I'm in what seems like the same boat (at Scottsdale/101.)
Could be worse. I had support asking me whether I was on 2.4ghz or 5ghz after clearly stating, twice, that my computer was plugged directly into the cable modem for testing. =)
I haven't had many issues in the last few years, if any really, but over the last week I noticed that there was sufficient packet loss that the monitoring/auto-reboot functions of my Ubiquiti network were triggering cable modem reboots to try and fix itself. I initially chalked it up to Ubiquiti's latest gizmo being a bit buggy, but it still triggered on the 20th, 21st, 22nd, and multiple times on the 25th. Simple video streaming of Youtube TV became a buffer nightmare, remote access to my home network got sluggish, and I noticed packet loss on continuous pings to google.com and 8.8.8.8. (Losing about 4-8% now.)
There have been zero changes to anything between the cable modem and the outside world (barring anything out at the street, who knows) and it's just been a gradual downward slope into oblivion. Today, I can still hammer out almost 700mbit down, but upstream is 2.41mbit. That's a kinda massive problem.
Just for giggles, I even did a full reset on the SB8200 (little reset button + paperclip) and let it do it's thing. Still managed to bring downstream online at docsis 3.1 (blue light) but upstream remains 3.0 (green light.) I honestly have no idea if thats always been that color scheme, or if upstream should also run blue/3.1 (maybe not since it's only supposed to be 35mbit anyway?) Regardless, the service is pretty much unusable at this point.
After dealing with the *** in chat who couldn't figure out that hardlining via ethernet used neither of the common wifi bands, and trying to sell me on 'advanced support' for $10/mo, I did speak with a support rep on the phone (is it bad that I still have 623-594-1000 memorized from back in the Cox@Home days?) and they actually did show an outage/issue in my area. The Cox app and ensuing sms alerts just before that corroborated the story. Unfortunately that has since cleared and things continue to degrade.
I also took a look at the modem's status page and while I deal with networks and IT for a living, the channels and signal levels it reports might as well be Greek to me. Nothing pops out as an obvious issue, other than the lo screen showing something about some 'dynamic range window violation' repeating a few times. (Lots to see about that on google, but nothing useful so far.)
In any case, packet loss begins immediately after leaving my network, so the issue would appear to be between the cable modem and the upstream connection at Cox. No luck getting past the script-reading support types to actually take action on any of this though.
I also had support asking me to unplug my router after telling them I was hardwired into the modem. Like come on man.
What are our options at this point? Every single skype work call/meeting I have cuts in and out every 5 seconds. Jitter, ploss, and ping spikes are terrible throughout the day and can provide ample proof to back up all of these claims.
Is cox going to actually fix their infrastructure or are we just out of luck? Only other option I have in this area (Gilbert and the 60) is Century Link DSL. I have seen 50+ cox trucks in my neighborhood over the last month so I am assuming my neighbors are experiencing issues as well.
The best option I can see is to organize a campaign around informal FCC complaints. There are a lot of customers out here not getting what they're paying for, and I think plenty of us are fed up.
My post here was to hopefully get some eyes at Cox to look at the case, so I'll be a little more patient in that regard, because just like Thoreau I'm totally over trying to get the folks on the phones reading scripts to get to the root of things. The last time I tried it they wanted to send another tech out so I just hung up.
I'd be surprised if this is anything more than my local area, but en route to a work site to swap out a Palo Alto 820 (so please for the love of god, Cox, don't try to treat me like a novice when it comes to networking?) I got a text message from cox stating:
"The outage is now clear. Are you still having issues? Please try unplugging the power of the device and plugging it back in. Need help? Text AGENT to chat"
So, I went back into my controller and told the monitoring bits to go back to cycling power to the modem if connectivity loss is detected. Proceeded to do some work, and during that time, no alerts about it being triggered... cautious YAY!
Took til now for me to get a spare moment to dig in further, but I just VPNed back to home and the first thing I noticed is that Ping Plotter (glad you posted those screenshots as I had almost forgotten about that very awesome piece of software due to lack of use in recent years!) was showing half the latency to google as prior. The time graph was also showing WAY less red. It's still not perfect, but definitely a vast improvement and probably about what I'd say is acceptable for residential service.
Loaded up ye olde speedtest.net, and hello dolly, 9ms ping, 926.43 down, 30.32 up. still a little lacking on the upstream, but that's getting back up to the 35 target. I'd not be surprised if my area (and any others affected by the same issue that caused my problems) weren't all busily pushing traffic to play catch up now that the freeway is open again, so that could just be a little mid day congestion at work there.
Edit: subsequent speed tests are 859/34 and 902/35, so i think things are actually back to normal... *fingers crossed*
