Difference between revisions of "SMS Notifications"

From ZoneMinder Wiki
Jump to navigationJump to search
 
(21 intermediate revisions by the same user not shown)
Line 1: Line 1:
These instructions are to assist with email or text message setup.
These are general email instructions. Eventually I will merge them to the Email page.


For SMS, You setup email to point to a [https://en.wikipedia.org/wiki/List_of_SMS_gateways text gateway]. Setup a filter that monitors the cameras for alarms, and when an alarm
For SMS, You setup email to point to a [https://en.wikipedia.org/wiki/List_of_SMS_gateways text gateway]. Setup a filter that monitors the cameras for alarms, and when an alarm
Line 68: Line 68:
* Once an email has been sent for an event, that event is marked and an email for it '''won't be sent again'''[https://forums.zoneminder.com/viewtopic.php?f=8&t=7774&p=27386&].
* Once an email has been sent for an event, that event is marked and an email for it '''won't be sent again'''[https://forums.zoneminder.com/viewtopic.php?f=8&t=7774&p=27386&].


* Cron may require setting an environment variable of EMAIL=sendingaddress@email.com in the crontab
* Cron may require setting an environment variable of EMAIL=sendingaddress@email.com in the crontab. When testing this from the shell, you must do the same but with export, e.g. "export EMAIL=sendingaddress@email.com"


=== Example Filter ===
=== Example Filter ===
Line 74: Line 74:
See [[Filters]]
See [[Filters]]


== Send email if only on a certain runstate (Filter on Run States) ==
=== Send email if only on a certain runstate (Filter on Run States) ===


Notifications are easy if you want alarms emailed between a static time, for example between 12AM and 6AM. If you have a situation where you may not want the alarms to send you text messages on occasion, you can filter on run states.  
Notifications are easy if you want alarms emailed between a static time, for example between 12AM and 6AM. If you have a situation where you may not want the alarms to send you text messages on occasion, you can filter on run states.  
Line 82: Line 82:
Note that in the author's opinion, run states can become unwieldy for large installations. It's best used with smaller setups only.
Note that in the author's opinion, run states can become unwieldy for large installations. It's best used with smaller setups only.


== Send email if a part of ZM goes offline ==
== Examples of Emails for Monitoring ==
=== Send Email if ZM is Not Running ===
Here is a minimal script to check that Zoneminder is running. Note that this does not check for mysql errors or hard drive errors (which are possible). Do not put any spaces in the subject or email. Also, you may want to monitor that the server is actually powered on, from another computer (as if the server shuts down for any reason, this will obviously not run).
<pre>
#!/bin/bash
 
 
SUBJECT="Name_of_server_ZM"
EMAIL="emailaddress"
 
/etc/init.d/zoneminder status 2> /dev/null
if test $? -ne 0 ; then
        echo "Detected ZM not running"
        echo "ZM is not running" | mutt -s $SUBJECT $EMAIL
else
        echo "ZM is running, do nothing"
fi
</pre>


For any serious deployment you will want to monitor ZM in case it goes offline, or a hdd fails. While it's possible to use a separate system for this (e.g. https://en.wikipedia.org/wiki/Comparison_of_network_monitoring_systems), you can also simply have email alerts in cron. Here is an example script that can watch two hdds (adjust as needed) and the status of ZM and Mysql.  
alt.
<pre>
<pre>
#!/bin/bash
#usage: feed $1 subject
/etc/init.d/zoneminder status > /root/zm_up_log
zmdc.pl check | grep stopped
echo "" >> /root/zm_up_log
if [ $? -eq 0 ]; then
/etc/init.d/mysql status >> /root/zm_up_log
#send email
ps -Ao user,fname,pmem | grep -e zmc -e zma | grep -v zmaudit >> /root/zm_up_log
  echo "ZM is not running" | mutt -s $1 email@address.com
echo "" >> /root/zm_up_log
else
echo "" >> /root/zm_up_log
  echo "do nothing"
lsblk >> /root/zm_up_log
fi
echo "" >> /root/zm_up_log
</pre>
echo "" >> /root/zm_up_log
===Send Email if MySQL is Not Running===
dmesg | grep -e sda -e sdb >> /root/zm_up_log
<pre>
#usage: feed $1 subject
mysqladmin processlist | grep zmuser
#returns 0 if zmuser detected
if [ $? -eq 0 ]; then
#send email
  echo "do nothing"
else
  echo "mysql is not running" | mutt -s $1 user@address.com
fi


echo "ZM_Status" | mutt -s "ZM Status" destination@email.com -a /root/zm_up_log
</pre>
</pre>
This script can be run in crontab as often as you wish. It should also be easy to filter the script to email only if something is out of the ordinary. In fact, let's do that next.


