Re: [weewx-user] Re: weewx, raspberry pi and sd cards - the lifetime for card

2016-11-07 Thread Chris Thompstone
In case anyone might be interested, my setup:

Odroid u3 (now discontinued, but other better ones now available)
Sqlite
Debian wheezy running on the eMMC card (maybe eMMC better life span than 
sd?)
using ramlog to store all logs into ram, I think it maybe writing 
occasionally or before reboots.
the weewx website is stored on tmpfs (ram) so no writes here.
The only writes to storage I make is to the sqlite database every 5 mins.
Obviously automated backups are made elsewhere of the database.

This has run happily for over 2 years so far.
No problems touch wood.

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] Off topic - sql issue

2016-11-07 Thread Andrew Milner
Hi guys - I'm having a problem getting my head around some sql - using 
MySQL.
I have a table with rows containing 
datetime, id, reading 

for example - and I'm trying to create a view which contains the latest 
readings for all ids - what is the sql to achieve this??
select max(datetime), id, reading group by id; 

appeared to work - it gave me the latest datetimes for all ids, but the 
readings did not correspond to the datetimet!!!

Anyone offer any help??


-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Re: [weewx-user] Noob question... where to find the actual data (temp) within weewx?

2016-11-07 Thread gjr80
No problems Jim, I have been using weewx for a few years now and I still 
keep learning things from the guides, particularly the Customization Guide.

Gary

On Tuesday, 8 November 2016 10:52:00 UTC+10, Jim W. wrote:
>
> Gary, Thanks! and yes I did read the entire weewx/docs guide... however it 
> was very early into trying to implement weewx and before I had much 
> understanding of how weewx worked. I should (and will) re-read the 
> documentation now that I have a little better understanding of how things 
> work.
>
> On Sunday, November 6, 2016 at 10:01:52 PM UTC-5, gjr80 wrote:
>>
>> Hi,
>>
>> I think you will get nothing other than an error message. 
>> $windDir.ordinal_compass will give you an ordinal compass direction for the 
>> latest wind direction. Have a read about wind ordinals in the Customization 
>> Guide http://weewx.com/docs/customizing.htm#unit_conversion_options
>>
>> Gary
>>
>>

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] Re: WU import missing rain data

2016-11-07 Thread gjr80
Just to tie this off it was fixed at comit c027bb4 
.
 


The bug will only come into play if you are using wee_import under 2.6.0 or 
2.6.1 *AND* you want to do a WU import *AND* WU natively presents you with 
US Customary/Imperial units (I think this is USA geolocated users only). If 
this is the case any WU imports will ignore any rain data. The workaround 
is to download the current wuimport.py from the weewx master branch on 
GitHub, move the old bin/weeimport/wuimport.py aside and copy the 
downloaded version in its place. Something like the following should do the 
job (paths and permissions aside):

$ wget -P /var/tmp https:
//raw.githubusercontent.com/weewx/weewx/master/bin/weeimport/wuimport.py
$ mv /home/weewx/bin/weeimport/wuimport.py /home/weewx/bin/weeimport/
wuimport_orig.py
$ cp /var/tmp/wuimport.py /home/weewx/bin/weeimport/wuimport.py

If you have already completed a WU import that was affected by the bug then 
the only (practical) remedy is to delete the records concerned, obtain the 
fixed wuimport.py and re-import, unless of course there was no rain during 
the period you imported!

Gary

On Monday, 7 November 2016 22:35:37 UTC+10, gjr80 wrote:
>
> That'll do it, not sure how I missed that, probably something to do with 
> WUs ever consistent approach to life: PressureIn and HourlyPrecipIn but 
> dailyrainin.
>
> Thanks Andrew.
>
> Gary
>
> On Monday, 7 November 2016 22:06:30 UTC+10, Andrew Chilvers wrote:
>>
>> Ok, figured this out, it's a case sensitivity bug in wuimport.py. The 
>> code is trying to map dailyrainIn, with a capital I, and WU is returning 
>> dailyrainin, all lowercase.
>>
>> Maybe one of the devs can update the code.
>>
>> Thanks
>>
>> On Sunday, November 6, 2016 at 4:50:44 PM UTC-5, gjr80 wrote:
>>>
>>> Hi,
>>>
>>> Going to need a little more info.
>>>
>>> You say 'all the rain data is missing', what makes you say this? Have 
>>> you looked in your weewx archive? Are your reports missing data? Are there 
>>> any errors, warnings or other messages in your log from the import session? 
>>> What import config file did you use and what changes did you make to it?
>>>
>>> Gary
>>>
>>>

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Re: [weewx-user] Noob question... where to find the actual data (temp) within weewx?

