System is unreachable - several false alaerts

[Michael OT]

My assumption is that the hardware status polling is using ICMP ping to check this? Polling is set to 5 minutes with ICMP ping timeout 5 seconds with two retries. Checking the box post alert we find that it's been up for days.

Does extending the ICMP ping retries cause the retry to occur in the same polling or the next one(s)? Should I shorten the hardware status polling interval at the same time? (96 servers monitored). What are other people using for these ICMP ping setting to address a WAN that has a once a day slight dropout? I am trying to avoid delaying real alerts as a priority but want to reduce false alerts >50% or more.

[Rob Buxton]

You can shorten the scheduled interval - which means servers unreachable are detected earlier.

The timeout / retries can be adjusted to make HPSIM take longer to flag there's an issue. You could do this per server for just those servers on the WAN that are causing the issue.
You could experiment with the server specific settings to find a suitable match.