thr3ads.net - Nut upsuser - [Nut-upsuser] Email Alerts for Multiple UPSs with upssched [Oct 2017]

If this information is useful, please help other people find it:
Share via:

Garrett Michael Hayes

2017-Oct-31 15:09 UTC

[Nut-upsuser] Email Alerts for Multiple UPSs with upssched

Greetings Gentlebeings,

I have started deploying a number of NUT instances on Raspberry Pi machines
(Raspbian Jessie) across my network. The purpose of these "nutcases"
as we refer to them is to monitor the status of the UPSs supporting our network
switching infrastructure and notify us of power and battery events via email.
No shutdown or other actions are required.

Everything is going spiffy as all get-out so far. But now I'm running up
against a configuration issue I don't quite know how to address.

Thus far each "nutcase" has been configured to monitor a single UPS.
Now I want to deploy one to monitor three UPSs in a larger switch closet.
Adding the additional UPSs to ups.conf and upsmon.conf is fairly
straightforward. But I'm darned if I can figure out how to configure the
email alerts to let me know WHICH UPS is the source of an event.

I'm using upssched as my NOTIFYCMD, and it in turn is calling a little shell
script that reads the flag that's passed and sends an email. What I
can't figure out is how (if there is a way) to determine the originating
UPS. It seems that only the status flag makes it across to the shell script.
Or at least if it's passing more than that, I don't know how to retrieve
it.

I'm sure I must be missing something simple. Any thoughts/pointers would be
appreciated!

Garrett Michael Hayes

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20171031/7134c5d9/attachment.html>

O'Shaughnessy, Mike

2017-Oct-31 16:45 UTC

head link

[Nut-upsuser] Email Alerts for Multiple UPSs with upssched

I'm running a single "nut-server" that monitors seven ups units.

In upsmon.conf:
MONITOR ups-1   0 <User> <Pw> master
MONITOR ups-2   0 <User> <Pw> master
MONITOR ups-3   0 <User> <Pw> master
MONITOR ups-4   0 <User> <Pw> master
MONITOR ups-5   1 <User> <Pw> master
MONITOR ups-6   1 <User> <Pw> master
MONITOR ups-7   0 <User> <Pw> master

NOTIFYCMD /sbin/upssched

NOTIFYFLAG ONLINE       SYSLOG+WALL+EXEC
NOTIFYFLAG ONBATT       SYSLOG+WALL+EXEC
NOTIFYFLAG LOWBATT      SYSLOG+WALL+EXEC
NOTIFYFLAG FSD          SYSLOG+WALL+EXEC
NOTIFYFLAG COMMOK       SYSLOG+WALL+EXEC
NOTIFYFLAG COMMBAD      SYSLOG+WALL+EXEC
NOTIFYFLAG SHUTDOWN     SYSLOG+WALL+EXEC
NOTIFYFLAG REPLBATT     SYSLOG+WALL+EXEC
NOTIFYFLAG NOCOMM       SYSLOG+WALL+EXEC
NOTIFYFLAG NOPARENT     SYSLOG+WALL+EXEC

In upssched.conf
CMDSCRIPT /bin/upssched-cmd

AT LOWBATT ups-1        EXECUTE ups-1-low
AT LOWBATT ups-2        EXECUTE ups-2-low
AT LOWBATT ups-3        EXECUTE ups-3-low
AT LOWBATT ups-4        EXECUTE ups-4-low
AT LOWBATT ups-5        EXECUTE ups-5-low
AT LOWBATT ups-6        EXECUTE ups-6-low
AT LOWBATT ups-7        EXECUTE ups-7-low

Upssched-cmd:

echo "upssched-cmd entry was $1"

case $1 in
        ups-1-low)
                /ws/tools/shutdownUps-1
                logger -t upssched-cmd "UPS 1 low battery"
                ;;
        ups-2-low)
                /ws/tools/shutdownUps-2
                logger -t upssched-cmd "UPS 2 low battery"
                ;;
      ups-3-low)
                /ws/tools/shutdownUps-3
                logger -t upssched-cmd "UPS 3 low battery"
                ;;
        ups-4-low)
                /ws/tools/shutdownUps-4
                logger -t upssched-cmd "UPS 4 low battery"
                ;;
        ups-5-low)
                /ws/tools/shutdownUps-5
                logger -t upssched-cmd "UPS 5 low battery"
                ;;
        ups-6-low)
                /ws/tools/shutdownUps-6
                logger -t upssched-cmd "UPS 6 low battery"
                ;;
        ups-7-low)
                /ws/tools/shutdownUps-7
                logger -t upssched-cmd "UPS 7 low battery"
                ;;
Esac

Scripts like 'shutdownUPS-N' do this:

*         Foreach linux box on the ups, log in and issue "shutdown -h
now"

*         Foreach VMWare server on the ups, log in and issue
"/sbin/shutdown.sh && /sbin/poweroff"

Hope that helps!