2016-11-07 Thread Andrew Milner
Jim - you need to get yourself into the habit of always referring to the 
documentation whenever you have issues and questions with weewx - and 
especially with new releases etc as very often things may change between 
releases!!  ALWAYS read the update guide for each new release in particular 
as it will tell you what is new/fixed in the new release.  With weewx the 
documentation is worth its weight in gold and the best way of understanding 
and appreciating weewx and its possibilities is to become familiar with 
knowing at least what is within the manuals even if, when reading, things 
may not be fully understood.

On Tuesday, 8 November 2016 02:52:00 UTC+2, Jim W. wrote:
>
> Gary, Thanks! and yes I did read the entire weewx/docs guide... however it 
> was very early into trying to implement weewx and before I had much 
> understanding of how weewx worked. I should (and will) re-read the 
> documentation now that I have a little better understanding of how things 
> work.
>
> On Sunday, November 6, 2016 at 10:01:52 PM UTC-5, gjr80 wrote:
>>
>> Hi,
>>
>> I think you will get nothing other than an error message. 
>> $windDir.ordinal_compass will give you an ordinal compass direction for the 
>> latest wind direction. Have a read about wind ordinals in the Customization 
>> Guide http://weewx.com/docs/customizing.htm#unit_conversion_options
>>
>> Gary
>>
>>

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Re: [weewx-user] Noob question... where to find the actual data (temp) within weewx?

2016-11-07 Thread Jim W.
Gary, Thanks! and yes I did read the entire weewx/docs guide... however it 
was very early into trying to implement weewx and before I had much 
understanding of how weewx worked. I should (and will) re-read the 
documentation now that I have a little better understanding of how things 
work.

On Sunday, November 6, 2016 at 10:01:52 PM UTC-5, gjr80 wrote:
>
> Hi,
>
> I think you will get nothing other than an error message. 
> $windDir.ordinal_compass will give you an ordinal compass direction for the 
> latest wind direction. Have a read about wind ordinals in the Customization 
> Guide http://weewx.com/docs/customizing.htm#unit_conversion_options
>
> Gary
>
>

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Re: [weewx-user] Re: weewx, raspberry pi and sd cards - the lifetime for card

2016-11-07 Thread vince
On Monday, November 7, 2016 at 11:50:31 AM UTC-8, ln77 wrote:
>
> On my RPi installs of jessie and wheezy, /var/run is just a symlink to 
> /run, and /run is a tmpfs filesystem — it’s created by an init script, not 
> mounted out of fstab, but it is tmpfs.  
>
>
ummm - yup it is, I forgot to look at that.  Thanks. 
And it's already tmpfs, as you mentioned (wheezy 8.6).

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Re: [weewx-user] Re: weewx, raspberry pi and sd cards - the lifetime for card

2016-11-07 Thread Frank Bandle
Hi Vince

for the weewx directory, your NOAA reports will disappear every time you 
reboot, so wouldn't weewx (very slowly) regenerate them every reboot ?
I didn't notice where your archive directories are located.  Do you copy the 
old one back into place via rc.local or something on every reboot ?
We did not use NOAA reports on these raspberries. The raspberries we use only 
collect data and send to our servers. 
Archive - the weewx.db is still on the SD Card in standard path  /var/lib/weewx 

we run jessie on the raspberries with no problem running this fstab. 
Maybe /var/run to tmpfs causes problems on other systems - here no problems 

Does a 'systemctl' command on your system show a couple failed processes ?

No :-) 



- frank



-- 
Dipl. Met. Frank Bandle
Sent with Airmail

Am 7. November 2016 um 17:40:40, vince (vinceska...@gmail.com) schrieb:

On Monday, November 7, 2016 at 3:48:11 AM UTC-8, fraban wrote:
we solved the problem with raspberries and SD Cards with writings most things 
to "RAMDISK" /  tmp Files.  So RAM Drive is the solution for the problem.
 
# For Debian Jessie - RAM Drive using weewx
tmpfs           /tmp            tmpfs   defaults,nosuid,mode=0755,nodev,noatime 
  0       0
tmpfs           /var/log        tmpfs   defaults,nosuid,mode=0755,nodev,noatime 
  0       0
tmpfs           /var/tmp        tmpfs   defaults,nosuid,mode=0755,nodev,noatime 
  0       0
tmpfs           /var/run        tmpfs   defaults,nosuid,mode=0755,noatime       
        0       0

hmmm - this is really interesting, you can stick syslogs and /tmp and /var/tmp 
in tmpfs just by editing fstab ?  Cool.

