These are chat archives for nightscout/intend-to-bolus

9th
Jun 2018
garykidd
@garykidd
Jun 09 2018 00:53
@scottleibrand is there an easy spot other than at the jst connector to which I can solder my battery leads
?
Zakm0n
@Zakm0n
Jun 09 2018 02:12
Jun 08 21:07:03 Aps pump-loop.log:  Starting oref0-pump-loop at Fri Jun  8 21:07:02 CDT 2018 with 7 second wait_for_silence:
Jun 08 21:07:06 Aps logger-loop.log:  06/08 21:07:06 Sending BG of 121 to pump via meterid 000000
Jun 08 21:07:10 Aps logger-loop.log:  2018/06/08 21:07:10 waking pump
Jun 08 21:08:14 Aps pump-loop.log:  Waiting up to 4 minutes for new BG: .......glucose.json newer than pump_loop_completed
Jun 08 21:08:44 Aps pump-loop.log:  mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?
Jun 08 21:08:46 Aps pump-loop.log:  Radio check failed. mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?
So, i'm using logger with a g5 and oref0.7.0 dev
and this is bad.
Fakemeter is crashing my radio
alimhassam
@alimhassam
Jun 09 2018 09:46
Seems like you still had code from dev running
fidoman
@efidoman
Jun 09 2018 10:37
@Zakm0n - I'm using Logger with g5 and oref0.7.0 dev on both edison/exp and rpi hat boards as well and fakemeter has never done this to my systems. It does seem coincidental though. Did you try reflashing the radio on your edison/exp?
fidoman
@efidoman
Jun 09 2018 10:49
Another question ... Are you using cgm type of xdrip? The reason I ask is because Logger is sending a BG and openaps is showing that it isn't seeing that BG. That is an error that is outside of any radio communications to the pump. I only see that happen when the xdripAPS connection from Logger to Openaps isn't setup right or isn't working. I believe you just did a oref0 setup reinstall. Perhaps, it didn't setup xdripAPS right? or cgm type in runagain.sh isn't set to xdrip?
fidoman
@efidoman
Jun 09 2018 10:55
@scottleibrand - That gitter download bash script is awesome. I'm going to do that for a few gitter channels so I can grep search. Thanks!
Zakm0n
@Zakm0n
Jun 09 2018 11:06
Cgm type is definitely xdrip. I removed fakemeter and the rig happened flawlessly overnight
fidoman
@efidoman
Jun 09 2018 11:13
@Zakm0n - OK, seems like we may need @ecc1 assistance here.
alimhassam
@alimhassam
Jun 09 2018 11:23
@Zakm0n make sure you run the steps to change branch correctly, seems like you still have old mmeowlink code running from the above log you posted
Or verify that you have only one loop running
john36m
@john36m
Jun 09 2018 12:25
Running 2 rigs at home. My Edison Explorer combo that I have been using for quite some time, and my newly assembled Pi Zero with HAT, running 0.7.0-dev. The odd thing I want to report today is the battery charge percent shown on Nightscout. This morning after I unplugged the Edison rig from its charging station and came to the kitchen to eat breakfast it showed 2% remaining. I said to my self, oh well old battery let me replace it. So I replaced it and the alternate battery showed 66% at the first reading and then 5 minutes later it was down to 22% and now it is also at 2%. So, is this a bizarre coincidence of 2 batteries failing or is it a software issue? When the Edison is plugged in it shows the usual 66% give or take. And is it "our problem" or a Nightscout problem?
richardshane
@richardshane
Jun 09 2018 13:24
When I transfer an Edison that was successfully running openaps to a new explorer board, where should I expect to have to pick up in the setup process? Reinstall Jubilinux? Bootstrap? Setup script? Somewhere else? Thanks.
Zakm0n
@Zakm0n
Jun 09 2018 14:07
@alimhassam I only have one rig. I followed the instructions on the page to change branch to a t, but I could try it again.
alimhassam
@alimhassam
Jun 09 2018 14:09
@Zakm0n Did you do all the steps including npm run global-install and re running oref0-setup? Are you sure you don't have 2 myopenaps directories? Note 0.7.0-dev is not recommended to be used yet on your main rig unless you're able to closely monitor it
David
@david-t1
Jun 09 2018 14:09
A52
fidoman
@efidoman
Jun 09 2018 14:57
@john36m - I saw the same thing. It turns out that both rigs were reporting battery still to NS at the same time. If you hover over the battery in NS, you may see two rigs reporting. That's what happened on my system. I haven't figured out an easy way to switch back and forth, so I end up just commenting out everything in crontab on the one I'm not using and the dual battery stuff in NS goes away.
john36m
@john36m
Jun 09 2018 15:47
@efidoman When I left the premises with one rig, my Edison, the battery readings were normal. I just unplugged the Pi Hat, and again the readings are normal. I will try your crontab fix, if I can learn how to do it. For now, the PiHat is NOT on battery and I leave it plugged in for nighttime use, and for range extension. so if I can turn off the feature, everything will be fine for now.
Dana Lewis
@danamlewis
Jun 09 2018 15:58
@richardshane you don't have to do anything other than power it on :smile:
NSewar
@nsewar01
Jun 09 2018 16:57
what does the on off switch on the PiHat rig do?
Martin Haeberli
@mhaeberli
Jun 09 2018 17:09
@nsewar01 - think of it as : if it’s not plugged into the wall, but just to the battery, it turns it on / off ; if it IS plugged into a wall, more or less nothing ...
(i think)
Scott Leibrand
@scottleibrand
Jun 09 2018 18:01
it's simpler to think of it as a battery disconnect switch
it just cuts electricity to/from the battery. so if you're plugged into a wall, turning the switch off prevents the battery from charging
richardshane
@richardshane
Jun 09 2018 18:19
@danamlewis yes, it was just that easy.....woohoo!
Martin Haeberli
@mhaeberli
Jun 09 2018 19:40
@scottleibrand :+1:
NSewar
@nsewar01
Jun 09 2018 20:16
@scottleibrand
thks
N3FM
@N3FM
Jun 09 2018 23:41
I'm looking into the 'Warning: clock input Invalid Date is unzoned' message that another user also reported. I traced it back to the /lib/iob/index.js block, where it appears to be going through 'inputs' records and finding an invalid date message instead of a date string. 1) Is this something that solves itself over time? 2) Has anyone else dug into this? 3) Where is that data stored and what code block writes it? (Minimed 722/Enlite CGM/Edison/Explorer on both 0.6.1 and 0.6.2-dev)
'''
No smb_enact needed. Temp refreshed: monitor/temp_basal.json: {"duration":118,"rate":0,"temp":"absolute"}
No bolus needed. Settings less than 15 minutes old. Edison on battery: 80%. pumphistory-24h refreshed
Pump profile refreshed; Could not parse autotune_data
No temptargets found.
Settings refreshed; Refreshed pumphistoryWarning: clock input Invalid Date is unzoned; please pass clock-zoned.json instead
Warning: clock input Invalid Date is unzoned; please pass clock-zoned.json instead
Warning: clock input Invalid Date is unzoned; please pass clock-zoned.json instead
Warning: clock input Invalid Date is unzoned; please pass clock-zoned.json instead
and meal.json
IOB: 10.037
Your instance of oref0 [0.6.2-dev, dev] is up-to-date.
Completed oref0-pump-loop at Sat Jun 9 19:38:16 EDT 2018
'''
The number of 'Warning' lines is increasing -- I'm assuming I am getting more lines in the inputs file.