Update 4.7j

[admin]

Update to Meteohub V4.7j (NSLU2, x86, SheevaPlug, DreamPlug) brings these features:

• Fixes a bug in Rainwise MK3 initialization, which causes stations not pre-configured into "automatically send data" mode (":A") to stay silent. Bug fix makes sure to configure the station in the right way at the beginning.
• Fixes a bug in template replacement mechanism that occurs on empty square brackets "[]" (replaced to empty string) and on square brackets having non-alphanum characters and a colon inside, like "[^\d\:]" (replaced to empty string).
• Adds a couple of system variabes (with example values attached, units and language are taken from dashboard settings):
  actual_system_version_text 4.7j
  actual_system_version_num 47
  actual_system_build_num 5315
  actual_system_platform_text fit-PC_Slim_(500_MHz)
  actual_system_language_text German
  actual_system_temp_unit f
  actual_system_hum_unit rel
  actual_system_press_unit hpa
  actual_system_wind_unit ms
  actual_system_rain_unit mm

Update Link: http://www.meteohub.de/files/update-v4.7j.upd

[admin]

After a long sequence of added features and bug fixes for version 4.7 I would like to bundle al these into a new release, including base images for installation on fresh devices.

If you experience bugs with 4.7j, please let me know shortly in order to fix these before freezing code for upcoming 4.8 release.

[wfpost]

Two things, but existing before 4.7j

1. Printing into my website the average temps for each months back the last 13 months, the variable
seqmonth1_th0_temp_c@13
starts to grow around the middle of the month. Which is weird because, this value is a fixed one giving the average as it was one year and one month back in time. At the beginning of the current month it was 11.7°C. Now it is 12.1°C

Recalculation is not the problem, as I saw this months ago, and re-calc does not fix it.
Up to seqmonth1_th0_temp_c@12 everything is fine, but the 13th does grow, which is wrong.

<td><b>[month1_th0_temp_c]&deg;C</b></td>
<td>[seqmonth1_th0_temp_c@1]&deg;C</td>
<td>[seqmonth1_th0_temp_c@2]&deg;C</td>
<td>[seqmonth1_th0_temp_c@3]&deg;C</td>
<td>[seqmonth1_th0_temp_c@4]&deg;C</td>
<td>[seqmonth1_th0_temp_c@5]&deg;C</td>
<td>[seqmonth1_th0_temp_c@6]&deg;C</td>
<td>[seqmonth1_th0_temp_c@7]&deg;C</td>
<td>[seqmonth1_th0_temp_c@8]&deg;C</td>
<td>[seqmonth1_th0_temp_c@9]&deg;C</td>
<td>[seqmonth1_th0_temp_c@10]&deg;C</td>
<td>[seqmonth1_th0_temp_c@11]&deg;C</td>
<td>[seqmonth1_th0_temp_c@12]&deg;C</td>
<td>[seqmonth1_th0_temp_c@13]&deg;C</td>
</tr>

2. The monthly averages for temperature show a weird behaviour.
I import my data to WSWIN and the monthly averages showing up there are 0.5 to 1°C higher than what meteohub displays.
The same is true for the data meteohub sends to Wunderground.
Bottomline: Wunderground and WSWIN coming up with the same monthly average temp values, but meteohub shows values about 0.5 - 1.0°C less.
That leads to the effect, that meteohubs yearly average temp is about 1°C lower than what WSWIN and WU do come up with.

What could be the reason? Some internal mis-calc?

[admin]

Could be a result of how Meteohub computes averages. Meteohub assumes that data pops up at more or less constant and regular intervals. It sums up all values and divides it by the number of samples. When you have more samples in winter month than in summer month, because you changed your weather station type i.e., than this assumption is false and values might show a drift in the direction to the higher sampling rate.

Meteoplug does compute that in another way. Does Meteoplug show the same effects on your data?

[semtex]

I upgraded to this latest version.
when restarting the system only a single beep and meteohub don't work
also ftp dont work
help please!!!!
(ALIX.1D)

[admin]

can you reach the web interface?
can you access via ssh on port 22?
I bit more detail would help.
I know of other ALIX users having no problems on updating to 4.7j - so it might be something very special with your rig.

[gm4jjj]

Only slight problem

Logfile shows:

2011.06.21 02:17:15 meteohub: weathernet: metoffice:
2011.06.21 02:32:14 meteohub: weathernet: metoffice:
2011.06.21 02:47:13 meteohub: weathernet: metoffice:

One entry for each transfer. The data is being successfully transferred to metoffice.

Many thanks.

[semtex]

can you reach the web interface?
can you access via ssh on port 22?
I bit more detail would help.
I know of other ALIX users having no problems on updating to 4.7j - so it might be something very special with your rig.

no access web interface
no access via ssh
when I turn on don't start nothing

[admin]

Only slight problem

Logfile shows:

2011.06.21 02:17:15 meteohub: weathernet: metoffice:
2011.06.21 02:32:14 meteohub: weathernet: metoffice:
2011.06.21 02:47:13 meteohub: weathernet: metoffice:

One entry for each transfer. The data is being successfully transferred to metoffice.

Many thanks.

Will be fixed with next update.

[iduskia]

hello
i've the same problem : after update, i've no access to the web interface, ni thru the network. Only one beep and a yellow led power activated
i need some help !!
iduskia

[admin]

@iduskia:
what platform are you using? ALIX? Sheeva?
Does it not signal it's IP by the LEDs as explained in the manual?
When your Meteohub does a reboot, it can get a new IP address from
your router unless you tell your router to keep the IP for that specicif MAC constant.

[iduskia]

I ve NSLU2 platform !!
i doesn't signal his IP at the boot : only one beep and a the status led flashing yellow..
very strange : il happened at the reboot after upgrading in vers 4.7J

[admin]

USB stick problems often show up when updates are done. Looks very much like a bad USB sick.
Please setp a new one as lined out in the manual.

[semtex]

I have changed the CF...is demaged
I can read the old CF
what is the directory to retrieve the configuration of the graphics ??

[admin]

"graphs" is for graph definitions
"weather/YYYYMM/raw" holds raw data of month MM of year YYYY