Yesterday we had about over 30 external nodes go into down status in Orion. These Nodes are monitored through the Internet. These Nodes we poll every 120 seconds and collect statistics every 5mins. We couldn't find any issue with our network as to why these Nodes went into Down status they were in down status for about 2mins and then went back into up status. However when you look at the Nodes response time and packet loss yesterday during the outage no packet loss is showin and the response time remains consistent. How can Nodes go into Down status with no packet loss been recorded for the node? Has anybody else seen these? I would think at least some packet loss would show up in a 2min outage.
Just curious: those '30' external nodes are in the same 'external' network ? Do you have more external nodes that did not have that issue?.. To see first if the problem is on your poller-server network or if it is on the 'internet' side, so you have something to compare it.