In their wisdom, the NWS has changed the point-printable forecast page that the old advforecast2.php V4.05 script used to swap the rows of temperatures with conditions, so this old script produced odd display results.
I've released V4.06 to correct that issue for anyone still using the OLD script.
The advforecast2.php V5.19 (current script) is unaffected as it uses api.weather.gov for data (and that hasn't changed).

If you need the V4.06 script for your site, download from the script page near the bottom of that page.  I do urge folks using the old script to convert to the V5.19 script instead as ongoing support for the V4.x script will be limited.

We are using dozens of Barani Gen III shields in Poland in various locations. We have spent thousands of euros on their product.  In some time we will show charts here on the forum that will shed light, because this shield is not perfect and has a lot of flaws that the Barani company does not talk about. We did compilations and comparisons throughout the year, the Slovaks advertised this as a replacement for the excellent Stevenson replacement, which we do not agree with and are entitled to this opinion.

There were comparisons with Apogee TS100 and Davis FARS24H, as well as Stevenson's WMO-compliant cage and SHT35 and PT1000 sensors. Comparisons were also made by a colleague who had several years of experience at state weather and hydrology service stations and saw quickly where the Barani Gen III was poor.

I myself had a Barani MeteoWind and it failed after 6 months of operation. The Barani company refunded the money and paid shipping both ways. I will not buy any more wind gauge from Barani company. I think their company uses too much marketing in their products.

All,

I currently have a rather simple setup to measure rain: a cheap tipping bucket gauge where I just count the tips and multiply them with the bucket size to get rain amounts. That results in a running total that I write into an influx database, from where I derive things like daily/weekly/monthly/yearly totals and the rain rate. I feed that total into weewx via mqtt, although while the hourly and such rain amounts are shown, I recently noticed that the running total (about 5000mm so far) doesn't make it into the database. Maybe the number is too high?

In any case, I'm considering moving to a better rain gauge (current placement isn't optimal) such as an Ecowitt WH40 (I already have the GW1100 gateway).

From what I gathered that gives me instant rain amounts, and also summarized totals. With that, the data stream would be reversed (GW1100 to weewx and from there via mqtt to influxdb). Is there a way to manage my running total in weewx? All I need is to store that total (5000+ mm) in the database and have any rain falling added to that number. Is that possible? I suppose that weewx would then derive hourly/daily etc. totals from the difference, as it does now.

Thanks!

Yeah, I read that. What I didn't realise is that the % value would be so different between sensors with the same calibration. Basically it seems to suggest that there is some of factory calibration, as one would expect a higher AD value to result in a higher % than another sensor with a lower AD (all things remaining the same as far as calibration as seen in WSView+).

It's not a big deal, I just found it weird.

Not to drift off-topic here, but I do wonder if there is a problem with the console.
Looking at the console picture, I never seen the exclamations points in colored background for temp / wind / indoor temp / barometer. 
Nor have I ever this this mentioned before.

Mind reading the WiKi (see signature) ? Read under the HP2551 main display/interface pictogram explanation - point 13 and a few paragraphs below