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

5th
Apr 2018
Dana Lewis
@danamlewis
Apr 05 2018 00:03
I've been thinking about ways to log settings, maybe at midnight every night, somehow in NS in a way that makes easy to look back and analyzes
Right now I think you'd have to log changes in care portal
Manually, which of course is not ideal
live4sw
@live4sw
Apr 05 2018 00:09
@danamlewis I saw your tweet about hormones & bg. Unrelated to that I was curious, how do you have two sets of BGs in your NS? two receivers both uploading from the same sensor? Two receivers from different sensors?
Dana Lewis
@danamlewis
Apr 05 2018 00:14

Reminder - I have a proposed fix for the log rotate stuff. I merged it to dev, so feedback on new dev installs regarding that welcome. It's also an open Pr for master here, for anyone to test: openaps/oref0#962

Once that gets tested and evaluated, we can do a patch release. Should solve some of the full rig issues some people have been encountering

@live4sw yea - I run 2 receivers off the same sensor. They drifted so needed calibration which is why it's so obvious
I have one plugged in to a rig, and the other uploads via share
live4sw
@live4sw
Apr 05 2018 01:45
Ah, nice! Thx for the clarification.
Dana Lewis
@danamlewis
Apr 05 2018 01:46
:+1:
Dave Acklam
@dcacklam
Apr 05 2018 01:53
@efidoman wouldn't settings/profile.json have the current in-effect curve in it (since it more or less contains all of the in-effect config values)?
fidoman
@efidoman
Apr 05 2018 09:47
:+1: It does confirm that openaps is using ultra-rapid curve, thanks for the suggestion.
KensingtonB
@KensingtonB
Apr 05 2018 17:13
having trouble trying to update a rig that stopped working. when doing fresh install, getting this error:
Sorry, here is the error:~/src/oref0# git fetch
fatal: Not a git repository (or any of the parent directories): .git
Worried my file sizes are too large, but not sure how to fix that... 118M ./root/src/decocare
313M ./root/src
349M ./root
148M ./var/log/openaps
725M ./var/log
825M ./var
1.3G .
Scott Leibrand
@scottleibrand
Apr 05 2018 17:16
no, that's just because ~/src/oref0 itself is not currently a git repo
probably easiest to delete and re-clone it. cd ~/src/ && rm -rf oref0 && git clone https://github.com/openaps/oref0.git
KensingtonB
@KensingtonB
Apr 05 2018 17:20
Thanks!
Mike
@libxmike
Apr 05 2018 17:26
@danamlewis are you still using G4 transmitter?
Scott Leibrand
@scottleibrand
Apr 05 2018 17:34
(Dana's on a call) yes, she is. G4 gives us more control over sensor restarts and looping off raw data when appropriate.
KensingtonB
@KensingtonB
Apr 05 2018 17:42
Hmmm. Got this error. Will try again. module.js:340
throw err;
Error: Cannot find module 'oref0/lib/profile/'
at Function.Module._resolveFilename (module.js:338:15) ...etc.
No, same error
module.js:340
throw err;
^
Error: Cannot find module 'oref0/lib/profile/'
at Function.Module._resolveFilename (module.js:338:15)
at Function.Module._load (module.js:280:25)
at Module.require (module.js:364:17)
at require (module.js:380:17)
at Object.<anonymous> (/root/src/oref0/bin/oref0-get-profile.js:19:16)
at Module._compile (module.js:456:26)
at Object.Module._extensions..js (module.js:474:10)
at Module.load (module.js:356:32)
at Function.Module._load (module.js:312:12)
at Function.Module.runMain (module.js:497:10)
Could not run oref0-get-profile
Mike
@libxmike
Apr 05 2018 17:51
@scottleibrand did you consider G5 and xDrip+?
Dana Lewis
@danamlewis
Apr 05 2018 17:51
@libxmike no.
