I am using nagios ver. 4.0.8 .
I want to set interval between ping times is 10 seconds like below:
define command{
command_name check-host-alive
command_line $USER1$/check_ping -t 10 -H $HOSTADDRESS$ -w 3000.0,80% -c 5000.0,100% -p 5
}
But not exactly (about 90 seconds). Can you help me?
Thanks
You are looking at things the wrong way.
define command{
command_name check-host-alive
command_line $USER1$/check_ping -t 10 -H $HOSTADDRESS$ -w 3000.0,80% -c 5000.0,100% -p 5
}
The thing you post is a Nagios command. the '-t 10' is not the interval, but the timeout argument. This means if the check_ping does not get a result in 10 seconds, the command will timeout.
To define the check interval, you need to look at the host (or service) configuration file.
For example:
define host {
host_name bogus-router
alias Bogus Router #1
address 192.168.1.254
parents server-backbone
check_command check-host-alive
check_interval 5
retry_interval 1
max_check_attempts 5
check_period 24x7
process_perf_data 0
retain_nonstatus_information 0
contact_groups router-admins
notification_interval 30
notification_period 24x7
notification_options d,u,r
}
The interval between checks in this example is 5 minutes (check_interval). It is not possible to set intervals of less then one minute with Nagios. If you want to have more granular (free) monitoring, check out InfluxDB, Telegraf and Grafana.
Related
I have the following service definition:
define service{
use my-service ; Name of service template to use
host_name dra
service_description https://www.example.com
check_command check_http!-I my.ip.address --ssl -H www.example.com
notifications_enabled 1
retry_check_interval 2
normal_check_interval 5
contact_groups myadmins
}
The service check keeps failing with
Name or service not known
HTTP CRITICAL - Unable to open TCP socket
However, if I run http_check from the command line, I get a 200 OK result:
/usr/lib/nagios/plugins/check_http -I my.ip.address --ssl -H www.example.com -v
.....
HTTP OK: HTTP/1.1 200 OK - 9176 bytes in 0.074 second response time |time=0.073543s;;;0.000000 size=9176B;;;0
Note also that the URL in question works just fine from a browser, the certificate is valid, etc. I also use the exact same service definition for a bunch of other sites, and they all work fine. The only thing I can think of is that this remote host is running on DigitalOcean and has a "Floating IP" assigned to it. I tried replacing my.ip.address above (and also in the host definition of the nagios config file) with either the Floating IP or the "standard" IP assigned to the host, and it makes no difference.
How is it possible that the same command would fail when run by nagios, but succeed when run manually?
The answer to my question is: don't use check_http, use
use check_https_hostname, and
make sure that the host_name stanza is the actual hostname
which requires matching the host_name stanzas in all the service and host definitions in the same cfg file.
So:
define service{
use my-service ; Name of service template to use
host_name www.example.com
service_description https://www.example.com
check_command check_https_hostname
notifications_enabled 1
retry_check_interval 2
normal_check_interval 5
contact_groups myadmins
}
Here is why: it becomes clear by looking at the definitions of check_http and check_https_hostname which are in the /etc/nagios-plugins/config/http.cfg file in my installation.
# 'check_http' command definition
define command{
command_name check_http
command_line /usr/lib/nagios/plugins/check_http -H '$HOSTADDRESS$' -I '$HOSTADDRESS$' '$ARG1$'
}
# 'check_https_hostname' command definition
define command{
command_name check_https_hostname
command_line /usr/lib/nagios/plugins/check_http --ssl -H '$HOSTNAME$' -I '$HOSTADDRESS$' '$ARG1$'
}
You will notice that the -H and -I arguments in check_http get the same value $HOSTADDRESS$, while in check_https_hostname they get $HOSTNAME$ and $HOSTADDRESS$, respectively.
The fact that I built my original command as check_http!-I my.ip.address --ssl -H www.example.com did not really matter. In the end, the /usr/lib/nagios/plugins/check_http command got two values for -I and two for -H, and the second pair was being ignored.
