Posting to webhook failed: The underlying connection was closed:

[Horse]

So I broke my fundamental rule of never upgrade multiple things unless you have to for interoperability.

Last week MS installed an update on my WIndows 10 Pro machine that hosts HS4, after the reboot I was greeted with a HS4 update available so I installed HS4 Pro 4.1.15.
While I was at it I thought I'd check Node Red install for updates as its been a while. Updated NR to 1.3.2 and the node-red-contrib-homeseer to 1.06.
Everything is working as it was before, until I checked the log and I get

HomeSeer      3
Posting to webhook failed: The underlying connection was closed: A connection that was expected to be kept alive was closed by the server. URL: http://192.168.1.6:1880/homeseer/webhook

From the HS4 windows machine I can access Node Red and everything still seems to be updating. I have control of devices and status works, just lots of these messages in the log that I can't seem to associate with what is trying to be posted and why I get 4 of these messages in a row, with no device updates received that should trigger a update to NR.

Anyone have a clue where I can look?

[Krumpy]

Open a web browser and enter the following below on the address bar. You may also want to do this from your HomeSeer machine and a different machine. What happens?
http://192.168.1.6:1880/homeseer/webhook

Remember that this is web technology. So, standard web troubleshooting applies. In this case, if I recall correctly, NR posts to the HomeSeer webhook service. The issue is therefore on the HS side. Firewall? You can also try to telnet to 192.168.1.6 port 1880 from different machines.

[Horse]

Krumpy, Yep that give me the expected can't GET to webhook message and II tried turning off the Windows FW just in case that changed something. As I mentioned, the NR integration all seems to be working just fine, all devices status updates, and control works. I just get errors in the HS log.

[kenm]

I think there's a little confusion here. http://192.168.1.5:1880/homeseer/webhook is the NR server, not HS4's listener port. You can't do a GET on that port.

Horse - How often are you getting this error message? In my experience, it's been caused by NR crashing and then being restarted. Are you running NR with some kind of a launcher (or systemd for Linux) or as a Windows service that auto-restarts?

[Horse]

kenm Correct the url is NR and doing a GET results in

Cannot GET /homeseer/webhook

which I expected, but at least tells me http TCP 80 is open and reachable. The errors are currently coming in every minute of so, haven't yet seen a pattern. NR is running in Docker and there are no errors in the NR container and its not crashing. I have copious amounts of event logging to see issue. As mentioned, everything is working from what i can tell, I have numerous dashboards with temp, humidity, PM2.5, VOC, CO2, pump and water alarms etc. I just get log messages in HS. Next step is to packet capture and see if there is any additional information I can glean.

NR all working, no messages
HS4 all working just log messages "From Homeseer Type 3"

[Horse]

To update and close this.
HS was having periodic issues with the TCP connection to NR:1880 as seen with WireShark TCP retransmits, but ping to the host was 100%. Other Docker containers running on the Docker Host VM were not having issues, however still looked like a docker network issue. Rebooted the Docker VM and the network connectivity issue to the NR container disappeared. Previous restarts to just NR container did not help.
Thanks for those that made suggestions, all very weird when it started, and why.