Mike
@libxmike
Apr 05 2018 17:52
I'm testing for last 11 days with G5 my first Dexcom sensor
99% capture rate
upload to two NS instances (one for Combo testing)
Dana Lewis
@danamlewis
Apr 05 2018 17:52
I have a lot of issues with phones connecting to receivers when flying (e.g. the app time sync's and freaks out and won't reconnect unless you get internet and the phone gets it's clock back) that I know it would be a problem as well with G5, and I see so many people with issues getting their G5 transmitter to connect to their phone. Between that, still wanting raw, and wanting to hot swap and restart etc, I choose to keep using G4
Mike
@libxmike
Apr 05 2018 17:54
I see, I'm using the G5 (refurbished) with Galaxy S8 and not a single problem with connection
Dana Lewis
@danamlewis
Apr 05 2018 17:55
glad you found something that works well for you! :+1:
Mike
@libxmike
Apr 05 2018 17:55
aslo connecting openaps over BT to same phone for 8h a day
Scott Leibrand
@scottleibrand
Apr 05 2018 18:00
@KensingtonB did you do cd ~/src/oref0 && npm run global-install after re-cloning?
KensingtonB
@KensingtonB
Apr 05 2018 18:06
Yes, but I can try again.
Bunch of errors, ended by "not OK" error. How do you do the highlighted text so its easier to read copy/pasted results?
DaisyHK
@laurathiessen
Apr 05 2018 18:11
sorry guys but trying to buy reaspberry pi in prep for the new hats and such. Can someone tell me which pi I will need? I understand the hat is from hamshield and you need a raspberrry pi wireless WH like this one is what i need for my mobile system https://www.buyapi.ca/product/raspberry-pi-zero-wireless-wh-pre-soldered-header/ but what are the details for the pi0 or pi 3, such as does it need to be zero or zero W? have also heard rumblings that the raspberry pi model B is what you need.... getting so confused with all these options, and on facebook they indciate to ask on gitter for details... or will @danamlewis have another post soon with these details and I should just be patient?
Dana Lewis
@danamlewis
Apr 05 2018 18:12
I'd wait for the full post that'll explain pros/cons of each, if it were me :) because those will have links, batteries, etc
Pi's are readily available, so there won't be a big time crunch by waiting
DaisyHK
@laurathiessen
Apr 05 2018 18:12
@danamlewis oh ok... any idea of when that will be.... as you can tell i am super excited/imptient lol
k I will just grab a hat then? cause that wont change correct?
Dana Lewis
@danamlewis
Apr 05 2018 18:13
:) probably within a few weeks, by the time HAT is ready
yes, feel free to pre-order the HAT
that won't change
DaisyHK
@laurathiessen
Apr 05 2018 18:13
k... darn.... my impatience will probably get the better of my BG soon lol ;)
KensingtonB
@KensingtonB
Apr 05 2018 18:16
Thanks, hope this works to show the error messages
> oref0@0.6.1-dev global-install /root/src/oref0
> npm install && sudo npm install -g && sudo npm link && sudo npm link oref0

Killed

npm ERR! oref0@0.6.1-dev global-install: `npm install && sudo npm install -g && sudo npm link && sudo npm link oref0`
npm ERR! Exit status 137
npm ERR! 
npm ERR! Failed at the oref0@0.6.1-dev global-install script.
npm ERR! This is most likely a problem with the oref0 package,
npm ERR! not with npm itself.
npm ERR! Tell the author that this fails on your system:
npm ERR!     npm install && sudo npm install -g && sudo npm link && sudo npm link oref0
npm ERR! You can get their info via:
npm ERR!     npm owner ls oref0
npm ERR! There is likely additional logging output above.
npm ERR! System Linux 3.10.98-poky-edison
npm ERR! command "/usr/bin/nodejs" "/usr/bin/npm" "run" "global-install"
npm ERR! cwd /root/src/oref0
npm ERR! node -v v0.10.29
npm ERR! npm -v 1.4.21
npm ERR! code ELIFECYCLE
npm WARN This failure might be due to the use of legacy binary "node"
npm WARN For further explanations, please read
/usr/share/doc/nodejs/README.Debian

npm ERR! 