This did break "thanks" to Cloudflare, because the IP address dynamically assigned by Cloudflare to my www.example.com was not the same as the actual host IP address that I had specified in my host definition.
Finally, I wanted to mention that what helped me figure this out was setting
debug_level=-1
debug_verbosity=1
in my /etc/nagios3/nagios.cfg file and then looking through /var/log/nagios3/nagios.debug.
Also, check out all the different variants of the check_http commands in /etc/nagios-plugins/config/http.cfg. There are some very useful ones.
I have formed a command for fetching established port connection using nagios check_by_ssh module.
I am able to get the output when I run the command, however after placing the command in the commands.cfg file I am seeing "check_by_ssh: skip-stderr argument must be an integer " in the GUI. Any suggestion on this would be of great help.
Command:
/usr/local/nagios/libexec/check_by_ssh -l fuseadmin -H <hostname> -C "netstat -punta | grep -i ESTABLISHED | wc -l | awk '{if (\$0>2500) {print \"CRITICAL: Established Socket Count: \"\$0} else {print \"OK: Established Socket Count: \"\$0}}'" -i ~/.ssh/id_dsa -E
OK: Established Socket Count: 67
Commands.cfg:
define command {
command_name netstat_cnt_estanblished_gt_2500_fuse01
command_line /usr/local/nagios/libexec/check_by_ssh -l fuseadmin -H a0110pcsgesb01 -C "netstat -punta | grep -i ESTABLISHED | wc -l 2>&1 | awk '{if (\$0>2500) {print \"CRITICAL: Established Socket Count: \"\$0} else {print \"OK: Established Socket Count: \"\$0}}'" -i ~/.ssh/id_dsa -E
}
Service Definition
#netstat_cnt_estanblished_gt_2500_csg2.0
define service{
use generic-service ; Name of service template to use
host_name <hostname>
service_description Netstat Established Count
event_handler send-service-trap-fms
event_handler_enabled 1
check_command netstat_cnt_estanblished_gt_2500_fuse01
max_check_attempts 1
notifications_enabled 1 ; Service notifications are enabled
check_period 24x7 ; The service can be checked at any time of the day
max_check_attempts 3 ; Re-check the service up to 3 times in order to determine its final (hard) state
check_interval 2 ; Check the service every 10 minutes under normal conditions
retry_interval 2 ; Re-check the service every two minutes until a hard state can be determined
contact_groups fuse_users ; Notifications get sent out to everyone in the 'admins' group
notification_options w,u,c,r ; Send notifications about warning, unknown, critical, and recovery events
notification_interval 30 ; Re-notify about service problems every hour
notification_period 24x7
}
**I have changed the actual hostname to due to compliance
here it says:
check_by_ssh: print command output in verbose mode
right now it is not possible to print the command output of ssh. check_by_ssh
only prints the command itself. This patchs adds printing the output too. This
makes it possible to use ssh with verbose logging which helps debuging any
connection, key or other ssh problems.
Note: you must use -E,--skip-stderr=<high number>, otherwise check_by_ssh would
always exit with unknown state.
Example:
./check_by_ssh -H localhost -o LogLevel=DEBUG3 -C "sleep 1" -E 999 -v
Meaning: you should just have to specify a number after "-E", like -E 999, in your definition (like the example in above code-block says)
... even though, it's confusing (maybe a bug?), because the command help of check_by_ssh says:
-E, --skip-stderr[=n]
Ignore all or (if specified) first n lines on STDERR [optional]
I have currently the following two service defined as below:
define service {
use my-webapp-service
hostgroup_name all
service_description System check - PING
check_command check_ping!100.0,20%!500.0,60%
}
define service {
use my-webapp-service
hostgroup_name all
service_description System check - Swap Usage
check_command check_nrpe!check_swap
check_interval 1
}
What I want is output string to be:
System check - PING - "Actual hostname where this alarm got fired off"
System check - Swap Usage - "Actual hostname where this alarm got fired off"
I think this could be possible but I just don't know how to make it possible.
