't NUT including upsmon (and other software which
is not included by suppliers of the hosts) for a number of reasons. I don't
need to inform not-NUT hosts about any UPS events, etc... but only shutdown all
the hosts after upssched timer expired. So I think shutdown them via ssh would
be enough and creating some replacement for upsmon is too much.
>_______________________________________________
>Nut-upsuser mailing list
>Nut-upsuser at lists.alioth.debian.org
>http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/nut-upsuser

On Thu, 14 Jul 2016, Dmitri Stepanov wrote:
> Some of those hosts haven't NUT including upsmon (and other software 
> which is not included by suppliers of the hosts) for a number of 
> reasons. I don't need to inform not-NUT hosts about any UPS events, 
> etc... but only shutdown all the hosts after upssched timer expired. So 
> I think shutdown them via ssh would be enough and creating some 
> replacement for upsmon is too much.
Perhaps it is possible to have a micro upsmon in the form of a cron job in 
every slave which, every minute, runs something like

  rprice at maria:~> X=$( echo "GET VAR my-ups battery.charge" |
netcat -w1 upsd-server 3493 )
  rprice at maria:~> if [[ "$X" =~ \"([0-9]*)\" ]] ; then
Y=${BASH_REMATCH[1]}; else Y=999 ; fi
  rprice at maria:~> if [[ "$Y" -gt 35 ]] ; then echo "Carry
on $Y" ; else echo "Shutdown $Y" ; fi
  Carry on 100

Different slaves could have different low battery values. I'm sure that a 
netcat expert would be able to find a way of executing netcat only once.

Roger

>???????, 14 ???? 2016, 23:58 +03:00 ?? Roger Price < roger at
rogerprice.org >:
>
>On Thu, 14 Jul 2016, Dmitri Stepanov wrote:
>
>> Some of those hosts haven't NUT including upsmon (and other
software
>> which is not included by suppliers of the hosts) for a number of 
>> reasons. I don't need to inform not-NUT hosts about any UPS events,
>> etc... but only shutdown all the hosts after upssched timer expired. So
>> I think shutdown them via ssh would be enough and creating some 
>> replacement for upsmon is too much.
>
>Perhaps it is possible to have a micro upsmon in the form of a cron job in 
>every slave which, every minute, runs something like
>
>??rprice at maria:~> X=$( echo "GET VAR my-ups battery.charge"
| netcat -w1 upsd-server 3493 )
>??rprice at maria:~> if [[ "$X" =~ \"([0-9]*)\" ]] ;
then Y=${BASH_REMATCH[1]}; else Y=999 ; fi
>??rprice at maria:~> if [[ "$Y" -gt 35 ]] ; then echo
"Carry on $Y" ; else echo "Shutdown $Y" ; fi
>??Carry on 100
>
>Different slaves could have different low battery values. I'm sure that
a
>netcat expert would be able to find a way of executing netcat only once.
Cool. I've never think via this way. I'm sure I'll use it. Thank you
for the hint.
But a problem could be with netcat software (if it installed or not on the
not-NUT slaves, moreover there are a number of Windows hosts). Unfortunately I
have no ability to break in configuration and installed software of the not-NUT
hosts too deeply (licenses of the suppliers of the hosts, etc...).

Thanks to Kris Jordan, I guess the problem is the uid of the upssched (CMDSCRIPT
and my shutdown script and ssh) process. I hope I 'd check it as soon as I
get connection to the system.
>
>Roger
>
>_______________________________________________
>Nut-upsuser mailing list
> Nut-upsuser at lists.alioth.debian.org
> http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/nut-upsuser
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20160715/e3100c60/attachment.html>