How to monitor the running status of zabbix_agentd?

[Bill Wang]

How to monitor the running status of zabbix_agentd?

[marcherren]

Use the key agent.ping from the Zabbix Agent

Check the agent usability. Always return 1. Can be used as a TCP ping.

[Bill Wang]

Use the key agent.ping from the Zabbix Agent

Check the agent usability. Always return 1. Can be used as a TCP ping.

Is this same as the key "icmpping"?

[marcherren]

No, icmpping is well an icmp ping (Internet Control Message Protocol)

agent.ping is an application check on a specific port.

icmp addresses your network device at a generic level

[Bill Wang]

No, icmpping is well an icmp ping (Internet Control Message Protocol)

agent.ping is an application check on a specific port.

icmp addresses your network device at a generic level

So this could monitor the agentd is running or not, cause I found sometimes the zabbix_agentd stopped unexpectedly.

[harmonica]

Use the key agent.ping and on trigger use function nodata

[stelios]

Hi,

I'm using the following trigger:

{mytemplate:agent.ping.nodata(20)}=1

but I'm not getting any emails/alerts when zabbix agent is down.

Any ideas?

[kustodian]

Hi,

I'm using the following trigger:

{mytemplate:agent.ping.nodata(20)}=1

but I'm not getting any emails/alerts when zabbix agent is down.

Any ideas?

How often do you run agent.ping since that trigger will report a problem if that item doesn't receive any data for 20 seconds. If you ping the agent every 60 seconds, you should use nodata(2m) at least.

[lynggaard]

all data stopped?

So this could monitor the agentd is running or not, cause I found sometimes the zabbix_agentd stopped unexpectedly.

I am seeing the same thing with passive agents in 2.0.3, suddenly they stop providing data, not only agent ping but all metrics. However zabbix get command works fine

Is this the same you are seeing