ESP8266 Community Forum

Or at least avoided.

I moved the ESP over to a different wifi network, and the problem went away. It was connected to the internet via an old cisco/linksys 2000 that was just used to extend wifi signal. That router was then hardwired into an internet connected router from verizon fios. When it connected directly to the verizon wireless, or to a completely different network the problem went away.

Poking around on the linksys setting, not knowing squat about networks, I saw a dynamic RIP setting that was off. I turned it on, and now connecting to the linksys is working as well. I'll probably dig a little more to try to understand/verify better, but now that it's working some of that natural motivation will slip I'm sure. Since I shot in the dark, I wouldn't be surprised if the problem came back.

I used the example NTP and had a similar behaviour. I solved this by created a function to update the internal time variables.

Assuming UDPNTPClient is of type WiFiUDP this function calls UDPNTPClient.begin(....) at the start and UDPNTPClient.stop() at the end.

This works reliably.

Also there is no need to call the NTP server repeatedly to get the time. You can use Ticker to update the internal time variables. I find an update every 12 hours is sufficient.

After several hours of frustration looking into a possible code bug I came across this post.
I would never have had the idea of looking into the router defaults for RIP
It now is working after enabling RIP
I thank you thank you!