Hello everyone, I have tried searching for this but can’t find much, I think this is relevant to the correct tag/category.
I am looking to move away from a single telegraf instance polling my network devices but want to try to avoid 2 instances polling the same network device. I can obviously split the SNMP configuration between the 2 instances so that telegrafA polls 50% of the network devices but I’d rather keep configurations centralized and not have to do this.
The main disadvantage I can see if I did go that route, when 1 instance is down for whatever reason (eg. maintenance) then it will stop collecting half the SNMP data for my network.
I also run the gNMI telemetry input, I believe I can easily tackle the scaling for the telemetry input with a local load balancer but these telegraf instances will not be in the same local network, so I may need to come up with a few ways to approach the . I am just wondering how other people have approached this scenario?
