meteohub.de

Posted: **Sat Aug 30, 2008 3:48 pm**

DEUTSCH:

anbei ein experimentelles Update mit ein paar kleineren Änderungen und Bug-Fixes.

Link: http://www.meteohub.de/joomla/index.php ... &Itemid=29

Änderungen gegenüber V2.7:

* Behebt eine Fehler bei der Auswertung der Regenrate der PCR800 Sensors beim Empfang via RFXCOM.
* Behebt die Vertauschung des Normaldrucks und des Stationsdrucks beim Auslesen der Daten der WMR-100.
* Fügt einige mathematische Operationen zum HTML-Template Ersetzungsprozess hinzu. Details finden sich im Handbuch (Abschnitt 3.4).
* Fügt die Variable "actual_lunar_phase_es" mit spanischen Texten zu den Mondphasen hinzu.
* Fügt die Variable "actual_date0_dayofweek_local" und "actual_date0_dayofweek_utc" mit Anzeige des Wochentages (0 = Sonntag) hinzu.
* Stellt sicher, dass beim Einspeisen von Daten in WD-Live die tägliche Regensumme <= der monatlichen Regensumme <= der jährlichen Regensumme ist.
* Behebt einen Fehler bei der Anzeige der Regenmenge der letzten 7 Tage und der letzten 12 Monate in WD-Live.
* Fügt "bartrend" Informationen für die Speisung von "hetweeractueel.nl" hinzu.
* Behebt einen mit 2.7 eingefügten Fehler bei der Übermittlung von Solarstrahlungsdaten an Wetternetzwerke.
* Enthält experimentellen Code zur Neuinitialisierung der USB Schnittstelle bei potentiellen USB Vantage Verbindungsabbrüchen.

-------------------------------------

ENGLISH:

here is an experimental update with some minor changes and bug-fixes.

Link: http://www.meteohub.de/joomla/index.php ... &Itemid=29

Changes since V2.7:
* Fixes a bug on recording PCR800 rain rate when used in RFXCOM mode.
* Fixes an unwanted interchange of local pressure and sealevel pressure when reading data from WMR-100.
* Adds some math capability to the HTML-template replacement mechanism. Details are explained in the manual.
* Adds variable "actual_lunar_phase_es" which contains Spanish descriptions of moon phases.
* Adds variable "actual_date0_dayofweek_local" and "actual_date0_dayofweek_utc" to display day of week (0 = sunday).
* Makes sure that when feeding wd live, daily rainsum <= monthly rainsum <= yearly rainsum.
* Fixes a bug in display of total rain for last 7 days and last 12 months in WD live.
* Adds "bartrend" information for reports to "hetweeractueel.nl".
* Fixes a bug introduced with 2.7 that inhibits reporting of solar radiation data to weather networks
* Includes experimental code that should help Meteohub to recover a connection drop from an USB Vantage.

Posted: **Sat Aug 30, 2008 6:02 pm**

Hi Boris,
welcome back.

Just applied 2.7a and ran into severe problems.
meteohub uploads my left control frame called links.html every 15 minutes, because that prints in the extrema of the day.
Since 2.7a the html code processed and uploaded is cut off nearly of all its content.

I guess the reason is the newly added operators {}
They seem to be in severe conflict with raw HTML code that uses the "Gescheifte Klammern" also.

The code for my links.html page is given below.
With 2.7a every text following BlurLinks() is cut off when processed and uploaded by meteouhb to the ftp-server.

This new operators are great, but as long as they are part of basic HTML-syntax this is a problem, I think.

Cheers,
Wolfgang
___

<script>

</script>
<link rel="stylesheet" href="format.css" type="text/css">

</head>
<body bgcolor="#2d5770" text="white" topmargin="0" leftmargin="0" marginwidth="0" marginheight="0" scroll="auto" >

<table border="0" cellpadding="0" cellspacing="0" width="100%" >

<tr>
<td width="20" > </td><td>

<table border="0" cellpadding="0" cellspacing="0" width="100%" >
<tr>
<td>
<br>
» <b>Sonnenstand und Mondphase</b>
<br> <br>
<center><font color="#f25f1f">Sonnenaufgang: <b>[actual_sunrise_standard_local] Uhr</b> - Sonnenuntergang: <b>[actual_sunset_standard_local] Uhr</b></font></center>
<center><font color="#f25f1f" size"-2">Morgendämmerung: [actual_sunrise_civiltwilight_local] Uhr - Abenddämmerung: [actual_sunset_civiltwilight_local] Uhr </font></center>
<center><font color="#f25f1f">Länge des Tages: <b>[actual_daylength_standard_hhmm] h</b></font></center>
<br><center><font color="#f25f1f">[actual_lunar_phase_de] - Sichtbarkeit:[actual_lunar_phase_percentage] %</font></center>
<center><font size="-4">(0% = Neumond - 100% = Vollmond)</font><br><img src="http://www.myway.de/honsolgen/moon.png" width="50" height="50"></center>
<center>
<font size="-3"><a href="http://de.wikipedia.org/wiki/Sonne">Sonne</a> &
<a href="http://de.wikipedia.org/wiki/Mond">Mond</a > @ Wikepedia</font>
<br>
<img src="http://www.myway.de/honsolgen/suncal.png"></center>
</tr>
</table>
</td>
<td width="20" > </td>
</tr>
</table></tr></table>

Posted: **Sat Aug 30, 2008 6:32 pm**

Just went back to 2.7 and the bug went away.
So, it´s definitely 2.7a and the new operators changing the raw HTML Code during the upload process.

