So the TEE file, the weewx.log file, and WU table seem to jive. I'm just not capturing baro data with regularity.
WU site: https://www.wunderground.com/personal-weather-station/dashboard?ID=KCASANJO594#history TEE file: ~~~ mac=0004a369e0d6&id=c2&pv=0&lb=0&ac=0®=1809&lost=0000&baro=1010&ptr=0&wfor=1&p=1 mac=0004a369e0d6&id=c2&pv=0&lb=0&ac=0®=1809&lost=0000&baro=1011&ptr=1&wfor=1&p=1 mac=0004a369e0d6&id=c2&pv=0&lb=0&ac=0®=1809&lost=0000&baro=1011&ptr=0&wfor=1&p=1 mac=0004a369e0d6&id=c2&pv=0&lb=0&ac=0®=1809&lost=0000&baro=1011&ptr=0&wfor=1&p=1 mac=0004a369e0d6&id=c2&pv=0&lb=0&ac=0®=1809&lost=0000&baro=1012&ptr=0&wfor=1&p=1 ~~~ weewx.log: ~~~ Nov 27 08:15:15 raspberrypi weewx[17396]: interceptor: MainThread: mapped packet: {'barometer': 1010.0, 'usUnits': 17, 'dateTime': 1480263316} Nov 27 12:16:16 raspberrypi weewx[17396]: interceptor: MainThread: mapped packet: {'barometer': 1011.0, 'usUnits': 17, 'dateTime': 1480277776} Nov 27 12:27:15 raspberrypi weewx[17396]: interceptor: MainThread: mapped packet: {'barometer': 1011.0, 'usUnits': 17, 'dateTime': 1480278436} Nov 27 13:57:13 raspberrypi weewx[17396]: interceptor: MainThread: mapped packet: {'barometer': 1011.0, 'usUnits': 17, 'dateTime': 1480283834} Nov 27 14:03:13 raspberrypi weewx[17396]: interceptor: MainThread: mapped packet: {'barometer': 1012.0, 'usUnits': 17, 'dateTime': 1480284194} ~~~ BTW, I find it interesting that you ask about readings. I'm at 220M. The data the equipment reports is never the same as the data that WU reports. For instance, the final reading above is 1012 from the station, but WU reports is as what is probably a more accurate 30.08". 1012 looks to be 29.77"... -SR On Sunday, November 27, 2016 at 1:26:38 PM UTC-8, Sam Roza wrote: > > THis is currently happening to my barometric pressure, so I'm actively > looking into it. > > What does the POST for pressure look like? I'm trying to remember if is > shows in mb or what? > > I'm getting wind data and thermo/hygro data, but can't tell if the > pressure is buried in the thgr POST or not? > > ~~~ > T 192.168.1.73:2850 -> 54.228.205.96:80 [AP] > > mac=0004a369e0d6&id=84&rid=d6&pwr=0&htr=0&cz=0&oh=54&ttr=0&ot=16.0&ch=3&p=1 > > ~~~ > > _SR > > On Thursday, November 24, 2016 at 10:03:44 AM UTC-8, Sam Roza wrote: >> >> OK, so there were several hours again where I recorded no data into >> weewx, out to WU/CWOP, but looking a tthe logs, I see packets arriving, >> being processed, and being sent out to external services. I looked at the >> tee'd file after some time thinking about it, and I see a big gap in the >> data where the temp sensor doesn't report (I think that's 'ttr' in the POST >> data). Wind is an issue, too. >> >> Any clue what would cause 2 pieces of hardware to stop reporting at >> basically the same time? Lost data starts after 0443 and begins at 0746. >> >> I'm including the following files: >> >> - unadulterated syslog >> - filtered syslog >> - tee file (no timestamps, so hard for me to see value/know when the >> issue occurred) >> >> Hmm. Will think on this over turkey and alcohol. >> >> -SR >> >> The system just started working on its own, with no action on my part. >> >> On Friday, November 18, 2016 at 1:50:18 PM UTC-8, Sam Roza wrote: >>> >>> I will try it with the tee. I couldn't get this to work without the sed >>> command, IIRC. >>> >>> -SR >>> >>> On Friday, November 18, 2016 at 12:00:49 PM UTC-8, mwall wrote: >>>> >>>> On Friday, November 18, 2016 at 2:37:28 PM UTC-5, Sam Roza wrote: >>>>> >>>>> I use interceptor with the following script: >>>>> >>>>> ~~~ >>>>> ngrep -l -q -d eth0 '0004a369e0d6' | sed -u '/mac=/!d' | xargs -n 1 >>>>> curl http://localhost:9999 -s -d >>>>> ~~~ >>>>> >>>>> I can tee it into a file if you think we're losing data. >>>>> >>>> >>>> when debug=1, the interceptor will report any data it receives via POST >>>> or GET >>>> >>>> so the question is whether the ngrep-sed-curl process is filtering out >>>> data from the gateway >>>> >>>> the sed command >>>> >>>> sed -u '/mac=/!d' >>>> >>>> skips any line that does not contain mac=, so if somehow the lines from >>>> the t/h and wind sensor were scrambled/shifted, that command might remove >>>> them. >>>> >>>> you could temporarily try a tee before the sed: >>>> >>>> ngrep -l -q -d eth0 '0004a369e0d6' | tee /var/tmp/dump.txt | sed -u >>>> '/mac=/!d' | xargs -n 1 curl http://localhost:9999 -s -d >>>> >>>> if you see another lapse in data, check the tee output file for strings >>>> that contain x=y&a=b data but no mac= >>>> >>>> or you could simply eliminate the sed altogether, but i'm not sure how >>>> curl would handle that, and although i have made the interceptor more >>>> robust to funky inputs that it used to be, it is still probably not robust >>>> enough. >>>> >>>> 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.