npm ERR! Additional logging details can be found in:
npm ERR!     /root/src/oref0/npm-debug.log
npm ERR! not ok code 0
Scott Leibrand
@scottleibrand
Apr 05 2018 18:17
@libxmike we also have some serious concerns about trusting a completely DIY CGM receiver as our only way to tell if BG is dangerously low. with the official Dexcom receiver, you can trust it to give you a < 55 mg/dL alarm if anything goes wrong, and their algorithm is pretty idiot-proof. we've seen issues where xDrip+ can be miscalibrated and result in dangerous situations, so I'd always recommend having a non-DIY source of data in parallel to anything else you want to do. that isn't always possible (for example if you can't afford to buy new Dexcom transmitters every 6 months), but we highly recommend prioritizing safety over convenience when it is possible.
Scott Leibrand
@scottleibrand
Apr 05 2018 18:23
@KensingtonB it might be easiest to just reflash jubilinux at this point: it seems you have some node version weirdness, and that's harder to troubleshoot than just to start over
KensingtonB
@KensingtonB
Apr 05 2018 18:24
Ok, thanks.
Martin Haeberli
@mhaeberli
Apr 05 2018 18:24
any suggestions on how to fix an E17 in a 512 pump (or was it maybe a 522)? seriously - asking for a friend
Scott Leibrand
@scottleibrand
Apr 05 2018 18:25
does that error have a description? I don't remember the error codes
Mike
@libxmike
Apr 05 2018 18:26
@scottleibrand I have different opinion, most people here in EU I know using xDrip+ as collector with different sensors, Eversence, Libre or Dexcom w/o issues , I used for over a year Libre with DIY transmitter with openAPS and after my G5 test probably will go back to it or use my refurbished G5 if the Dex sensor will last at least 3 weeks.
Martin Haeberli
@mhaeberli
Apr 05 2018 18:27
I don’t know - this article maybe suggests - I’ll forward to my friend...
and found this article
(friend is trying to start looping on that pump, which is the right firmware version, but…)
Scott Leibrand
@scottleibrand
Apr 05 2018 18:32
@libxmike I don't doubt that xDrip+ works better most of the time. my concern is with those rare cases where it does something unexpected, it's better to have a completely independent system that can give you a heads up something is wrong. it's kinda similar to the two recent self-driving fatalities: it doesn't matter if your system is better most of the time, if you don't have a working redundant system when it inevitably fails, and don't alert the human that it's not working properly, you can get into a very dangerous situation.
Mike
@libxmike
Apr 05 2018 18:37
I trust more xDrip+ as Dexcom as I can have few handsets as failover, as you know in xDrip you have alarms, but my opinion rests not on experience with Dexcom receiver as I never had one, BUT all what we do here, openAPS or AAPS (as I'm testing) is DIY, all what I can do with xDrip I could not with Dexcom
Paul Heltzel
@pheltzel
Apr 05 2018 18:47
Hi all. I'm seeing major spikes at mealtime and still after adjusting up basal schedule with autotune. My spikes (180-200) are lasting for ~6 hours even with very little carb intake. I'm frequently seeing a 0 temp basal right after a meal and then my BG goes up very quickly and stays there.
spike.jpg
autotune-pump-settings.png
I'm wondering if enabling UAM/SMB might be worth trying. My first thought was to turn on %basal and just focus on getting my basal rate up. It was more like 1U most of the day, with 1.5U from 4am to 12 pm -- and my BGs were pretty well controlled.
Scott Leibrand
@scottleibrand
Apr 05 2018 19:02
yes, turning on UAM would probably help there.
Paul Heltzel
@pheltzel
Apr 05 2018 19:07
I think I might step back, just make sure my basal schedule is working for me and then come back using UAM/SMB.
I'm just looking at a snapshot where I'd normally have 1.5 U/hr basal and it's 0, but with a bolus that looks right. I feel like I'm missing that basal, and if I'm at around 100mg/dL when I eat there's very little spike.
