Whatever it is, it's related to RPis talking to a Davis Logger via
USB. I have a couple of powered USB hubs lying around. I can try putting
one of them on in the next day or so . . . now that you mention it, I
never had any trouble with an RPi 3. However, it did/does manifest
itself on
I did not expect the errors to go away, but I did expect a wakeup to clear
them. Could you try again, except this time set debug=1, then restart
weewxd? That will show the wakeup attempts.
On Tue, May 17, 2022 at 1:19 PM WindnFog wrote:
> I spoke too quickly, Tom. I did a cold power off,
I spoke too quickly, Tom. I did a cold power off, powered on, and got
the ubiquitous "error: Expected to read 99 chars; got 0 instead" 3 times
followed by the reset (twice,) and then it settled down and seems OK. This
is what it's been doing on my system for some months. I can live with
Me either, Tom. Since it's intermittent, I might be jumping the gun
in claiming that it fixed things. We'll have to wait for the others to
weigh in. I've got some outside things to do today, but once I get a few
hours, I'll warm and cold reboot it to see if I can reproduce the error.
It
I don't see why the patch would stop the "Expected 99, got 0" errors, just
make them less traumatic.
For your find_rain_gust_day script, why not just use $month.rain.maxsum and
$month.wind.maxtime?
Look in the template NOAA-%Y.txt.tmpl for clues on how to calculate all
these things using tags.
That seems to have solved the problem completely, Tom. However, based
on what you wrote, I thought it only made it more tolerant of logger
errors. Maybe something in the re-write changed the timing slightly. Or I
misinterpreted what you said it did.
Either way, I didn't see the
All,
Can you give this version of the driver a try? Rather than restart from the
top, it tries a wakeup, then a retry. If that doesn't work after 3
attempts, then it does a restart.
https://raw.githubusercontent.com/weewx/weewx/Issue_772/bin/weewx/drivers/vantage.py
-tk
On Mon, May 9, 2022 at
I wish mine would settle down. I do the stop/start jobs because mine
eventually exits the engine.
Dave
May 9 12:56:30 HP-Iron weewx[3016530] INFO weewx.restx: Wunderground-PWS:
Published record 2022-05-09 12:55:00 CDT (1652118900)
May 9 12:56:55 HP-Iron weewx[3016530] ERROR
Here's my 2-cents worth. I have been fighting this error for 1-2 years
with a Vantage VUE and a roughly 6-year-old USB logger. It's always on
startup (or after a reboot.) It happens a few times and then stops. So
here are the things I've done to "fix" it.
(1) Cleared the logger and reset
Also, I bought a new USB data logger as I thought that could be the issue.
It was not.
Dave
On Monday, May 9, 2022 at 8:32:18 AM UTC-5 David Hathaway wrote:
> If you need another system to test on, mine reliably has this issue. I
> talked about it in Unable to wake up Vantage console
>
Thanks for that, John.
I think we're saying the same thing because a new LOOP request starts with
a wake-up. My intention was to catch the error lower down in the stack,
then restart the loop.
It's good to hear that the strategy works!
-tk
On Sun, May 8, 2022 at 4:30 PM 'John Kline' via
Hi Tom,
You don’t need to restart or try a wake up. You just need to return from
genDavisLoopPackets so another request is made for packets.
Below is what I have been running. You can simplify it as the bad_read
variable is not necessary since it has NEVER occurred that two bad reads
Tom
It's a USB Data Logger (old type of USB). AC powered with battery backup,
no taught wires and the logger is housed securely.
Just over an hour has gone by with no more errors, as you say, it may just
be the reset/catch up that caused the second issue. I'll have a look again
on the morrow
Now that I think about it, there is an experiment I've been meaning to try.
Right now, the Vantage driver tries 3 times, then gives up and restarts the
program from the top.
The experiment would be to do something not quite so drastic. Rather than
restart the program, try a Vantage "wake up"
The "Expected 99, got 0" error is pretty common, especially after the
logger has done a catch up. I have not noticed any correlation with the age
of the logger.
What kind of connection are you using? USB? Serial? Serial through a USB
converter? The last (converters) are notoriously flaky.
-tk
Spoke too soon. The error appears to back back up, is this a data logger
issue?:
May 8 20:55:18 raspberrypi weewx[8172] INFO weewx.restx: AWEKAS: Published
record 2022-05-08 20:55:00 BST (1652039700)
May 8 20:55:18 raspberrypi weewx[8172] INFO weewx.restx: PWSWeather:
Published record
16 matches
Mail list logo
