So, another Quantum article, this one being a little more odd, in that it's a technical issue relating to only the 10gig port of the Q1000K SmartNID, although I suspect the same behavior applies to the C6500XK.

Before we begin, I want to talk about pfSense:

Setting aside the fact I love it, and it's what I deploy for installations, I want to talk about market share, then tie that in with the 10gig port discussion. pfSense controls about 6.2 percent of corporate market share, so about half of cisco, which is still pretty huge. For home installations, market share is much smaller, however, it is huge among tech-savy home users. A large percentage of those folks have multi-gig services, and utilize the 10gig port of their C6500XK(issue not replicated on this hardware), and their Q1000K. For business users, the use of the 10gig port is the most common implementation, as they very often have multi-gig services as well.

So, business customers on pfSense, there are a lot on the 10gig port. For home users, a lot of tech-savy users also sport pfsense installations on the 10gig port, even if they run it at gigabit speeds.

Now I want to talk about gateway monitoring:

For those who don't know, pfSense uses ICMP packets to repeatedly ping a gateway, or another low-latency host, I use their nameservers, and monitor the status of the connection. The size of these ICMP packet conents is called the data payload. By default, pfSense uses a data payload of 1, the smallest possible non-zero/empty payload, which creates minimal traffic.

Now, are you a pfSense admin, or pfSense home user of Quantum Fiber who has recently jacked into their 10 gig ports? Are your gateway alarms going nuts with latency alarms, lost packets and the gateway is being marked offline repeatedly?

Then you are a victim of the 10gig port issue with Quantum's infracstructure, thus the XGS side.

The reason your pings and icmp traceroutes are working, is they default to a payload of 64.

Now here is the rub. For whatever reason, Quantum is refusing to respond the pfsense default payload of 1.

The fix is pretty easy:

This should correct your gateway monitoring timeouts and perceived packet loss.

So, my comment to Quantum Fiber is thus:

For what possible reason would you not respond to all non-zero payload on this side of your network, even though you do on the 1gig/GPON side? This oversight on your part has forced a HUGE market segment of business and home user's to be forced to waste time investigating and correcting this. Way to think it through!