I'll take a breather and dig into the docs more a bit on oref1. Thanks again!
Scott Leibrand
@scottleibrand
Apr 05 2018 19:16
:+1:
a lot of people focus in on post-meal zero basal as being the cause of any later highs; it's apparently not very intuitive (at least given what/how we visualize insulin) to understand the relative impacts of basal insulin vs. bolus insulin activity, and to understand how insulin and carb absorption timing relate.
Paul Heltzel
@pheltzel
Apr 05 2018 19:27
Sure, that makes sense. And in some cases when I'm looking at the treatment schedule the basal rate was 2x where I'd normally have it set shortly before.
I'm just spiking and staying spiked and can't figure it out.
Martin Haeberli
@mhaeberli
Apr 05 2018 19:44
so - trying to use an rpi to flash a new edison. did sudo /etc/init.d/dphys-swapfile stop ; that was fast ; did 1sudo /etc/init.d/dphys-swapfile start - took forever, never finished - so I rebooted the pi. now have a sad pi that doesn’t boot. Advice sought
Martin Haeberli
@mhaeberli
Apr 05 2018 20:26
memo to self - run df -h first to make sure there is room for the swapfile!
[and memo to write doc PR re same]
James Babcock
@jimrandomh
Apr 05 2018 20:46
@pheltzel In the screenshot there, zero-basal took off about 1U of insulin (based on its predictions and model of how quickly carbs are absorbed); by comparison, over the preceding three hours you had 36.2U of insulin from meal boluses.
That rise looks like something where, in order to prevent it, you'd want to bolus a bit more in advance of when you eat, or eat lower-glycemic-index meals
Paul Heltzel
@pheltzel
Apr 05 2018 21:04
@jimrandomh thanks, James. I've barely been eating any carbs. Mostly lean meat and vegetables. But my BGs are staying around 200. I took a long walk today. No carbs at lunch though I bolused and planned to wait. Can't get my NG below 200 all day.
Even just some lunch meat is making my bg spike like crazy. Can't figure it out. I'm going to get back in range and come back to it when I've had some time there.
My bgs are typically 85% in range. 6.8 A1c or thereabout always. The 722 is new to me but I thought I had the basal scheduled dialed in. Took a little adjustment from omnipod.
Juan Mejías
@juanjuanmejias_twitter
Apr 05 2018 21:15
image.png
Still far from perfect, but I'm looping!
Scott Leibrand
@scottleibrand
Apr 05 2018 21:16
:+1:
Juan Mejías
@juanjuanmejias_twitter
Apr 05 2018 21:16
Thank you everybody who's been helping me get here
Martin Haeberli
@mhaeberli
Apr 05 2018 21:16
:+1:
statuur
@statuur
Apr 05 2018 21:17
Hi all, i'me trying to setup a 'read only' configuration with an edison and explorer block since my daughters pump isn't compatible. openaps first-upload works but after that the oref0-pump-loop.sh script doesn't update glucose readings from the sensor. And obviously quite a lot of error messages. Where to ask for technical assistance?
Martin Haeberli
@mhaeberli
Apr 05 2018 21:17
uname -a Linux jubilinux 3.10.98-poky-edison #5 SMP PREEMPT Tue Apr 4 22:42:44 EDT 2017 i686 GNU/Linux looks to be jubilinux-v0.1.1 - any reasons to use v0.2.0 instead ?
Scott Leibrand
@scottleibrand
Apr 05 2018 21:17
@statuur is this using Enlite? configure for MDT CGM?
@mhaeberli nope, either is fine
Martin Haeberli
@mhaeberli
Apr 05 2018 21:18
k thx - what benefits to 0.2.0
statuur
@statuur
Apr 05 2018 21:18
@scottleibrand yes
Scott Leibrand
@scottleibrand
Apr 05 2018 21:18
@mhaeberli really none for OpenAPS
Martin Haeberli
@mhaeberli
Apr 05 2018 21:18
:+1:
statuur
@statuur
Apr 05 2018 21:19
@scottleibrand i have an own hosted nightscout setup @ http://109.71.51.137
Scott Leibrand
@scottleibrand
Apr 05 2018 21:19
This message was deleted
@statuur might be best to start by pasting an appropriate pump-loop.log snippet
from where it's trying to refresh the CGM data from the pump
statuur
@statuur
Apr 05 2018 21:22

@scottleibrand
Radio ok. Listening: .No interfering pump comms detected from other rigs (this is a good thing!)