I did this in a vagrant vm and it's making systemd rather unhappy, although the 
system seems to run ok:
initial login after bootup is 'very' slow
systemctl reports user@MYUID.service failed
systemctl reports systemd-update-utmp.service failed
systemctl reports 'Failed to start System Logging Service' every 90 secs or so
systemctl reports watchdog timeout for systemd.logind
systemctl reports watchdog timeout for systemd.journald
I did a little more research with some controlled reboots and found that 
systemd really doesn't like /var/run in tmpfs.

not putting /var/run into tmpfs made the problems above go away, and 
'systemctl' is clean
Does a 'systemctl' command on your system show a couple failed processes ?





tmpfs /var/www/weewx  tmpfs   defaults,nosuid,mode=0755,nodev,noatime         0 
      0
tmpfs           /var/log/nginx  tmpfs   defaults,nosuid,mode=0755,nodev,noatime 
        0       0

for the weewx directory, your NOAA reports will disappear every time you 
reboot, so wouldn't weewx (very slowly) regenerate them every reboot ?
I didn't notice where your archive directories are located.  Do you copy the 
old one back into place via rc.local or something on every reboot ?


Cool idea putting that stuff in tmpfs rather than hard-allocating a ramdisk and 
maybe not using it...that'll give me more free space on the pi that does my 
weathercam.  Thanks !



--
You received this message because you are subscribed to a topic in the Google 
Groups "weewx-user" group.
To unsubscribe from this topic, visit 
https://groups.google.com/d/topic/weewx-user/atRpRzKrlyk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to 
weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] Re: weewx, raspberry pi and sd cards - the lifetime for card

2016-11-07 Thread vince
On Monday, November 7, 2016 at 3:48:11 AM UTC-8, fraban wrote:

> we solved the problem with raspberries and SD Cards with writings most 
> things to "RAMDISK" /  tmp Files.  So RAM Drive is the solution for the 
> problem. 
>  
>
# For Debian Jessie - RAM Drive using weewx
> tmpfs   /tmptmpfs   
> defaults,nosuid,mode=0755,nodev,noatime   0   0
> tmpfs   /var/logtmpfs   
> defaults,nosuid,mode=0755,nodev,noatime   0   0 
>
tmpfs   /var/tmptmpfs   
> defaults,nosuid,mode=0755,nodev,noatime   0   0
>
tmpfs   /var/runtmpfs   defaults,nosuid,mode=0755,noatime   
> 0   0 
>

hmmm - this is really interesting, you can stick syslogs and /tmp and 
/var/tmp in tmpfs just by editing fstab ?  Cool.

I did this in a vagrant vm and it's making systemd rather unhappy, although 
the system seems to run ok:

   - initial login after bootup is 'very' slow
   - systemctl reports user@MYUID.service failed
   - systemctl reports systemd-update-utmp.service failed
   - systemctl reports 'Failed to start System Logging Service' every 90 
   secs or so
   - systemctl reports watchdog timeout for systemd.logind
   - systemctl reports watchdog timeout for systemd.journald

I did a little more research with some controlled reboots and found that 
systemd really doesn't like /var/run in tmpfs.

   - not putting /var/run into tmpfs made the problems above go away, and 
   'systemctl' is clean

Does a 'systemctl' command on your system show a couple failed processes ?



tmpfs /var/www/weewx  tmpfs   defaults,nosuid,mode=0755,nodev,noatime   
>   0   0
> tmpfs   /var/log/nginx  tmpfs   
> defaults,nosuid,mode=0755,nodev,noatime 0   0
>


   - for the weewx directory, your NOAA reports will disappear every time 
   you reboot, so wouldn't weewx (very slowly) regenerate them every reboot ?
   - I didn't notice where your archive directories are located.  Do you 
   copy the old one back into place via rc.local or something on every reboot ?


Cool idea putting that stuff in tmpfs rather than hard-allocating a ramdisk 
and maybe not using it...that'll give me more free space on the pi that 
does my weathercam.  Thanks !


-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] Re: No data from station

2016-11-07 Thread Thomas O


On Sunday, November 6, 2016 at 6:41:44 PM UTC+1, mwall wrote:
>
> On Sunday, November 6, 2016 at 8:58:36 AM UTC-5, Thomas O wrote:
>>
>> I see the same issue with 0.15rc1 from master and WMR300.
>>
>> Any data you need to further debug ; I am trying to diff 0.9 and 0.15rc1 
>> but I didn't find anything obvious yet.
>>
>> Can i provide some more traces ?
>>
>
> thomas,
>
> please post your usb configuration 'dpkg -l | grep usb'
>

I will unfortunately I have access to the hardware only on weekends.
 