Mike

From: Nut-upsuser [mailto:nut-upsuser-bounces+mike.oshaughnessy=gd-ms.com at
lists.alioth.debian.org] On Behalf Of Garrett Michael Hayes
Sent: Tuesday, October 31, 2017 11:10 AM
To: nut-upsuser at lists.alioth.debian.org
Subject: [Nut-upsuser] Email Alerts for Multiple UPSs with upssched

Greetings Gentlebeings,

I have started deploying a number of NUT instances on Raspberry Pi machines
(Raspbian Jessie) across my network.  The purpose of these "nutcases"
as we refer to them is to monitor the status of the UPSs supporting our network
switching infrastructure and notify us of power and battery events via email. 
No shutdown or other actions are required.

Everything is going spiffy as all get-out so far.  But now I'm running up
against a configuration issue I don't quite know how to address.

Thus far each "nutcase" has been configured to monitor a single UPS. 
Now I want to deploy one to monitor three UPSs in a larger switch closet. 
Adding the additional UPSs to ups.conf and upsmon.conf is fairly
straightforward.  But I'm darned if I can figure out how to configure the
email alerts to let me know WHICH UPS is the source of an event.

I'm using upssched as my NOTIFYCMD, and it in turn is calling a little shell
script that reads the flag that's passed and sends an email.  What I
can't figure out is how (if there is a way) to determine the originating
UPS.  It seems that only the status flag makes it across to the shell script. 
Or at least if it's passing more than that, I don't know how to retrieve
it.

I'm sure I must be missing something simple.  Any thoughts/pointers would be
appreciated!

Garrett Michael Hayes

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20171031/72dfb815/attachment-0001.html>

Garrett Michael Hayes

2017-Oct-31 17:18 UTC

head link

[Nut-upsuser] Email Alerts for Multiple UPSs with upssched

Ah!  So I can detect the source UPS in the upssched.conf and pass it as a
distinct event to the shell script.  That sounds like an excellent approach.

Thanks!

GMH

From: O'Shaughnessy, Mike [mailto:Mike.OShaughnessy at gd-ms.com]
Sent: Tuesday, October 31, 2017 12:46 PM
To: Garrett Michael Hayes <Garrett at VerbalImaging.com>; nut-upsuser at
lists.alioth.debian.org
Subject: RE: Email Alerts for Multiple UPSs with upssched

I'm running a single "nut-server" that monitors seven ups units.

In upsmon.conf:
MONITOR ups-1   0 <User> <Pw> master
MONITOR ups-2   0 <User> <Pw> master
MONITOR ups-3   0 <User> <Pw> master
MONITOR ups-4   0 <User> <Pw> master
MONITOR ups-5   1 <User> <Pw> master
MONITOR ups-6   1 <User> <Pw> master
MONITOR ups-7   0 <User> <Pw> master

NOTIFYCMD /sbin/upssched

NOTIFYFLAG ONLINE       SYSLOG+WALL+EXEC
NOTIFYFLAG ONBATT       SYSLOG+WALL+EXEC
NOTIFYFLAG LOWBATT      SYSLOG+WALL+EXEC
NOTIFYFLAG FSD          SYSLOG+WALL+EXEC
NOTIFYFLAG COMMOK       SYSLOG+WALL+EXEC
NOTIFYFLAG COMMBAD      SYSLOG+WALL+EXEC
NOTIFYFLAG SHUTDOWN     SYSLOG+WALL+EXEC
NOTIFYFLAG REPLBATT     SYSLOG+WALL+EXEC
NOTIFYFLAG NOCOMM       SYSLOG+WALL+EXEC
NOTIFYFLAG NOPARENT     SYSLOG+WALL+EXEC