== Send email if HDD Fails ==
=== Send Email if HDD Fails ===
This script will give you 1 email an hour, if a hdd has an error. This is an alternative way of checking hdd status instead of S.M.A.R.T. .
<pre>
<pre>
#!/bin/bash
#!/bin/bash
# 20 * * * *  root /root/hdderror.sh (this script)
# 20 * * * *  root /root/hdderror.sh (this script, and the root assumes you are using /etc/crontab)
# must run as root for access to dmesg in debian
# must run as root for access to dmesg in debian
# chmod +x  
# chmod +x  


LOGFILE=/root/file.log
LOGFILE=/root/file.log
SUBJECT="e.g. Home Camera System HDD Error"
SUBJECT="e.g._Home_Camera_System_HDD_Error"


echo "" > $LOGFILE
echo "" > $LOGFILE
Line 121: Line 146:
#send email
#send email
   echo "HDD Error..." | mutt -s $SUBJECT destination@email.com -a $LOGFILE
   echo "HDD Error..." | mutt -s $SUBJECT destination@email.com -a $LOGFILE
  echo "email sent"
else
else
   echo "do nothing"
   echo "do nothing"
fi
</pre>
=== Send Email if Camera is Offline ===
This script will ping an ip address 10 times, then if there is an error, it will try again 10 more times. If it still cannot connect, an email will be sent. There is a lock file, so that only one email is sent every couple hours.
<pre>
#!/bin/bash
SERVERIP=$1
LOGFILE=$1_$(date +%A)_LOG
HISTORYFILE=$1_$(date +%A)_LOCKFILE
NOTIFYEMAIL=email_address
#setup this script in cron once an hour (or as frequent as you want), and also
#crontab requires historyfile / lockfile to be blanked (echo "" > file) each day or each hour, whatever you prefer.
#make the log directory:
#mkdir /var/log/networkalerts
#this script will run like this:
#e.g. $ test_up.sh <ipaddress> (test_up is this script)
# in /etc/crontab
#0 * * * *  root /root/test_up.sh 192.168.1.1
#0 */5 * * * root rm /var/log/networkalerts/*LOCKFILE
#0 0 * * *  root rm /var/log/networkalerts/*$(date +%A)*LOG
#check for lock file
if test -s /var/log/networkalerts/$HISTORYFILE
then
exit
fi
#keep track of time
date >> /var/log/networkalerts/$LOGFILE
ping -c 10 $SERVERIP >> /var/log/networkalerts/$LOGFILE
#if return val is error 1 (but not 2) (see man on ping regarding count and deadline)(success return is zero)
if test $? == 1
  then
  sleep 60
else
exit
fi
#reduce false alerts, run it once again to be certain.
ping -c 10 $SERVERIP >> /var/log/networkalerts/$LOGFILE
  if test $? == 1
  then
    echo "$SERVERIP is down" | msmtp -a default $NOTIFYEMAIL
    #lock file / history file
    echo "alertsent" > /var/log/networkalerts/$HISTORYFILE
else
exit
fi
fi


</pre>
</pre>
</pre>
=== Basic Server Monitor ===
These would go in /etc/crontab. You might run this once a week, to monitor a server.
<pre>
1 0 * * 1 root dmesg | tail -10 > /tmp/hdd_report
2 0 * * 1 root lsblk  >> /tmp/hdd_report
3 0 * * 1 root ls -lt /videos/ >> /tmp/hdd_report
4 0 * * 1 root df -h >> /tmp/hdd_report
5 0 * * 1 root echo "zm server" | mutt -s "zm server" user@email.com -a /tmp/hdd_report
</pre>
==Troubleshooting==
===Error sending message, child exited 127 (Exec error.).===
If when sending something like the following you get the above error:
echo "message" | mutt  -s "cameras_email" user@email -a /tmp/attachment.txt
Error sending message, child exited 127 (Exec error.).
You need to install msmtp-mta.


==Resources==
==Resources==
Line 138: Line 232:
* [https://forums.zoneminder.com/viewtopic.php?f=9&t=12646 Receive Email on Runstate change]
* [https://forums.zoneminder.com/viewtopic.php?f=9&t=12646 Receive Email on Runstate change]
* [https://forums.zoneminder.com/viewtopic.php?f=40&t=29755&p=116634#p116634]
* [https://forums.zoneminder.com/viewtopic.php?f=40&t=29755&p=116634#p116634]
 
* https://linux.die.net/man/8/logcheck - logcheck / logfile scanner.