>
> also, please post the exception stack when you run 0.15rc1
>
> this is rather perplexing. i thought the localization that alberto (es) 
> and roland (de) are using might be causing the problem, but i'm not sure.  
> the driver checks for 'No data available' and 'No error' messages from 
> libusb.  (i have not yet figured out a way to check for an error code that 
> works across different libusb versions)
>
>
You may be on something as my localization is "fr"


Thomas.

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] Re: Interceptor Driver - rainin dailyrain - Acurite 5 in 1

2016-11-07 Thread Brad Tucker
It has been loaded. Ill report back in the morning if all goes well.

Thanks Matt,
Brad

On Monday, November 7, 2016 at 7:57:39 AM UTC-8, mwall wrote:
>
>
>
> On Monday, November 7, 2016 at 12:08:35 AM UTC-5, Brad Tucker wrote:
>>
>> Good Evening Matt,
>>
>> I thought tcpflow would go forever but forever happened to only be 12 
>> hours ;) After running for a very long time everything crashed. Ive been 
>> keeping a mindful eye on it so I caught it pretty early... Attached is a 
>> syslog. Let me know if you see anything I should look for. The error codes 
>> are the same as the last crash. Not ure that helps any...
>>
>
> hi brad,
>
> please try the latest interceptor.py.  i added rejection of unexpected cgi 
> formats at commit 2ad77b7
>
> m 
>

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] Re: Interceptor Driver - rainin dailyrain - Acurite 5 in 1

2016-11-07 Thread mwall


On Monday, November 7, 2016 at 12:08:35 AM UTC-5, Brad Tucker wrote:
>
> Good Evening Matt,
>
> I thought tcpflow would go forever but forever happened to only be 12 
> hours ;) After running for a very long time everything crashed. Ive been 
> keeping a mindful eye on it so I caught it pretty early... Attached is a 
> syslog. Let me know if you see anything I should look for. The error codes 
> are the same as the last crash. Not ure that helps any...
>

hi brad,

please try the latest interceptor.py.  i added rejection of unexpected cgi 
formats at commit 2ad77b7

m 

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] Re: weewx 2 different drivers

2016-11-07 Thread mwall


On Monday, November 7, 2016 at 7:28:24 AM UTC-5, martin swanepoel wrote:
>
> Running 2 instances of weewx. 1 is the sdr driver but this station don't 
> have a pressure sensor. 2 Hook up a bmp085 onto the raspberry for the 
> second instance to get barometric data and inside temperature.
>
> Question now how do I integrate the 2 together on one www server. What 
> happens now is they overwrite the same pages. So sometimes I see the 
> barometric graphs and the next time the graphs from the sdr equipment. Any 
> suggestions ?
>

hi martin,

here are three different approaches:

option 1: make data collection from bmp085 a service instead of a driver.  
run a single instance of weewx with the sdr driver.  collect data from the 
bmp085 using a service that augments the loop and/or archive data.  display 
data in reports as usual.

option 2: run two instances of weewx to collect data, but one instance 
knows about the other for reporting.  ensure that each instance is writing 
to its own database.  one instance of weewx will not generate any reports 
(remove the StdReport service from this instance).  the other instance uses 
the data_binding parameter in its reports to include data from the other 
instance (see the customization guide for details).

option 3: run two instances of weewx to collect data, plus a third instance 
to do the reporting.  each data collection instance has its own database 
and HTML_ROOT.  the reporting instance has yet another HTML_ROOT and uses 
data_binding to get data from the two collector databases.

although two weewx instances *can* share a single database, it is usually 
not a good idea to do so.  same goes for HTML_ROOT.

there are many other possible permutations, but that should get you started.

m

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] Re: WU import missing rain data

2016-11-07 Thread gjr80
That'll do it, not sure how I missed that, probably something to do with 
WUs ever consistent approach to life: PressureIn and HourlyPrecipIn but 
dailyrainin.

Thanks Andrew.

Gary

On Monday, 7 November 2016 22:06:30 UTC+10, Andrew Chilvers wrote:
>
> Ok, figured this out, it's a case sensitivity bug in wuimport.py. The code 
> is trying to map dailyrainIn, with a capital I, and WU is returning 
> dailyrainin, all lowercase.
>
> Maybe one of the devs can update the code.
>
> Thanks
>
> On Sunday, November 6, 2016 at 4:50:44 PM UTC-5, gjr80 wrote:
>>
>> Hi,
>>
>> Going to need a little more info.
>>
>> You say 'all the rain data is missing', what makes you say this? Have you 
>> looked in your weewx archive? Are your reports missing data? Are there any 
>> errors, warnings or other messages in your log from the import session? 
>> What import config file did you use and what changes did you make to it?
>>
>> Gary
>>
>>

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] Re: WU import missing rain data

2016-11-07 Thread Andrew Chilvers
Ok, figured this out, it's a case sensitivity bug in wuimport.py. The code 
is trying to map dailyrainIn, with a capital I, and WU is returning 
dailyrainin, all lowercase.

