Lost data on dell servers monitor

[walter]

Hi,

I tried to monitor Dell PowerEdge R920 through SNMP on my test environment. I found the graphs is not continues. But my network device monitor looks great on graphs. So I use tcpdump to find connections of udp port 161 from the Dell sever, things like where the blank area on graphs that the Zabbix server didn't send requests.
In my another Zabbix environment, everything looks well, both of the environment use the same SNMP template.

losing data environment:
zabbix version: 3.4.10
2 server: master/slave
2 db: Ver 15.1 Distrib 10.1.34-MariaDB master/slave + elasticsearch

health environment:
zabbix version: 3.4.4
1 server
2 db: Ver 15.1 Distrib 10.1.34-MariaDB master/slave






Thanks

[walter]

After I installed a proxy to monitor the dell server and change the timeout to 15s, the problem solved. So the reason is timeout, I don't know why everything works well on my prod environment, which timeout is 5s the same as the test environment before.

[ingus.vilnis]

Check zabbix_server.log and grep for the exact host name which is failing. Some more clues could be seen that way.

Compare those two instances. Do you have "Bulk requests" configured identically for the SNMP interface on both Zabbixes?

Since you have one server on the healthy environment but a two node cluster in the faulty then maybe it has something to do with the network and addressing configuration?