Curious that enabling rapid-fire improved reliability for you. I was having 
reliability issues posting to WU and was told that having it enabled led to 
problems. So I disabled, and things have been consistently good for me over 
the past month (i.e. since that change).

On Sunday, August 7, 2022 at 1:53:44 PM UTC-4 Doug Bo wrote:

> Another data point, my station is missing some location info and 
> barometric pressure on the Android app.  I have verified my upload data on 
> the WU station web page and everything looks good.  One thing WU "lost" is 
> my hardware type.  I have reset the station hardware info so we'll see what 
> happens.  I have also opened a ticket with them on this issue.  
> FWIW, a few years back WU moved some PWS stations to a new server and 
> messed up my station.  I wonder if they have performed some data house 
> cleaning and mucked up a few things?
>
> I'll update the thread when I have additional info.
>
> Thanks all for the input. 
> Doug B.
>
>
> On Sunday, August 7, 2022 at 9:02:10 AM UTC-7 [email protected] wrote:
>
>> I might have a hint about this. For quite some time, my VP2 (KAKWASIL2) 
>> was one of very few PWS in my area. I had the same issues you described, 
>> where it would disappear from the map but when viewed directly it would be 
>> uploading. I went through all the troubleshooting steps with WU, and we 
>> essentially came down to the sanity checks - if your weather is too far off 
>> the "normal" reports then it's considered "bad" and not displayed until 
>> "corrected". Thing is, my station is close to town, while the "official" 
>> airport weather is several miles away and across a small set of hills, so 
>> can be quite different. With no other reports to compare to, mine was 
>> considered "off".
>>
>> As more stations came online, and more data was available, mine became 
>> "more reliable" since more stations showed similar data. RapidFire adds 
>> more data into the mix, so if your data is consistent and within certain 
>> limits, then it's considered "more reliable". Also, if there's other PWS in 
>> the area and all show similar data it helps. Mine would drop for a day or 
>> two, then suddenly you'd query for zip 99654 and it would show as the 
>> official town weather again :-)
>>
>> WU algorithms are known to be a bit flaky with sanity checks, but have 
>> gotten much better over the years as more data has come online. Still, if 
>> you're in an area with somewhat different temp/wind patterns, it can cause 
>> WU's AI some headaches.
>>
>>
>> On Saturday, August 6, 2022 at 11:21:38 AM UTC-8 Alastair L wrote:
>>
>>> Hi
>>>
>>> Many thanks for all your responses and suggestions.  As a result I 
>>> believe I can report success, my station now appears reliably in the WU 
>>> maps.
>>>
>>> I checked out the time on both my Pi and the Vantage console against my 
>>> PC time and all were within seconds of each other.  So I tried I editing 
>>> the rapidfire setting in the config file as suggested and changed it from 
>>> false to true.  For the first time in ages my station (ILAIRG10) appeared 
>>> on WU maps and  continues to appear - great!   I've waited some hours 
>>> to confirm this is not just a fluke, and so far the station appears every 
>>> time I have I've checked.  This hasn't been possible in months. I have 
>>> wasted a lot of time fretting over this problem and it's great to have a 
>>> solution.
>>>
>>> I can see from the weewx log that weewx now sends data to WU every 2 
>>> seconds now instead of  every 5 minutes previously and that the WU data 
>>> table still updates every 5 minutes.  Curiously the time stamp on my WU 
>>> data  table changed as soon as the rapid fire option was made, from e.g.  
>>> 2:00PM, 2:05 PM, 2:10 PM etc to  e.g 2:34 PM, 2:39 PM, 2:44 PM etc.  These 
>>> post rapidfire change time stamps now precisely align with the other two 
>>> local stations (that I believe not running on Weewx), but what the 
>>> significance is escapes me.  The comment made regarding WU dropping 
>>> sites from the map if they aren't uploading reliably looks to have been 
>>> true for me, but why, to my likely simplistic understanding, firing data at 
>>> WU 150 times in 5 minutes, instead of a nice sedate once, improves 
>>> reliability surprises me.  I'm not really clear what the logic behind 
>>> this is but it works for me.  Perhaps someone can suggest an explanation?
>>>
>>> Anyway, I'm happy now and thanks once again to you all for your input. 
>>>
>>> Alastair
>>>
>>>>
>>>>

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/weewx-user/21cdd755-e20a-4efd-8729-9cd144126d4fn%40googlegroups.com.

Reply via email to