In upssched.conf
CMDSCRIPT /bin/upssched-cmd

AT LOWBATT ups-1        EXECUTE ups-1-low
AT LOWBATT ups-2        EXECUTE ups-2-low
AT LOWBATT ups-3        EXECUTE ups-3-low
AT LOWBATT ups-4        EXECUTE ups-4-low
AT LOWBATT ups-5        EXECUTE ups-5-low
AT LOWBATT ups-6        EXECUTE ups-6-low
AT LOWBATT ups-7        EXECUTE ups-7-low

Upssched-cmd:

echo "upssched-cmd entry was $1"

case $1 in
        ups-1-low)
                /ws/tools/shutdownUps-1
                logger -t upssched-cmd "UPS 1 low battery"
                ;;
        ups-2-low)
                /ws/tools/shutdownUps-2
                logger -t upssched-cmd "UPS 2 low battery"
                ;;
      ups-3-low)
                /ws/tools/shutdownUps-3
                logger -t upssched-cmd "UPS 3 low battery"
                ;;
        ups-4-low)
                /ws/tools/shutdownUps-4
                logger -t upssched-cmd "UPS 4 low battery"
                ;;
        ups-5-low)
                /ws/tools/shutdownUps-5
                logger -t upssched-cmd "UPS 5 low battery"
                ;;
        ups-6-low)
                /ws/tools/shutdownUps-6
                logger -t upssched-cmd "UPS 6 low battery"
                ;;
        ups-7-low)
                /ws/tools/shutdownUps-7
                logger -t upssched-cmd "UPS 7 low battery"
                ;;
Esac

Scripts like 'shutdownUPS-N' do this:

*        Foreach linux box on the ups, log in and issue "shutdown -h
now"

*        Foreach VMWare server on the ups, log in and issue
"/sbin/shutdown.sh && /sbin/poweroff"

Hope that helps!

Mike

From: Nut-upsuser [mailto:nut-upsuser-bounces+mike.oshaughnessy=gd-ms.com at
lists.alioth.debian.org] On Behalf Of Garrett Michael Hayes
Sent: Tuesday, October 31, 2017 11:10 AM
To: nut-upsuser at lists.alioth.debian.org<mailto:nut-upsuser at
lists.alioth.debian.org>
Subject: [Nut-upsuser] Email Alerts for Multiple UPSs with upssched

Greetings Gentlebeings,

I have started deploying a number of NUT instances on Raspberry Pi machines
(Raspbian Jessie) across my network.  The purpose of these "nutcases"
as we refer to them is to monitor the status of the UPSs supporting our network
switching infrastructure and notify us of power and battery events via email. 
No shutdown or other actions are required.

Everything is going spiffy as all get-out so far.  But now I'm running up
against a configuration issue I don't quite know how to address.

Thus far each "nutcase" has been configured to monitor a single UPS. 
Now I want to deploy one to monitor three UPSs in a larger switch closet. 
Adding the additional UPSs to ups.conf and upsmon.conf is fairly
straightforward.  But I'm darned if I can figure out how to configure the
email alerts to let me know WHICH UPS is the source of an event.

I'm using upssched as my NOTIFYCMD, and it in turn is calling a little shell
script that reads the flag that's passed and sends an email.  What I
can't figure out is how (if there is a way) to determine the originating
UPS.  It seems that only the status flag makes it across to the shell script. 
Or at least if it's passing more than that, I don't know how to retrieve
it.

I'm sure I must be missing something simple.  Any thoughts/pointers would be
appreciated!

Garrett Michael Hayes

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20171031/4a208b28/attachment.html>

Jim Klimov

2017-Oct-31 17:36 UTC

head link

[Nut-upsuser] Email Alerts for Multiple UPSs with upssched

On October 31, 2017 4:09:46 PM GMT+01:00, Garrett Michael Hayes <Garrett at
VerbalImaging.com> wrote:>Greetings Gentlebeings,
>
>I have started deploying a number of NUT instances on Raspberry Pi
>machines (Raspbian Jessie) across my network.  The purpose of these
>"nutcases" as we refer to them is to monitor the status of the
UPSs
>supporting our network switching infrastructure and notify us of power
>and battery events via email.  No shutdown or other actions are
>required.
>
>Everything is going spiffy as all get-out so far.  But now I'm running
>up against a configuration issue I don't quite know how to address.
>
>Thus far each "nutcase" has been configured to monitor a single
UPS.
>Now I want to deploy one to monitor three UPSs in a larger switch
>closet.  Adding the additional UPSs to ups.conf and upsmon.conf is
>fairly straightforward.  But I'm darned if I can figure out how to
>configure the email alerts to let me know WHICH UPS is the source of an
>event.
>
>I'm using upssched as my NOTIFYCMD, and it in turn is calling a little
>shell script that reads the flag that's passed and sends an email. 
>What I can't figure out is how (if there is a way) to determine the
>originating UPS.  It seems that only the status flag makes it across to
>the shell script.  Or at least if it's passing more than that, I
don't
>know how to retrieve it.
>
>I'm sure I must be missing something simple.  Any thoughts/pointers
>would be appreciated!
>
>Garrett Michael Hayes
Try $UPSNAME (this and other vars are RTFMed somewhere, maybe comments to
configs) - I used it in NOTIFYCMD handling by scripts directly, maybe it would
pass through upssched too?

