pi is online and weewx driver is working - but not posting

[Sean Jahnig]

Hi,

This is a different station that i setup about a month ago. This is using an Acurite 5-in-1 with and SDR dongle - no displays involved. The station has been online for the last 1 month with no problems. Today i woke up and the despite everything being online, the data is not being uploaded to windguru and wunderground. Attached is the log file after i set debug to 1.

When the station is working, these are the links it broadcasts to:

windguru link: https://www.windguru.cz/station/2146 (offline since 3am)

wunderground link: https://www.wunderground.com/dashboard/pws/IUMMAL1

From what i can see, the pi is seeing the sensor and it is collecting the data. What is odd is that when i open 192.168.1.150/weewx from a browser - the reports open but the time at the top shows '12/29/2019 11:25:00 AM' 

I have tried numerous reboots, no difference.

When i run 'date' on the pi, the reply is 'Sun 12 Jan 2020 08:44:15 AM +04' and it is accurate.

When i ping www.google.com i get a reply:

PING www.google.com (172.217.19.164) 56(84) bytes of data.

64 bytes from fjr02s09-in-f4.1e100.net (172.217.19.164): icmp_seq=1 ttl=53 time=5.69 ms

64 bytes from fjr02s09-in-f4.1e100.net (172.217.19.164): icmp_seq=2 ttl=53 time=5.24 ms

Any ideas?

[gjr80]

Hi,

The lack of posting to external services (at present) has nothing to do with connectivity. If you look at the log your system is not generating any archive records, the windy uploader is bound to the NEW_ARCHIVE_RECORD event so uploads are only triggered when a new archive record is created. Fix the archive record issue and I suspect all will be fine.

I would probably start looking at your archive interval, 100 seconds is somewhat unusual. Issue #469 was raised to deal with 'oddball' archive intervals, but it will not appear until 4.0. Whilst your archive interval is not quite 'oddball' by issue #469 standards, I would try making it something more conventional like five minutes, restart and see if that rectifies the archive record generation issue.

Gary

[Sean Jahnig]

Thanks for the reply.

I changed the archive interval back to 300 and rebooted. The system has been online for 3 hours now and it's still now posting.

Any other ideas?

--
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/rlau4OqQTtY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/92ecb14b-e6d4-4ca6-94c6-ce7226b48613%40googlegroups.com.

[gjr80]

OK, so no luck there. I would be looking at your sensor map next. In your sensor map I see wind_speed but in the log I see the sdr driver finds wind_speed_kph. I would work through the steps in the sdr driver installation instructions (https://github.com/matthewwall/weewx-sdr) from step 4 onwards to verify your sensor map is correct. Specifically, what does step 4 capture and what sensor map did you implement at step 5 based on that capture.

Gary

[Sean Jahnig]

Hi Gary,

Thanks for helping me troubleshoot. I found the problem.

In the sensor map i originally specified which items to capture by using the identifier '013E' which i obtained by running rtl_433 -M utc -F json -G and that was working for the last month until 3am this morning. Below is what i was using:

wind_speed.013E.Acurite5n1Packet

Now, upon a closer inspection of running rtl_433 -M utc -F json -G again i notice the identifier has changed to '0A0A'. So i updated the sensor map in /etc/weewx/weewx.conf to use the new identifier and it all started working.

wind_speed.0A0A.Acurite5n1Packet

Any ideas why that would've changed?

On Sun, Jan 12, 2020 at 12:12 PM gjr80 <gjrod...@gmail.com> wrote:

OK, so no luck there. I would be looking at your sensor map next. In your sensor map I see wind_speed but in the log I see the sdr driver finds wind_speed_kph. I would work through the steps in the sdr driver installation instructions  (https://github.com/matthewwall/weewx-sdr) from step 4 onwards to verify your sensor map is correct. Specifically, what does step 4 capture and what sensor map did you implement at step 5 based on that capture.

Gary

--
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/rlau4OqQTtY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to weewx-user+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/fe850799-6774-4458-9f7f-2fe94b208820%40googlegroups.com.

[gjr80]

No no idea, have not used SDR myself. Maybe Matthew or others have come across changing sensor ids.

Gary

[mwall]

On Sunday, January 12, 2020 at 3:56:32 AM UTC-5, Sean Jahnig wrote:

Any ideas why that would've changed?

it depends on the sensor.  i know that some temperature/humidity sensors have a new, apparently random identifier each time you power cycle them.  however, i have never seen the 5n1 sensor do that.

m

[Sean Jahnig]

Thanks for the reply Matthew.

We had a big storm last night so i guess the sensor must've power cycled itself? It is mounted far out of reach so i'm quite certain no one touched it.

Will keep an eye on it. 

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/rlau4OqQTtY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to weewx-user+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/a4bbc408-2fec-44ee-a40a-0b1c355f6495%40googlegroups.com.