I have been running continuous network monitoring using Obkio’s cloud agents on a site that routes internal traffic through VLAN-tagged subnets. The issue I am facing is that Obkio occasionally reports intermittent packet loss (1–5%) from cloud agents to local devices; but when I ping the same endpoints manually / via third-party tools; I see no loss at all.
I suspect it might be related to how tagged traffic is handled on our switches or firewalls, as some monitoring paths involve policy-based routing or NAT. However; Obkio is supposed to run at the network level; so I’d expect consistent results. Could the discrepancy be due to how probes are prioritized (e.g., low QoS marking or ICMP deprioritization)?
I’m Sam, a technical point of contact for Obkio’s users.
Regarding intermittent packet loss (1–5%) from cloud agents to local devices, which doesn’t appear when manually pinging or using third-party tools.
Obkio’s monitoring sessions primarily use UDP packets, while tools like ping rely on ICMP packets. It’s common to observe packet loss on one type of traffic but not the other. You can compare both UDP and ICMP (via traceroutes) within the Obkio app to see this firsthand. Therefore, this discrepancy could be normal.
Regarding whether discrepancies are due to probe prioritization (e.g., low QoS marking or ICMP deprioritization).
VLAN tagging, policy-based routing, or NATing our UDP packets should not affect the accuracy of the results and are common practices among Obkio users.
I suspect that the issue might be impacting only UDP traffic. This often occurs when firewalls or security equipment apply filtering or rate-limiting to specific traffic types.
I hope this is useful. Feel free to contact us via our support chat or email at support@obkio.com for specific assistance with your Obkio account and settings.