Not entirely sure still if this is something on my end, but I wanted to make a more formal report and follow up.
In the past few update threads I had mentioned it, and maybe one or two other people popped in to say they saw similar experiences, but I don’t believe it’s widespread.
I haven’t been able to track really when these disconnects happen, but they are only short enough to not allow messages to come / go, but I can send and see ride ons being sent / received (unless that’s just a visual thing on my end). Messaging capability usually returns within ~20 seconds or so; sometimes messages received over that time do appear, but not always; and no telling if a message I sent is actually sent or not.
What I noticed last night most importantly however, is it seems to be tied to number of visible users / data collection. Typically when riding with a pace bot, or on a larger group ride with 50+ people, I will have these issues maybe once every 10-15 minutes, almost guaranteed from my mental tracking.
Meanwhile, as a group ride started to decrease over time yesterday, the issue stopped happening completely.
I haven’t changed anything, I’ve tried making some packet buffer size changes on my network card to no avail, and have reset everything; but nothing seems to be fixing this issue that wasn’t present a month ago.
After uploading my log from yesterday, which ran across a few rides on pacer bots, and two group rides of varying sizes, I can see high spikes in UDP packets total while with the pacer bots; upwards of 1400 per minute according to Zwiftalizer, combined server to client (1050) and client to server (300). Towards the end of my last ride however when things got real quiet and only a few riders left, that number dropped to a baseline of maybe 600, at 300 both ways.
Zwiftalizer mentions zero transmission or reception errors, but, it does show disconnects in TCP network disconnects; although oddly not all of the “connection disconnected” occurrences.
Regardless; is this issue on my end that’s popped up seemingly out of nowhere?
Is this a global issue?
Is anyone else having this issue as of more recently?
I’m still trying to decide if this is a change on my end or not.