I am not sure if you are asking if there is another way to get current conditions?
Yes there is sort of by addressing the Tempest or devices themselves rather than the station.
I dont know what data points you were missing or if it makes any difference addressing the actual device rather than the station but it is what I do because I have many devices.
It does not return all of the same data or in the same format so it may not suit your purpose.
cheers Ian
Nope, what I’m asking about is an RCA - Root Cause Analysis when something breaks. What I have seen with issues with the API like this is that not only does my personal weather display break, but also the app. It pretty much renders everything useless. So when it broke this Friday, so much data was missing that is expected, the Python script behind my e-ink display just nopes out and crashes. I can fix this with code for a placeholder, yes. Anyhow, at the same time the app was pretty much toast because it also relies on API data.
In short, looking for yeah there was an outage, it’s planned maintenance, etc is helpful to the end user from a transparency and reliability standpoint.
From what I’ve noticed from similar outages recently, their maintenance window for deploying back-end server changes seems to be Friday evenings eastern time.
In my now-former IT career of four decades, we used to have a similar Friday evening maintenance window. It was great for the hourly “A-team” techs, who collected some nice overtime checks when problems with the rollouts happened. It was also least disruptive to our 24/7 user base, since most of them were off for the weekend or out partying until Sunday night. When HR decided to make all of the “A-team” techs salaried with no OT, the maintenance window was moved to Tuesday evening since there was no financial incentive for everyone to ruin their weekends if something went wrong with the rollouts. It also led to better pre-rollout testing and debugging when the OT pay went away.
That being said, the WF-T public/customer visibility of change control, version info, and release notes has not been very good for a few years now IMHO…
Everything is dead again. Confirmed this with my own as well as my father’s Tempest. The app is showing offline and our weather displays are failing on missing data. I’ve reached my point of motivation to code checks for the data before attempting to write to the screen.
Just came back again for now. Can someone please speak to this? There is a weather outbreak currently and this is an absolutely horrible time to lose all data.
Most quality server setups for many years now have rolling deployments and no downtime. At a minimum they also announce their maintenance windows, when the work (outage or slowdown) has started, and when the work has been completed.
