ICMP PING to Monitor hosts

[Argon0]

HI,

I've got a number (1500+) of Medical devices which I need to know when they are NOT on the network. ICMP ping is the only way I have of doing this across all the different makes/models, etc.. (NOT all support SNMP). So the ICMP is all good and working, I'm getting responses and graphs BUT... I need a dashboard showing availability - from what we can work out there isn't an easy way to do this...

I've tried top hosts using ICMP response time and ICNP LOSS as columns, but just get NO DATA. BUT I can see the individual graphs and the data in latest data...

What am I doing wrong?

P.S. I'd love to use the "host availability" widget but that's a no go, and the Problems widget DOES give me some joy, just not enough.

Cheers

Argon0

[vijayk]

I have same things. I tried all in Zabbix dashboard and then after integrates the Grafana with Zabbix and now i able to monitor all devices with single graph.
https://blog.zabbix.com/configuring-...h-zabbix/8007/

[Argon0]

How did you use Grafana to do this?

[vijayk]

Refer this blog: https://blog.zabbix.com/configuring-...h-zabbix/8007/

and Integration link: https://www.zabbix.com/integrations/grafana

[vijayk]

Did you tried Data overview in Zabbix dashboard?

[Argon0]

I've integrated Grafana, and got it showing some graphs, and the Zabbix Problems board

How do you specifically use it to show number of hosts up and down when using just ICMP?

[vijayk]

See below SS. Use item as /ICMP ping/ in Group which you want to monitor.

[Argon0]

TVM I'll give that a go!

[vijayk]

WELCOME...!

[Argon0]

Well, its giving me something... Do you know of any way to get something like this: Using the Ping stats (up and down, or ICMP Loss, ideally)...

Cheers

[vijayk]

Yes, you can use Data overview

in dashboard of Zabbix. You will the device status like this

[Argon0]

Yeah, that's in Zabbix, and gives too much info - just looking for a host availabilty widget - think I *may* have found it in Grafana - there are a number of Status dashboards available in plugins - testing..

[Argon0]

Nope nothing is giving me a point in time availabilty, but I'm thinking I need to do some SQL, which, TBH, is daunting...

[vijayk]

Which items you want to monitor?

[Argon0]

I have devices in a group - Med Devices - which I only monitor via ICMP. The three metrics I get are ICMP Ping (giving up and down), ICMP Loss - giving a percentage, and ICMP Response Time. I'm thinking ICMP Loss gives the best info for this.

So I'd use 75%+ ICMP Loss for Unavailable. Anything under for Available. I can't think when unknown would trigger, but I'd want total as well.

And that's it for this counter.... I'm currently getting this from a table, but the problem is it needs to be tied to a specific time to be usable at all..(I have since worked out how to hide columns)


I'd want to group (average) values from unique hosts per time period: so if I have 100 hosts which are pinged x times over y minutes I'd want 100 values for each ICMP Loss over that period, no matter how many times they were tested . The table currenlty gives a set of 5 results (as I'd expect) for each time period (I've filtered to last measure on the above). I used transform to heatmap to get the above with buckets of 5m x, and 25(%) y.

It's getting there, but is clumsy so far...