SNMP availability and device upTime item having delay response in Latest data!

[Vijay.kali]

While I'm testing for snmp status for devices in zabbix, SNMP availability is showing Up(1) even after device is down with ICMP status Down(0). I've also tested with snmpget command and it shows connection timed out while Latest data showing it as up(1) and the last check is updated for every 15s as configured in item interval.

I'm testing on Cisco router by reloading the device, it's taking 8 minutes for device to come back up(long enough). So when the device is reloaded the ICMP status shows down but SNMP availability shows Up.
After few minutes the SNMP availability is updated as Down(0) with Device upTime showing last check value when I reloaded device. Now after 8 minutes when router is up ICMP staus reflected instantly but SNMP again taking some time to reflect it.

Let me know if any other details require to analyse the issue.

Thanks!

[Hamardaban]

Most likely you have encountered the operation of this process https://www.zabbix.com/documentation...unreachability

[Vijay.kali]

I've checked your link and my zabbix configuration, it has default setting for unreachability. So as per the default settings the UnrechableDelay is 45 seconds the device should go into Unavailable host or in our case snmp availability should be down in max 60 seconds.

But here it's taking more than 5 minutes to show snmp down. It's a critical network we configured for and customers are asking for instant status or it should be within 15 seconds.

Please let me know if you need any further details.

[c1support]

What does it say in the zabbix server log?
According to https://www.zabbix.com/documentation...unreachability, we would expect something like

Zabbix agent item "system.cpu.load[percpu,avg1]" on host "New host" failed: first network error, wait for 15 seconds
Zabbix agent item "system.cpu.load[percpu,avg15]" on host "New host" failed: another network error, wait for 15 seconds

--> please also note the timestamps
--> if changing the debug level in zabbix_server.conf, please make sure to restart the service (https://www.zabbix.com/documentation.../zabbix_server)

[Vijay.kali]

I've figured out the issue now, it was due to high number of unreachable devices in zabbix environment which lead to 100% Utilization of Unreachable pollers. After disabling the unreachable devices(almost 50% of hosts were unreachable) the Unrechable poller utilization decreased significantly as show in screenshot. After making changes SNMP availability was quick enough.


Also updated zabbix configuration by increasing Unreachable pollers count, now after enabling all the devices the utilization of Unreachable pollers was about 60% and SNMP availability is responding as configured.

Thank you all for your support!

[Hamardaban]

This is the proper way!