I'll echo this sentiment. We were seeing a large number of false positives because of lack of SNMP response from some devices even though they were responding to pings without problem. Setting a delay of 1 reduced the number of alerts we were seeing substantially. We do still get a couple devices that will fail both checks, usually right after the discovery run, and we'll get an alarm on that, but it clears itself next poll. Damn Cat6509s and Nexus 1000Vs. ;-)

Sent from my Verizon 4G LTE smartphone

-------- Original message -------- From: "Moerman, Maarten" mmoerman@ebay.com Date: 6/27/16 4:39 AM (GMT-06:00) To: Observium Network Observation System observium@observium.org Subject: Re: [Observium] Delay the alert trigger in case of loss only one ping

-- This message has been scanned by E.F.A. Project and is believed to be clean.