Some ideas / feedback after long haul travel for discussion

[mschmiedel]

I had the chance to do a longer haul travel with my ZS EV, and I had the gateway running for a trip of over 3800km without major issues (I did not need to restart the gateway at any time).

Some of the issues I had:

1. the first day of our trip, the SAIC API was failing and the charging state would not refresh. This causes the APRP integration not to work. Since the only mandatory thing needed is the SOC, and this can be read from the VehicleState (with less precision), I would propose to modify the gateway / ABRP integration to be able to use a fallback in case the Charging State is absent. If you think this is a good idea, I would add MRs for this feature soon... Also the Gateways starts polling a lot on the vehicle which could cause issues with the 12V battery, handling this case seems like a good idea...

2. I also noticed that with my recent changes (refreshPeriods), when the gateway is in "once a day" refresh policy, changing things like locked status or turning on the AC will work fine, but new status is not pushed back to mqtt, causing for "wrong" states, in my case in HomeAssistant. I will open an issue for this and try to fix it.

3. When driving across country borders, the car will loose Internet connection for up to 10-15 minutes. In my case it did not cause any troubles, ABRP just started getting status a few minutes later. I am just wondering if at some point the gateway could interpret the offline car status as "engine off" and enter "once a day" refresh policy? We should verify this scenario to make sure the gateway will not stop polling on a running car if it exceeds the grace period...