[[Category:Dummies_Guide]]
[[Category:Dummies_Guide]]

Latest revision as of 02:54, 10 December 2024

These are general email instructions. Eventually I will merge them to the Email page.

For SMS, You setup email to point to a text gateway. Setup a filter that monitors the cameras for alarms, and when an alarm is set off you have it send an email to the text gateway (or email address). Simple.

You can get images from the alarm embedded in the text (covered in install steps).

The following is used in 1.36 / Debian 10 and newer.

Installation

Follow How to get ssmtp working with Zoneminder.

There are two ways to send emails in options - email. One is sending a short email, and one is designed for longer messages. In the code (as of 1.30 lines 818 and 918), in zmfilter.pl these are correspondingly sub SendEmail and sub SendMessage. They are distinct functions. SSMTP/MSMTP support is in SendEmail but not in SendMessage.

MSMTP Specific Install

Debian 10 and newer should use MSMTP.

apt-get install msmtp msmtp-mta mutt

Testing the email

First test via terminal outside of ZM. Confirm it is working there. Example configurations and calls can be found at: https://wiki.debian.org/msmtp and https://wiki.archlinux.org/index.php/MSMTP (for msmtp).

Once you get a working /etc/msmtprc configuration, try:

echo "hello" | msmtp -a default user@email.com

Watch out for the following:

  • Make sure app-armor is disabled for msmtp. The following commands will help: apt install apparmor-utils and then aa-disable /etc/apparmor.d/usr.bin.msmtp

Testing the email in Zoneminder

Then:

  1. run a filter in the background (see tips below) or execute.
  2. Enable logging on /var/log/zm folder.
  3. tail the ZMFilter log and syslog.
  4. to initiate an alarm (if alarms are used to test emails), use 'force alarm' on the monitor screen. That is, view a camera, and click force alarm.

You should see SSMTP send an email on the syslog (or try) and the zmfilter log will report the trigger.

Tips

  • Free email services may limit how many emails you are allowed to send.
  • Always make filters for 1 (or low number) results. If you fail to do this, you could try to send dozens of emails.
  • When you make the filter you want the 'current' alarms. You don't want old events. This is accomplished with a Date/Time greater than or equal to -1 minute.
  • If when testing the filter execute, the 'execute' button becomes unclickable, uncheck then recheck the email option.
  • If you send email as someone other than root you may need to include the user in revaliases.
  • test email attachments with one of these:
echo hello | mail -A /dir/to/attachment.log -s "email with file" destination@email.com
mutt -a /dir/to/attachment.log -s "email with file" -- destination@email.com
echo "test" | mutt -d 5 -s "Test Email" user@email.com -a /tmp/file_attachment
  • make sure email is checked off in filters.
  • service zoneminder restart may be needed after changes (depends upon ZM version)
  • test setting off alarms by clicking 'force alarm' in the camera monitor view, then unclick to remove alarm.
  • If you find false alerts with motion detection, consider using either ZMES, or passing the event to an object detection based script.
  • Always double check the filters are right, AFTER you save them.
  • Once an email has been sent for an event, that event is marked and an email for it won't be sent again[1].
  • Cron may require setting an environment variable of EMAIL=sendingaddress@email.com in the crontab. When testing this from the shell, you must do the same but with export, e.g. "export EMAIL=sendingaddress@email.com"

Example Filter

See Filters

Send email if only on a certain runstate (Filter on Run States)

Notifications are easy if you want alarms emailed between a static time, for example between 12AM and 6AM. If you have a situation where you may not want the alarms to send you text messages on occasion, you can filter on run states.

Starting from (ZM > 1.30.4) you can set filters based on run states, which means you can change the run state at any desired time (using Cron, or manually), and only if a runstate matches the filter will SMS be sent.

Note that in the author's opinion, run states can become unwieldy for large installations. It's best used with smaller setups only.

Examples of Emails for Monitoring

Send Email if ZM is Not Running

Here is a minimal script to check that Zoneminder is running. Note that this does not check for mysql errors or hard drive errors (which are possible). Do not put any spaces in the subject or email. Also, you may want to monitor that the server is actually powered on, from another computer (as if the server shuts down for any reason, this will obviously not run).