Maybe one of the devs can update the code.

Thanks

On Sunday, November 6, 2016 at 4:50:44 PM UTC-5, gjr80 wrote:
>
> Hi,
>
> Going to need a little more info.
>
> You say 'all the rain data is missing', what makes you say this? Have you 
> looked in your weewx archive? Are your reports missing data? Are there any 
> errors, warnings or other messages in your log from the import session? 
> What import config file did you use and what changes did you make to it?
>
> Gary
>
>

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] Re: weewx, raspberry pi and sd cards - the lifetime for card

2016-11-07 Thread fraban

Hi all

we solved the problem with raspberries and SD Cards with writings most 
things to "RAMDISK" /  tmp Files.  So RAM Drive is the solution for the 
problem. 
And we use 32GB or 16GB Cards Class 10 ( SAMUNG - no "NONAMES" ! ) 

We run all our raspberries - with the following entries in /etc/fstab -  So 
we reduced the writings on the SD Card to a minimum .  
Since we use these settings our raspberries are OK and run now more then 6, 
8 and 12 month without problems 24h a day. 

Regards Frank

pi@raspberrypi ~ $ more /etc/fstab
proc/proc   procdefaults  0   0
/dev/mmcblk0p1  /boot   vfatdefaults  0   2
/dev/mmcblk0p2  /   ext4defaults,noatime  0   1
# a swapfile is not a swap partition, no line here
#   use  dphys-swapfile swap[on|off]  for that
# For Debian Jessie - RAM Drive using weewx
tmpfs   /tmptmpfs   
defaults,nosuid,mode=0755,nodev,noatime   0   0
tmpfs   /var/logtmpfs   
defaults,nosuid,mode=0755,nodev,noatime   0   0
tmpfs   /var/tmptmpfs   
defaults,nosuid,mode=0755,nodev,noatime   0   0
tmpfs /var/www/weewx  tmpfs   defaults,nosuid,mode=0755,nodev,noatime   
  0   0
tmpfs   /var/log/nginx  tmpfs   
defaults,nosuid,mode=0755,nodev,noatime 0   0
tmpfs   /var/runtmpfs   defaults,nosuid,mode=0755,noatime   
0   0

-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[weewx-user] meteostick - experience since start of development the driver this year

2016-11-07 Thread fraban



Hi all - meteostick users together with weewx 

first let me thank Luc, Matthew and Kobuki for the great work with the 
driver we initiated now approx a year ago with the first steps.  In the 
meanwhile meteostick works stable and as we can say in most times good. s. 
above.
 
We have a few meteosticks in our small new weathernetwork - still we are in 
closed BETA -> if you want to have a look check: 
 https://stations.weather365.net .  
There are actually only a few stations in the new network.  We will go 
public beta in the next days - I will write a new post if we go this step 
and will invite you to participate. 

We observerd sometimes strange pressure jumps with meteostick.
Still what we found out with meteostick is, that we have some funny 
pressure spikes in the afternoon.  

This is from our Teststation Vantage PRO 2 with raspberry and meteostick 
https://stations.weather365.net/#/station1//  check out "last 30 days / 
Letzte 30 Tage" ...







We can zoom in and we will see pressure differences in a few minutes approx 
with 3 hPa ...  
see next graph. 








Question is:  What happens ?  We made some tests and we found the reason. 
It´s no new weather phenomen !

We installed the meteostick and rapberry near a window in the west of our 
building. In the atfternoon sun comes at approx 4pm - til 5pm and shines 
directly through the window on the meteostick behind. ( only a few minutes )
The pressure sensor is inside the transparent USB meteostick. The small 
amount of air behind the "acrylglass" stick will become warmer and warmer 
in these minutes in the sun "warm air" streams out of the stick and the 
pressure inside the stick decreases rapidly. This effect you can see in the 
pressure measurement in that time period. 
In the meanwhile we placed the stick in the shadow and we have no more 
funny pressure jumps in the afternoon when the sun is shining. 

So our advice using meteostick is - leave it in the shadow and protect it 
from sun :-) 

Regards 
Frank 


-- 
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 weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.