nut-debian/docs/man/upssched.txt

116 lines
3.7 KiB
Plaintext
Raw Normal View History

2011-01-26 11:35:08 +02:00
UPSSCHED(8)
===========
NAME
----
upssched - Timer helper for scheduling events from upsmon
SYNOPSIS
--------
*upssched*
NOTE: *upssched* should be run from linkman:upsmon[8] via the NOTIFYCMD.
2010-03-26 01:20:59 +02:00
You should never run it directly during normal operations.
2011-01-26 11:35:08 +02:00
DESCRIPTION
-----------
*upssched* was created to allow users to execute programs at times
relative to events being monitored by linkman:upsmon[8]. The original
2010-03-26 01:20:59 +02:00
purpose was to allow for a shutdown to occur after some fixed period
on battery, but there are other uses that are possible.
2011-01-26 11:35:08 +02:00
INTEGRATION
-----------
upssched needs to be called as the NOTIFYCMD in your linkman:upsmon.conf[5].
2010-03-26 01:20:59 +02:00
It determines what is happening based on the UPSNAME and NOTIFYTYPE
environment variables. You should never have to deal with them directly.
Set the EXEC flag on the events that you want to see in upssched.
For example, to make sure that upssched hears about ONLINE, ONBATT and
LOWBATT events, the flags would look like this:
NOTIFYFLAG ONLINE EXEC
NOTIFYFLAG ONBATT EXEC
NOTIFYFLAG LOWBATT EXEC
If you also want to continue writing to the syslog, just add it in:
NOTIFYFLAG ONLINE SYSLOG+EXEC
NOTIFYFLAG ONBATT SYSLOG+EXEC
NOTIFYFLAG LOWBATT SYSLOG+EXEC
2011-01-26 11:35:08 +02:00
For a full list of notify flags, see the linkman:upsmon[8] documentation.
2010-03-26 01:20:59 +02:00
2011-01-26 11:35:08 +02:00
CONFIGURATION
-------------
2010-03-26 01:20:59 +02:00
2011-01-26 11:35:08 +02:00
See linkman:upssched.conf[5] for information on configuring this program.
2010-03-26 01:20:59 +02:00
2011-01-26 11:35:08 +02:00
EARLY SHUTDOWNS
---------------
2010-03-26 01:20:59 +02:00
To shut down the system early, define a timer that starts due to an ONBATT
condition. When it triggers, make your CMDSCRIPT call your shutdown
2011-01-26 11:35:08 +02:00
routine. It should finish by calling `upsmon -c fsd` so that upsmon gets
2010-03-26 01:20:59 +02:00
to shut down the slaves in a controlled manner.
Be sure you cancel the timer if power returns (ONLINE).
2011-01-26 11:35:08 +02:00
DEBOUNCING EVENTS
-----------------
2010-03-26 01:20:59 +02:00
If your UPS goes on and off battery frequently, you can use this program
to reduce the number of pager messages that are sent out. Rather than
2011-01-26 11:35:08 +02:00
sending pages directly from linkman:upsmon[8], use a short timer here.
2010-03-26 01:20:59 +02:00
If the timer triggers with the UPS still on battery, then send the page.
If the power returns before then, the timer can be cancelled and no page
is necessary.
2011-01-26 11:35:08 +02:00
BACKGROUND
----------
2010-03-26 01:20:59 +02:00
This program was written primarily to fulfill the requests of users for
the early shutdown scenario. The "outboard" design of the program
(relative to upsmon) was intended to reduce the load on the average
2011-01-26 11:35:08 +02:00
system. Most people don't have the requirement of shutting down after 'N'
2010-03-26 01:20:59 +02:00
seconds on battery, since the usual OB+LB testing is sufficient.
This program was created separately so those people don't have to spend
CPU time and RAM on something that will never be used in their
environments.
The design of the timer handler is also geared towards minimizing impact.
It will come and go from the process list as necessary. When a new timer
is started, a process will be forked to actually watch the clock and
eventually start the CMDSCRIPT. When a timer triggers, it is removed from
the queue. Cancelling a timer will also remove it from the queue. When
no timers are present in the queue, the background process exits.
This means that you will only see upssched running when one of two things
is happening:
2011-01-26 11:35:08 +02:00
- There's a timer of some sort currently running
- upsmon just called it, and you managed to catch the brief instance
2010-03-26 01:20:59 +02:00
The final optimization handles the possibility of trying to cancel a timer
when there are none running. If the timer daemon isn't running, there
are no timers to cancel, and furthermore there is no need to start
2011-01-26 11:35:08 +02:00
a clock-watcher. So, it skips that step and exits sooner.
FILES
-----
linkman:upssched.conf[5]
2010-03-26 01:20:59 +02:00
2011-01-26 11:35:08 +02:00
SEE ALSO
--------
2010-03-26 01:20:59 +02:00
2011-01-26 11:35:08 +02:00
linkman:upsmon[8]
2010-03-26 01:20:59 +02:00
2011-01-26 11:35:08 +02:00
Internet resources:
~~~~~~~~~~~~~~~~~~~
2010-03-26 01:20:59 +02:00
The NUT (Network UPS Tools) home page: http://www.networkupstools.org/