Would sincerely appreciate your guidance on that.
Many Thanks
Output are handled by scripts. Default behavior is that script don´t return hostname, because it is not necessary.
If you wanna add hostname in output, you must edit already exist scripts or create new one.
Here is basic info how create script for Nagios - http://www.yourownlinux.com/2014/06/how-to-create-nagios-plugin-using-bash-script.html
For your needs you must add $HOSTNAME to echo. For instance:
echo "$HOSTNAME - WARNING- $output"
If you want the script that is executing to be aware of the hostname, you'll need to pass the hostname as an argument to the Nagios command. That also means that the script will need to accept the hostname as an argument. Take for example:
define service {
use my-webapp-service
hostgroup_name all
service_description System check - PING
check_command check_ping!100.0,20%!500.0,60%
}
check_ping probably looks something like:
define command {
command_name check_ping
command_line $USER1$/check_ping -H $HOSTADDRESS$ -w $ARG1$ -c $ARG2$ -p 5
}
The problem here is that the executable at $USER1$/check_ping doesn't know that you want to pass the host's name as an argument. So you'll need to make a wrapper script. I'm not going to write the script for you, but to give you a hint, the command definition would look something like:
define command {
command_name check_ping_print_hostname
command_line $USER1$/my_check_ping_wrapper.sh --hostname $HOSTNAME$ -H $HOSTADDRESS$ -w $ARG1$ -c $ARG2$ -p 5
}
And then the script at $USER1$/my_check_ping_wrapper.sh is obviously going to need grab that --hostname argument, and then pass the other arguments directly to check_ping, wait for the output, and then amend the output with the information given in the --hostname arg.
Hope this helps!
I have a nagios system, working well and i wanted to check a specific url with check_http.
The command is defined:
define command{
command_name check_http_with_folder
command_line $USER1$/check_http -H $HOSTADRESS$ -u http://$HOSTADRESS$$ARG1$
}
and i call it correct ... But it throws me an
"Name or service not known"
When i call it from my nagios machine from command line, it works well and i get an status result 200, so all okay.
The problem is now, that i want the nagios command working and not throwing an error.
Any Ideas?
P.S. The problem is only in the part with the -u xxx param, without it (in the normal check_http command without -u) it all works well.
You've misspelled $HOSTADDRESS$ in your command definition. It needs 2 D's. Also, you might want to ensure there is a slash in between $HOSTADDRESS$ and $ARG1$ in the value you pass in to your -u command argument, or make sure that $ARGS1$ has a preceding slash.
Building on Joe's observations...
Note the corrections:
define command{
command_name check_http_with_folder
command_line $USER1$/check_http -H $HOSTADDRESS$ -u $ARG1$
}
Then the $HOSTADDRESS$ should be just that. For example, www.example.com. And $ARG1$ should be the location at the host only. For example, /blog/index.php. The check_http check will build it into an actual http request.
I've deployed a new instance of Nagios on a fresh install of CentOS 7 via the EPEL repository. So the Nagios Core version is 3.5.1.
After installing nagios and nagios-plugins-all (via yum), I've created a number of hosts and service definitions, have tested my configuration with nagios -v /etc/nagios/nagios.cfg, and have Nagios up and running!
Unfortunately, my host checks are failing (although my service checks are working perfectly fine).
Within the Nagios Web GUI / Dashboard, if I drill down into a Host page with the "Host State Information", I see this being reported for "Status Information" (IP address removed):
Status Information: /usr/bin/ping -n -U -w 30 -c 5 {my-host-ip-address}
CRITICAL - Could not interpret output from ping command
So in my troubleshooting, I drilled down into the Nagios Plugins directory (/usr/lib64/nagios/plugins), and ran a test with the check_ping plugin consistent with the way check-host-alive runs the command (see below for my check-host-alive command definition):
./check_ping -H {my-ip-address} -w 3000.0,80% -c 5000.0,100% -p 5
This check_ping command returns the following output:
PING OK - Packet loss = 0%, RTA = 0.63
ms|rta=0.627000ms;3000.000000;5000.000000;0.000000 pl=0%;80;100;0
I haven't changed the definition of how check_ping works, and can confirm that I'm getting a "PING OK" whenever the command is run the same way that check-host-alive runs the command, so I cannot figure out what's going on!
