Latency Uptime Warnings/Alerts and Juniper Switches

[emssa]

It may be worth mentioning as well to download or wget the plugin check_tcping from the Nagios exchange

https://exchange.nagios.org/components/ ... 2&cf_id=24

yum install gcc

unzip/tar

man unix

place into /usr/local/nagios/libexec/

chown nagios:nagios /usr/local/nagios/libexec/check_tcping
chmod 775 /usr/local/nagios/libexec/check_tcping

[kyang]

Thanks, I am aware

Great, since you've created the command with the plugin.

Now, if you wanted this on a specific host --> in the Core Config Manager --> click on a host you want to use the plugin with, and add the check_tcping to the check_command. Add your warnings and criticals.

Save --> apply configuration, and now it should work.

Here's my screenshot for a better image.

Capture.PNG

You may have to force an immediate check on that host if you don't see the results for check_tcping, but overall this worked for me.

I hope this helps.

[emssa]

I tested it via your steps

on the host screen under the command in ARG1 through ARG4 added similar 10,20,30,40...

Screenshot from 2017-12-13 12-55-23.png

Is there a more appropriate number that should be used for each ARG?

Are these ARG's high and low thresholds etc?

I then went down to the Run Check Command and tested for no errors

/usr/local/nagios/libexec/check_tcping -w 10,20 -c 30,40 Hostname
TCPING OK: Packet loss = 0.00%, RTA = 3.54 ms

I clicked save and applied the config when prompted

Now what is needed to push this out via Bulk Modifications tool?

[kyang]

The thresholds for warnings and critical are up to you. (I just threw in random numbers to see if it's working)

Assuming this plugin is the similar to check_ping --help

Code: Select all

THRESHOLD is <rta>,<pl>% where <rta> is the round trip average travel
time (ms) which triggers a WARNING or CRITICAL state, and <pl> is the
percentage of packet loss to trigger an alarm state.

check_tcping --help

Code: Select all

-w <rta>,<%loss> -c <rta>,<%loss> 

RTA is the average travel time, and %loss should be packet loss percentage.

I don't know your requirements, but I think the default settings for thresholds on check_ping in localhost is a good default. They should be OK, but it's up to you.

Code: Select all

-w 100.0,20% -c 500.0,60%

Once you have the thresholds you want, in the Bulk Modifcations Tools --> click "change command and arguments" Find check_tcping, add in your threshold ARGS.

Capture.PNG

Then select the hosts you want this plugin to be applied to. Then save changes.

I hope this helps.

[emssa]

Yes it does help, though we are now seeing 100% loss on switches and win servers, everything else seems happy.

Host Down: TCPING CRITICAL: Packet loss = 100.00%, RTA = (not available due to 100% packets lost)

Guess those have something specific that need to be done to allow tcping vs ICMP?

[kyang]

I can't be sure since it's an exchange plugin.

What happens if you use check_icmp on a win server or switch?

Does it work or is it critical?

[emssa]

tried switching it back to check_icmp (ping)

and all I get is:

check_ping: %s: Warning threshold must be integer or percentage!

[emssa]

nevermind found an article 20 deep that shows you have to use a percent sign..

Does Nagios have a page with all the commands and their default settings?

So put in 100.0,20% and 200.0,40% in ARG1 and ARG2

This made the errors go away

[emssa]

Just followed up with the switches and adding back the check_icmp and they are now clear as well

[kyang]

Great to hear that!

Is this resolved for you? Did you have any more questions on this or are we okay to lock this up?