Preflight OK. Done waiting for rigs with better signal.
Unsuccessful oref0-pump-loop at Thu Apr 5 23:09:57 CEST 2018
Error: pump is set to % basal type. The pump won’t accept temporary basal rates in this mode. Please change the pump to absolute u/hr so temporary basal rates will then be able to be set.

Starting oref0-pump-loop at Thu Apr 5 23:10:01 CEST 2018 with 13 second wait_for_silence:
MDT CGM configured; not waiting
Radio ok. Listening: .No interfering pump comms detected from other rigs (this is a good thing!)
Preflight OK.
Attempting to retrieve MDT CGM data from pump
MDT CGM data retrieved
MDT New cgm data reformatted
Profile less than 60m old; Profile valid. RefreshedWarning; adjusting DIA from 2 to minimum of 3 hours
Warning; adjusting DIA from 2 to minimum of 3 hours
<!--- Above Line repeats about 200 times each loop -->
Warning: Autotune has not been run. All microboluses will be disabled until you manually run autotune or add it to run nightly in your loop.
{"carbs":45,"nsCarbs":45,"bwCarbs":0,"mealCOB":0,"currentDeviation":2,"maxDeviation":4.36,"minDeviation":0.33,"slopeFromMaxDeviation":-0.999,"slopeFromMinDeviation":0.21,"allDeviations":[4,4,3,2,2,1,0],"lastCarbTime":1522948969000,"bwFound":false}
{"iob":0.419,"activity":0.0024,"basaliob":0,"bolusiob":0.419,"netbasalinsulin":0,"bolusinsulin":0.475,"time":"2018-04-05T21:18:52.000Z","iobWithZeroTemp":{"iob":0.419,"activity":0.0024,"basaliob":0,"bolusiob":0.419,"netbasalinsulin":0,"bolusinsulin":0.475,"time":"2018-04-05T21:18:52.000Z"},"lastBolusTime":1522960793000,"lastTemp":{"date":0}}
{"delta":0,"glucose":162,"short_avgdelta":1,"long_avgdelta":2.22,"date":1522962720000}
Autosens ratio: undefined; ISF from 102.6 to NaN; CR: 16
currenttemp: { duration: 0, rate: 0, temp: 'percent' } lastTempAge: 25382720 m tempModulus: 20 m
Error: could not calculate eventualBG.
Checking deliverAt: is within 1m of current time: Thu Apr 5 23:19:35 CEST 2018
date: option requires an argument -- 'd'
Try 'date --help' for more information.
Couldn't smb_verify_suggested
oref0-pump-loop failed. pump_loop_completed more than 15m old; waiting for 40s silence before mmtuning
Radio ok. Listening: .No interfering pump comms detected from other rigs (this is a good thing!)

Scott Leibrand
@scottleibrand
Apr 05 2018 21:23
so to fix the Warning; adjusting DIA from 2 to minimum of 3 hours thing, you can change your pump's DIA in the bolus wizard settings
Juan Mejías
@juanjuanmejias_twitter
Apr 05 2018 21:23
@cluckj - In case you're curious, the problem with the radio came from the pump. I managed to get hold of a 722 (again, thanks @marafie!), and as soon as I tried mmtune I got a signal of -37 . After that, no more issues :smile:
Scott Leibrand
@scottleibrand
Apr 05 2018 21:27
@statuur it's complaining that the pump is set to percentage temp basal mode. not sure if changing that to U/hr mode on the pump would help, but it might get it further along
statuur
@statuur
Apr 05 2018 21:29
@scottleibrand tnx! i'me a little hesitative about changing that. Have to find out how to 'convert' the existing settings to the same U/hr settings in order to have the same effect.
by the way, i'me so Grateful to the developers of Openaps. what an mazing job!
