Again telnet connection problems

[anderl78]

Hello!

I'm having telnet connection errors again. I've been using @skrue's fix for a while now and haven't had any problems since this fix. Which is very good! But sadly this is over now, i get around 1 or 2 connection lost a day (only fix is a reboot of HA)... I suspect the new firmware V14.0HAR.p1 is causing this. Has anyone had similar experiences? I'm using the latest Homeassistant 2024.9.2....

At the moment I'm playing around with longer timeouts for the connection and for the update frequency of the sensors. At the momentI can't say if this really helps. Other thing I'm thinkingabout is to make the telnet connectionmake even more robust , perhaps with complete stop and restart the connectionif errors occur - or something like that....

[anderl78]

Ok... Since changing the timeouts, I haven't had any connection problems for 3 days. I therefore suspect that the regular timeouts are set relatively tightly and that this may cause problems depending on the system, network, etc. Slightly longer timeouts should not be a problem for functionality and may make the connection a little more stable. I will therefore not make any further changes at the moment, but I will continue to monitor this issue.

For documentation purposes, I have changed the following parameters (all values ​​are seconds):

const.py: BRIDGE_TIMEOUT = 2.0 --> 10.0
sensor.py: SCAN_INTERVAL = timedelta(seconds=5) --> 30
hargassner.py: SCAN_INTERVAL = timedelta(seconds=5) --> 30

[anderl78]

OK, I've tried it for a while now. I have disconnections every now and then every few days. A repair is only possible by restarting Homeassistant. I have therefore extended the code again to catch this error and write additional LOG messages. It has been more stable for a few days and has been running without any problems so far. I'll check it out a bit more and then post a pull request when it's stable...

[Michael1101W]

Hi there,
is there any progress on the pull request? I also have irregular disconnections. Sometimes 2-3 weeks no disconnection, then two days in row. I changed the timeouts and scan interval as suggested. This helped certainly, but not fully solve the issue.
Thanks!

[simonputz]

@Michael1101W I made a fork from this repo , since there was so much re-work I had to do to get it working with my old HSV14 . Currently still on a local branch , testing. But since my set up is a bit different, I actually had to wire RS232 to the heater to get the data out, I also had to deal with and cover lots of serial as well as tcp connection scenarios (I use a small ESP32 to do RS232 to TCP server conversion at the heater and then pull that similar in the way @TheRealKillaruna code works for the more modern heaters.

I'll put it up on Github as soon as I can, maybe there is a fix in the connection handling parts of the code that can be easily merged.

I don't know how much time and energy @TheRealKillaruna has to maintain the current code base, but there is plenty space for improvement to make this more compliant and robust with HA, and I do intend to make my fork available as proper HACS following HA standards as well, but I'd focus a bit more on the older heater models as this is something I can test locally. But if I have time, I would also not be hesitant to refactor current code base a bit. The labeling mechanism of variables etc. is something you guys with the modern systems don't really have to deal with (as the heater exports a field config) , but it was super annoying for me as I had to put every single thing in manually into that XML structure.

[Murazor]

Hi there, I've downloaded and put in my HA the repo from Andler78 and moved from every day telnet connection error and lost of communication until HA reboot to some errors on telnet connection but as soon as possible the integration get back in connection with the Nano_PK.
I think for me the errors are from some issues on LAN because I get some other stuffs that disconnect from Internet/HA.