Also a script can call `set` to print out its env, as well as you can inspect
/proc/PID/environ on linux to get the envvars of a running process.

Jim
--
Typos courtesy of K-9 Mail on my Android

Garrett Michael Hayes

2017-Nov-01 15:34 UTC

head link

[Nut-upsuser] Email Alerts for Multiple UPSs with upssched

This was the secret sauce!  I first modified my BASH script to stuff the output
of the 'set' command into a variable, then passed that variable in to my
email command.  Then I triggered an event and saw what variables were there in
the body of the message.

Voila!

$UPSNAME and $NOTIFYTYPE were there and available.  I modified the script to
pump THOSE into the mail command, and the problem is (at least partially)
solved.

The next challenge is going to be sorting out which UPS is being seen on which
USB port.

Thanks everyone!!

GMH

-----Original Message-----
From: Jim Klimov [mailto:jimklimov at cos.ru] 
Sent: Tuesday, October 31, 2017 1:36 PM
To: nut-upsuser at lists.alioth.debian.org; Garrett Michael Hayes <Garrett at
VerbalImaging.com>; nut-upsuser at lists.alioth.debian.org
Subject: Re: [Nut-upsuser] Email Alerts for Multiple UPSs with upssched

On October 31, 2017 4:09:46 PM GMT+01:00, Garrett Michael Hayes <Garrett at
VerbalImaging.com> wrote:>Greetings Gentlebeings,
>
>I have started deploying a number of NUT instances on Raspberry Pi
>machines (Raspbian Jessie) across my network.  The purpose of these
>"nutcases" as we refer to them is to monitor the status of the
UPSs
>supporting our network switching infrastructure and notify us of power
>and battery events via email.  No shutdown or other actions are
>required.
>
>Everything is going spiffy as all get-out so far.  But now I'm running
>up against a configuration issue I don't quite know how to address.
>
>Thus far each "nutcase" has been configured to monitor a single
UPS.
>Now I want to deploy one to monitor three UPSs in a larger switch
>closet.  Adding the additional UPSs to ups.conf and upsmon.conf is
>fairly straightforward.  But I'm darned if I can figure out how to
>configure the email alerts to let me know WHICH UPS is the source of an
>event.
>
>I'm using upssched as my NOTIFYCMD, and it in turn is calling a little
>shell script that reads the flag that's passed and sends an email. 
>What I can't figure out is how (if there is a way) to determine the
>originating UPS.  It seems that only the status flag makes it across to
>the shell script.  Or at least if it's passing more than that, I
don't
>know how to retrieve it.
>
>I'm sure I must be missing something simple.  Any thoughts/pointers
>would be appreciated!
>
>Garrett Michael Hayes
Try $UPSNAME (this and other vars are RTFMed somewhere, maybe comments to
configs) - I used it in NOTIFYCMD handling by scripts directly, maybe it would
pass through upssched too?

Also a script can call `set` to print out its env, as well as you can inspect
/proc/PID/environ on linux to get the envvars of a running process.

Jim
--
Typos courtesy of K-9 Mail on my Android

Possibly Parallel Threads

Search for more possibly parallel threads

Nut upsuser - Oct 2017 - Email Alerts for Multiple UPSs with upssched

[Nut-upsuser] Email Alerts for Multiple UPSs with upssched

[Nut-upsuser] Email Alerts for Multiple UPSs with upssched

[Nut-upsuser] Email Alerts for Multiple UPSs with upssched

[Nut-upsuser] Email Alerts for Multiple UPSs with upssched

[Nut-upsuser] Email Alerts for Multiple UPSs with upssched

Possibly Parallel Threads