[Nut-upsuser] Configuration tunning
antoine.gatineau at alcatel-lucent.com
Fri Nov 6 13:09:04 UTC 2009
Thanks for the quick reply Arjen!
Of course I will tune the pollinterval cleverly for my system to shutdown
safely taking the worst margin * 3.
You say that the is no relationship between pollinterval and other
parameters. Does that mean that if I configure pollinterval=60 and
POLLFREQ=30 there will be no error in upsmon?
nut-upsuser-bounces+antoine.gatineau=alcatel-lucent.com at lists.alioth.debian.
[mailto:nut-upsuser-bounces+antoine.gatineau=alcatel-lucent.com at lists.alioth
.debian.org] De la part de Arjen de Korte
Envoyé : vendredi 6 novembre 2009 12:22
À : nut-upsuser at lists.alioth.debian.org
Objet : Re: [Nut-upsuser] Configuration tunning
Citeren Antoine Gatineau <antoine.gatineau at alcatel-lucent.com>:
> Is there a maximum value for pollinterval in ups.conf? Is it possible
> that a high value conflicts with other linux components (usb timeout for
> Is there a relation between pollinterval and MAXAGE (in upsd.conf),
> and parameters in upsmon.conf
No (this used to be in ancient NUT versions, but this has long gone).
In almost all drivers, the pollinterval parameter determines how often the
UPS status is updated. If you increase the time between polls by increasing
pollinterval, the latency for updates will increase. You may miss critical
events on the UPS and worse, may be informed about the low battery status
too late (if at all).
Whether or not this affects your systems depends on the remaining runtime on
battery when the low battery alarm is raised. You might increase the
pollinterval to the remaining runtime when the battery low warning is raised
by the UPS minus the time it takes to shutdown the systems that are attached
to it. To be on the safe side, you should divide this by three to allow for
a couple of failed status polls.
Best regards, Arjen
Please keep list traffic on the list
Nut-upsuser mailing list
Nut-upsuser at lists.alioth.debian.org
More information about the Nut-upsuser