Below are the command definitions for check-host-alive as well as check_ping.
# 'check-host-alive' command definition
define command{
command_name check-host-alive
command_line $USER1$/check_ping -H $HOSTADDRESS$ -w 3000.0,80% -c 5000.0,100% -p 5
}
{snip}
# 'check_ping' command definition
define command{
command_name check_ping
command_line $USER1$/check_ping -H $HOSTADDRESS$ -w $ARG1$ -c $ARG2$ -p 5
}
Any suggestions on how I can fix my check-host-alive command definition to work properly and evaluate the output of check_ping properly?
Edit
Below is the full define host {} template I'm using:
define host {
host_name myers ; The name of this host template
alias Myers
address [redacted]
check_command check-host-alive
contact_groups admins
notifications_enabled 0 ; Host notifications are enabled
event_handler_enabled 1 ; Host event handler is enabled
flap_detection_enabled 1 ; Flap detection is enabled
failure_prediction_enabled 1 ; Failure prediction is enabled
process_perf_data 1 ; Process performance data
retain_status_information 1 ; Retain status information across program restarts
retain_nonstatus_information 1 ; Retain non-status information across program restarts
notification_period 24x7 ; Send host notifications at any time
register 1
max_check_attempts 2
}
For anyone else who runs into this issue, there's another option than changing permissions on ping. Simply change the host check command to use check_host rather than check_ping. While there are certainly some differences in the functionality, the overall end result is the same.
There are those who will say this isn't a good option because of the ability to range the check_ping command, but it should be remembered that host checks aren't even executed until all service checks for a given host have failed. Anyway, if you're interested in testing throughput, there are MUCH better ways of going about it than relying on ICMP, which is the lowest priority traffic type on a network.
I'm sure the OP is well on to other things by now, but hopefully someone else who has this issue will benefit.
I could not found the ping on /usr/bin/ping
# chmod u+s /bin/ping
# ls -al /bin/ping
-rwsr-xr-x 1 root root 40760 Sep 26 2013 /bin/ping*
Finally run the below command,
/usr/local/nagios/libexec/check_ping -H 127.0.0.1 -w 100.0,20% -c 500.0,60% -p 5
I was fairly certain that running chmod U+s /usr/bin/ping would solve the issue, but I was (and still am) wary about chmod'ing system files. It seems to me that there has to be a safer way to do it.
However, in the end, that's what I did - and it works. I don't like it, from a security standpoint.
I also had same problem and the above answers did not work for me. After some checking the issue further noticed that the reason is IP protocol. once I passed the correct IP protocol , It worked fine.
/usr/local/nagios/libexec/check_ping -H localhost -w 3000.0,80% -c 5000.0,100% -4
output
PING OK - Packet loss = 0%, RTA = 0.05 ms|rta=0.051000ms;3000.000000;5000.000000;0.000000 pl=0%;80;100;0
By default It's getting IPv6.
/usr/local/nagios/libexec/check_ping -H localhost -w 3000.0,80% -c 5000.0,100% -6
output
/sbin/ping6 -n -U -W 30 -c 5 localhost
CRITICAL - Could not interpret output from ping command
But when integrating with Nagios server, I could not able to pass this value as an argument. Therefore I have done below workaround in client side nrpe.cfg file
command[check_ping_args]=/usr/local/nagios/libexec/check_ping -H $ARG1$ -w $ARG2$ -c $ARG3$ -4
Here Host, warning and critical thresholds were passing by Nagios host as below,
define service{
use generic-service
hostgroup_name all-servers
service_description Host Ping Status
check_command check_nrpe_args!check_ping_args!localhost!3000.0,80%!5000.0,100%
}