No data sent to hub/station offline

[prez02]

Hey,
on my local aiscatcher page, under Statistics/System, is say Feeding Hub = Yes, but under "Data Throughput" it says not connected. Looking in the logs, it tries to connected to 185.77.96.227, port: 4242, bu there are always timeouts

2026-07-04 13:37:21.160][aiscatcher] TCP (185.77.96.227:4242): timeout connecting to server, reconnect.

Pinging that IP from a local pc fails. Aiscatcher is runningt via a shipfeeder container as part of adsb.im.

It does work from time to time, though.

Any ideas where to look to resolve this issue?

[prez02]

Ok, so I tried reconnecting to the internet, getting a new IP in the process, and now it is connected to the hub. Apparently the IP was blocked for some reason?

[prez02]

I spoke to soon, after a few hours the connection broke down again and has not recovered since then. Changing the IP does not make a difference, but a router reboot did?

[2026-07-05 02:07:11.252][aiscatcher] TCP (185.77.96.227:4242): send error 110 (Connection timed out). Reconnecting.

[jvde-github]

Bit puzzled why the connection would stop. So, it detects an error and should automatically reconnect which does not seem to happen.

In case there is a failure to connect, does something like this work:

nc  185.77.96.227 4242 

returning

{"status": "ok", "api": "8mVLWwsAawjYr4Rx", "url": "https://aiscatcher.org/hub"}

That would be the test that your local system can reach the server. Also manually restarting the AIS-catcher process or computer does not resolve it?

Otherwise I don't have an idea.

[prez02]

No, nc ... just hangs. I cannot ping 185.77.96.227 from the feeder or any other computer on my network,

Getting a new ip or completely rebooting does not help. Right now I even do not see the recent activity.

[prez02]

Now it works again, the nc command works as well. Activity from the last 7 days still gone.

Still cannot ping the ip, but maybe that is on purpose?!

I only get a few messages, maybe that is reason as it looks like it is offline and then blocked?

[jvde-github]

Good that it works. The activity of the last 7 days indeed got reset, for all stations unfortunately — a Linux security update rebooted the server. Nothing specific to your station and nothing I can restore. Something I still need to make more persistent.....

Ping doesn't work on aiscatcher.org, that's normal, the nc command is the right test. There is no blocking on the hub side and no other stations had issues during your outage. Since a reboot or AIS-catcher restart didn't cure it and plain nc also failed, the problem was not in your setup but somewhere between your ISP and the server, and it cleared on its own.

If it happens again, run the nc test from the feeder and from a phone hotspot at the same time — if the hotspot connects and your line doesn't, that's something to take to your ISP. A traceroute 185.77.96.227 while it's broken would also show where the connection dies.

[prez02]

Stopped overnight again. Ran traceroute, the packet successfully navigates the European backbone but dies immediately after leaving Serverius infrastructure 195.35.53.30, a Dutch connectivity and data center provider. I'll run it again when it works, to see what would be the next hop.

[jvde-github]

Weird, what is your station number (can also email), I will do some digging on the server for any errors?

[prez02]

Hi, 1048 baldeney

[prez02]

I was about to write that it seems to work fine two-ish days with one period of disconnection, but just now it was offline again. This time I tried to reconnect my internet connection until I got a new IP address, which did not help, but the timeout error stopped. I then recreated the container, and it shows as connected again on aiscatcher.rorg.

To me it seems that at some point for whatever reason the traffic or IP is deemed suspicous and is then blocked. Probably nothing one can do about it.

[prez02]

Went offline again after midnight. A reboot of the feeder did not do anything, as did a rebuild of the container. Only the second reset of the internet connection with a new ip fixed it.

[jvde-github]

Interesting.... What was the second reset of the internet? Do you mean the router?

[prez02]

The router menu has a button "reconnect" or „Neu verbinden“. It cuts the internet connection and then reconnects. Most of the time that leads to a new IP and new IPv6-prefix, but not always. It only worked after the second reconnect with a new IP address.

[prez02]

Quick update, so I set up new station on another pi using the install script. I also added the UDP port for the older station. It worked fine for one night, then the -X connection broke, and the UDP connection broke as well.

[prez02]

Now I am confused, the site feeding via UDP is now green and active as well since yesterday evening? Did you change that?
Apart from that, it think it is fair to say that nightly reconnect broke the -X connections. Since I disabled then, both types of connection are running smoothly.