#!/bin/bash


SUBJECT="Name_of_server_ZM"
EMAIL="emailaddress"

/etc/init.d/zoneminder status 2> /dev/null
if test $? -ne 0 ; then
        echo "Detected ZM not running"
        echo "ZM is not running" | mutt -s $SUBJECT $EMAIL
else
        echo "ZM is running, do nothing"
fi

alt.

#usage: feed $1 subject 
zmdc.pl check | grep stopped
if [ $? -eq 0 ]; then
#send email
  echo "ZM is not running" | mutt -s $1 email@address.com
else
  echo "do nothing"
fi

Send Email if MySQL is Not Running

#usage: feed $1 subject 
mysqladmin processlist | grep zmuser
#returns 0 if zmuser detected
if [ $? -eq 0 ]; then
#send email
  echo "do nothing"
else
  echo "mysql is not running" | mutt -s $1 user@address.com
fi

Send Email if HDD Fails

This script will give you 1 email an hour, if a hdd has an error. This is an alternative way of checking hdd status instead of S.M.A.R.T. .

#!/bin/bash
# 20 * * * *   root /root/hdderror.sh (this script, and the root assumes you are using /etc/crontab)
# must run as root for access to dmesg in debian
# chmod +x 

LOGFILE=/root/file.log
SUBJECT="e.g._Home_Camera_System_HDD_Error"

echo "" > $LOGFILE

# any text of "error" with regards to sda,sdb, etc that isn't a remount
dmesg | grep -e sda -e sdb -e sdc -e sdd -e sde | grep -i error | grep -v remount >> $LOGFILE
#returns 1 if nothing

if [ $? -eq 0 ]; then
#send email
  echo "HDD Error..." | mutt -s $SUBJECT destination@email.com -a $LOGFILE
  echo "email sent"
else
  echo "do nothing"
fi

Send Email if Camera is Offline

This script will ping an ip address 10 times, then if there is an error, it will try again 10 more times. If it still cannot connect, an email will be sent. There is a lock file, so that only one email is sent every couple hours.

#!/bin/bash 

SERVERIP=$1
LOGFILE=$1_$(date +%A)_LOG
HISTORYFILE=$1_$(date +%A)_LOCKFILE
NOTIFYEMAIL=email_address

#setup this script in cron once an hour (or as frequent as you want), and also
#crontab requires historyfile / lockfile to be blanked (echo "" > file) each day or each hour, whatever you prefer.
#make the log directory:
#mkdir /var/log/networkalerts
#this script will run like this:
#e.g. $ test_up.sh <ipaddress> (test_up is this script)
# in /etc/crontab
#0 * * * *   root /root/test_up.sh 192.168.1.1
#0 */5 * * * root rm /var/log/networkalerts/*LOCKFILE
#0 0 * * *   root rm /var/log/networkalerts/*$(date +%A)*LOG

#check for lock file
if test -s /var/log/networkalerts/$HISTORYFILE
 then
 exit
fi

#keep track of time
date >> /var/log/networkalerts/$LOGFILE
ping -c 10 $SERVERIP >> /var/log/networkalerts/$LOGFILE
#if return val is error 1 (but not 2) (see man on ping regarding count and deadline)(success return is zero)
if test $? == 1
  then
  sleep 60
 else
exit
fi

#reduce false alerts, run it once again to be certain.
ping -c 10 $SERVERIP >> /var/log/networkalerts/$LOGFILE
   if test $? == 1
   then
    echo "$SERVERIP is down" | msmtp -a default $NOTIFYEMAIL
    #lock file / history file
    echo "alertsent" > /var/log/networkalerts/$HISTORYFILE
 else
 exit
fi

Basic Server Monitor

These would go in /etc/crontab. You might run this once a week, to monitor a server.

1 0 * * 1 root dmesg | tail -10 > /tmp/hdd_report
2 0 * * 1 root lsblk  >> /tmp/hdd_report
3 0 * * 1 root ls -lt /videos/ >> /tmp/hdd_report
4 0 * * 1 root df -h >> /tmp/hdd_report
5 0 * * 1 root echo "zm server" | mutt -s "zm server" user@email.com -a /tmp/hdd_report

Troubleshooting

Error sending message, child exited 127 (Exec error.).

If when sending something like the following you get the above error:

echo "message" | mutt  -s "cameras_email" user@email -a /tmp/attachment.txt
Error sending message, child exited 127 (Exec error.).

You need to install msmtp-mta.

Resources