Posted: **Sat Aug 30, 2008 10:48 pm**

Not sure if this is 2.7a related, but I don't recall seeing it before...

Just after installing 2.7a I deleted some readings for th3 from earlier this afternoon when I had been carring out an experiment with the sensor.

I almost expected to be told that a recompute would be done, but I didn't see such a message and the System Information page showed no evidence of a recompute.

A bit later I noticed that the time scale for recent records on my WD graphs was looking very compressed, and that WSWIN was way behind with its sync. Checking in WD I saw that the logging protocol was only responding intermittently. Checking in the MH export folder I saw that the import.csv wasn't being written as often as usual.
Checking System Information the System Load:was showing figures of 9 and above!

I stopped WD and WSWIN, and the clientraw uploads.

4 and a half hours since 2.7a install now. System Load is now down to low figures, so I've restarted WD (I want to be sure it is running at midnight - if not some values don't get computed).

My guess is that maybe a recompute was going on? But System Info => Process => Weather Data Recomputation has simply stated completed the whole time.

Graphs for recent time scales, e.g. today still only show straight lines for times prior to around about the time of reboot / record deletion. Maybe the recompute is not yet complete?

Should a delete cause a recompute (I'd not be surprised if it did). If so should System Info show that the recompute is happening?

Or was it maybe something else entirely?

Posted: **Sat Aug 30, 2008 11:08 pm**

skyewright wrote:Graphs for recent time scales, e.g. today still only show straight lines for times prior to around about the time of reboot / record deletion.

Just done some further tests. It appears only to be graphs that use '5 minute time buckets' that have these straight lines!

Maybe tomorrow I'll try doing an intentional Recompute...

Posted: **Sun Aug 31, 2008 12:35 am**

It doesn't seem like this fixed the solar radiation issue with WUnderground.

Posted: **Sun Aug 31, 2008 1:25 am**

@wfpost: checked your code, Meteohub does not process this right. Bug will be fixed with 2.7b

Posted: **Sun Aug 31, 2008 1:39 am**

Okay... Thanks! John at WU support sent me one of the strings that my station is sending in and the solarradiation argument isn't there. If it helps:

Hello,

Here is the raw text being sent from meteohub to our server:
/weatherstation/updateweatherstation.php?action=updateraw&ID=KLAALEXA5&PASSWORD=xxxxxxx&dateutc=2008-08-30%2023:19:53&realtime=1&rtfreq=5&winddir=34&windspeedmph=0&windgustmph=8&windspdmph_avg2=1&humidity=56&tempf=89.2&rainin=0.00&dailyrainin=0.00&baromin=29.88&dewptf=71.4&UV=0.0&indoorhumidity=0&softwaretype=meteohub
HTTP/1.0

See how every variable is listed (I hid your password)? But,
solarradiation is missing.
...
-John

Posted: **Sun Aug 31, 2008 1:42 am**

@bsparks: I removed one more bug.
Unfortunately, I don't have a vantage with solar sensor
to test the changes. Please give 2.7b a chance, which
will be released shortly.

Posted: **Sun Aug 31, 2008 2:42 am**

No problem. Meteohub is recognizing the sensor and is showing its value, but it just isn't getting to Wunderground. Will install update as soon as it is available and let you know. It's getting dark here, so I might have to wait until the morning to see if the update fixes the issue.

Posted: **Sun Aug 31, 2008 3:06 pm**

skyewright wrote:Maybe tomorrow I'll try doing an intentional Recompute...

A full recompute (4hrs 20 minutes) has sorted this, so probably not 2.7a related. :)

(but still nice to know if a delete does cause a recompute)

Posted: **Sun Aug 31, 2008 7:32 pm**

no, it does not.

Posted: **Sun Aug 31, 2008 8:22 pm**

admin wrote:no, it does not.

Strange, because something very processor intensive was happening for around 4 hours followinng the install. :S

May or may not be related but I've (just) discovered a problem with logging protocol requests (about to post about this in the Meteohub Application thread...).

Posted: **Sun Aug 31, 2008 8:42 pm**

that might be the initialization process for the
http datalogger, that goes with low priority once
over all weather data after each reboot.

meteohub.de

Meteohub Update 2.7a (experimental)

Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)

Re:Meteohub Update 2.7a (experimental)