Juan Mejías
@juanjuanmejias_twitter
Apr 05 2018 21:31
@danamlewis saw it coming...
I think that's related to the pump (an x12), not the Edison
Scott Leibrand
@scottleibrand
Apr 05 2018 21:33
@statuur do you use temp basals much?
the temp basal type setting just controls the way you enter temp basals: it doesn't affect your existing settings: that schedule is already in U/hr
statuur
@statuur
Apr 05 2018 21:34
@scottleibrand never used it before
Martin Haeberli
@mhaeberli
Apr 05 2018 21:35
really liking the new general wifi etc setup script for new Edison : at http://openaps.readthedocs.io/en/dev/docs/Build%20Your%20Rig/edison-install.html
Scott Leibrand
@scottleibrand
Apr 05 2018 21:35
@statuur then I don't think you'll notice the effects of the change on your normal use of the pump
statuur
@statuur
Apr 05 2018 21:36
@scottleibrand i'll adjust the settings tomorrow and monitor the effect. tnx!
Scott Leibrand
@scottleibrand
Apr 05 2018 21:36
:+1:
Martin Haeberli
@mhaeberli
Apr 05 2018 21:56
#ntp-wait -n 1 -v Waiting for ntpd to synchronize... No! ntpd did not synchronize - wondering why since internet works ...
Scott Leibrand
@scottleibrand
Apr 05 2018 22:04
what kind of Internet connection are you on? any chance it's partially filtered?
James Babcock
@jimrandomh
Apr 05 2018 23:25
@pheltzel Is the cannula age in the screenshot accurate? If so that might be it
Martin Haeberli
@mhaeberli
Apr 05 2018 23:41
NTP sync works now ...
when running , say, 3 rigs, maybe not all are in the same place (maybe one in bedroom, primary and spare carried along) what if anything needs to be done to keep autotune data harmonized?
Scott Leibrand
@scottleibrand
Apr 05 2018 23:44
if all three start out the same and run every night, they should come up with the same answers. since sometimes ours get turned off, we have an rsync cron job, but setting that up (if you want it) is going to be specific to your environment
and even if they do get out of sync, they should gradually converge on similar results. so you don't have to do anything unless you want to make multi-rig debugging easier.
Martin Haeberli
@mhaeberli
Apr 05 2018 23:50
i’m not yet ready to make ‘all three start out the same’. can I find the rsync cron job via google in rtd? (thinking of having a cloud service be the target / master for rsync ing…)
Scott Leibrand
@scottleibrand
Apr 05 2018 23:51
no, I didn't document any of the rsync stuff as it's done via my personal server
Martin Haeberli
@mhaeberli
Apr 05 2018 23:51
cn u send rsync script and i’ll parse it out somehow?
Scott Leibrand
@scottleibrand
Apr 05 2018 23:53
this copies everything (not just autotune stuff) up to the server: * * * * * rsync -e "ssh -4" -rtuv --exclude ".*/" --min-size=2 /root/myopenaps/ me@myserver:~/myopenaps/
55 23 * * * (mkdir -p /root/myopenaps/autotune/; rsync -e "ssh -4" -rtuv --exclude ".*/" --min-size=2 me@myserver:~/myopenaps/autotune/ /root/myopenaps/autotune/ && cp /root/myopenaps/autotune/profile.json /root/myopenaps/settings/autotune.json) 2>&1 | tee -a /var/log/openaps/autotune.log
and ^^ that copies down the autotune directory each night before running autotune
could make that 55 3 now to run at 3:55am instead of 11:55pm - I never updated that one after changing autotune to run at 4am
you could also adapt those to run between rigs if you don't have a server
and you'd have to figure out passwordless ssh access between them via ssh keys
Martin Haeberli
@mhaeberli
Apr 05 2018 23:57
:+1: already have passwordless … for tunnels thx