It should work the same as before. Have you checked if the port is open and accessible? Have you checked troubleshooting guide?
I deploy Traccar via a templated systemd unit (using Ansible), and the only thing changed from 4.12 to 4.13 was the Docker image tag - the firewall rules were not touched, the port forwarding via the Docker bridge stayed the same. Yes, my setup is completely automated, and i can guarantee that the Docker image is the only things changed between deployments.
Yes, it used the port checking tool, and for 4.13 it still says "Port is open" - 5027/tcp, btw.
So i got myself another Teltonika FMB920 because i don't want to rip the one installed in my bike out, just to test at home.
And i have the exact same behaviour as described above!
With the 4.12 Docker image and port 5027 open, everything is fine. I stop the container, change the systemd unit's tag to the 4.13 version, start that new container, and the tracking stops. Switching the tag back to 4.12: the tracking works.
The only strange thing i noticed are log entries like this:
2021-08-05 19:50:10 INFO: [2209e4d0: mavlink2 < 172.17.0.1] HEX: 5f6c6f6769746563682d726576657273652d626f6e6a6f75722e5f7463702e6c6f63616c2e0a3434333239
mavlink2 is on port 5224, which i DO NOT have opened/forwarded to the Docker container (only 5027, 5055 and 8082).
I could stick to 4.12, but i think it would be a better idea to find out why 4.13 doesn't work with my Teltonika devices anymore.
Just to make sure, i've also tested with the Traccar Client app and that one works fine with 4.13 (but on port 5055, of course).
So it's only Teltonika on port 5027 that is being "ignored" by 4.13.
Alright, i don't know why, maybe using the Traccar Client app unclogged something, but all of a sudden the Teltonika device also comes back online.
I guess for now this is solved then. ;-)
I've been using a Teltonika FMB920 in my motorcycle for a couple of months now (even with OBD2 data via Bluetooth, pretty cool!), but when i upgraded to 4.13, that FMB920 did not come online again. I only noticed after a couple of days, so first thing i did was revert back to 4.12 - and the device came back online and submitted all the packets it had buffered.
While running 4.13, everything looked normal. The webinterface worked as expected, there was no funny output in the logs, it just looked like the FMB920 was not even trying to connect.
Were there any changes that could have caused this?
btw, i am/was running the official Docker images, "traccar/traccar:4.12-alpine" and "traccar/traccar:4.13-alpine"