nightscout/intend-to-bolus

Discussions around DIY closed looping with #OpenAPS

cameronrenwick

@cameronrenwick

where are people finding version 14.0.3? When I go to http://nightscout.github.io/pages/update-fork/ and enter my username it says I'm running the newest version (but my version is 13)... is there a secret hiding spot for 14?

Foxy7

@Foxy7

there's a really simple way

https://nightscout.github.io/update/update/

@cameronrenwick

cameronrenwick

@cameronrenwick

lol.. just found that now! thx @Foxy7 !

cameronrenwick

@cameronrenwick

ok so is it possible that the release update tool for nightscout can be wrong? I'm for sure running NS v13 but the tool to see if I'm running the most current version says my repo is up to date... hmm

Foxy7

@Foxy7

i guess it could be wrong, but it's worked for me every time so far.

Carl

@c-robertson

@cameronrenwick Yeah, strange, the update tool worked as expected for me (I was on 13.0.1 when I started). I think ensuring that you log into GitHub with an identity that has a fork of the official NS repo is key, assume you are doing so?

cameronrenwick

@cameronrenwick

well.. good question. I haven't actively used (to my knowledge) a non-official NS repo... I'm gonna dig a bit further

cameronrenwick

@cameronrenwick

ok so I found a fairly quick fix on this. I went to my heroku site, scrolled down to manual deploy, selected "master" in the dropdown box, and clicked deploy. It built the new site (moving from NS13 to NS14) and all is now well. Gotta say, 14 is very peppy... I'm glad as 13 really was boggy on my end. For those still on 13.. upgrade!

Martin Haeberli

@mhaeberli

Interesting project: https://www.bunniestudios.com/blog/?p=5921

Milan Lakhani

@milan.lakhani_gitlab

Got a couple of functional safety questions, if the CGM (or Insulin pump) has been off for 3 hours and then is connected with OpenAPS, and OpenAPS tries to pull for instance Glucose data from the last 40 minutes, might it use the data from 3 hours ago instead? This may not be a big problem with NightScout for the CGM, but could a similar thing happen with the Insulin pump where OpenAPS pulls old operating data in absence of newer data?

Would it be an issue if a bug caused OpenAPS to communicate with the Insulin pump too frequently, or at the wrong time interfering with the insulin pump doing other things?

Eric

@ecc1

@milan.lakhani_gitlab most (all?) sources of CGM data use timestamps, and openaps checks that the rig, CGM, and pump have clocks that are "close enough". Pump history data (used to calculate IOB for example) is similarly timestamped.

Eric

@ecc1

Medtronic pumps give priority to an ongoing bolus and will reject attempts to set temp basal rates while the bolus is in progress. But there is an error (A52) that can be triggered if the bolus wizard is being used at just the wrong time when a remote command comes in. And communication with the pump will drain the battery much faster than non-APS mode

Foxy7

@Foxy7

Hi, i get these messages all the time. And in the past two days my daughter's rig has stopped working/run out of batter just as she's left school, ~15:15 and about 5hr earlier than normal. Is there a way to check the battery status, it seems to not be on NS anymore. I'm using Dev. I do see a battery simbol on NS but it's for the mobile phone with XDrip+ on. is used to be the case i could hover over it and see the phone and the rig battery %.

Sep 24 16:17:33 MealsDev2 pump-loop.log No bolus needed. jq: error: Could not open file monitor/edison-battery.json: No such file or directory
Sep 24 16:22:38 MealsDev2 pump-loop.log jq: error: Could not open file monitor/edison-battery.json: No such file or directory

ArvoNiilo

@ArvoNiilo

I*m trying to flash an old Edison but get stuck here: "---@---Air jubilinux % ./flashall.sh
/bin/bash: ./flashall.sh: Interrupted system call" . Is there any way out or is the Edison ready for the carbage can?

Foxy7

@Foxy7

@ArvoNiilo not sure if you are using PC or MAC, but i've found these tips to work to flash a dodgy rig.

https://openaps.readthedocs.io/en/latest/docs/Resources/Edison-Flashing/PC-flash.html#errors-when-flashing-the-rig

thanksegon

@thanksegon

Hi, folks. I am suddenly dead in the water. Heroku is reporting an error in the logs.

2020-09-25T20:22:14.722251+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/api/v1/status.json?t=1601065333548" host=hankthepank2.herokuapp.com request_id=fafca828-d769-43ff-8626-223d9fe0629c fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:16.552132+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/sw.js" host=hankthepank2.herokuapp.com request_id=05dfff47-c2ed-45b4-9828-4aa1db351517 fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:16.591388+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/socket.io/?EIO=3&transport=polling" host=hankthepank2.herokuapp.com request_id=9cf74ee6-1ca1-4596-97c9-4e72cd0ced43 fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:21.819236+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/socket.io/?EIO=3&transport=polling" host=hankthepank2.herokuapp.com request_id=4bc9efbc-931a-4524-b2d5-3b1a24ba0b1b fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:24.785443+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/api/v1/status.json?t=1601065343611" host=hankthepank2.herokuapp.com request_id=bd49a4be-ab2d-4ce7-a1c2-f63490fc582a fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:26.710144+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/sw.js" host=hankthepank2.herokuapp.com request_id=15a3099f-eb09-443b-8829-0bf93a7e3c51 fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:27.039069+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/socket.io/?EIO=3&transport=polling" host=hankthepank2.herokuapp.com request_id=9aaca117-673c-4177-823f-b053d1ed3c82 fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:32.261412+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/socket.io/?EIO=3&transport=polling" host=hankthepank2.herokuapp.com request_id=b73ac4c3-6405-4399-a8a0-dca60c602e22 fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:32.547397+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/" host=hankthepank2.herokuapp.com request_id=937865ad-44f6-440b-86d2-a06aea6abcc1 fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:37.497710+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/socket.io/?EIO=3&transport=polling" host=hankthepank2.herokuapp.com request_id=505a928b-6b03-4603-8202-758382aff0bf fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:41.086914+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/pebble?units=mgdl&count=19" host=hankthepank2.herokuapp.com request_id=dab79870-b073-4b00-b9e9-0e1382d3878c fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:42.727424+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/socket.io/?EIO=3&transport=polling" host=hankthepank2.herokuapp.com request_id=02d83b75-8599-4217-a234-2c0d50c51833 fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:47.957172+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/socket.io/?EIO=3&transport=polling" host=hankthepank2.herokuapp.com request_id=17931319-3fd8-4deb-9706-5b1f903b9517 fwd="64.223.121.252" dyno= connect= service= status=503 bytes= protocol=https
2020-09-25T20:22:53.178393+00:00 heroku[router]: at=error code=H10 desc="App crashed" method=GET path="/socket.io/?EIO=3&transport=polling" host=hankthepank2.herokuapp.com request_id=cd7b8626-7f03-4604-be73-6405

Does any of this make sense to someone? It is way over my paygrade.

thanksegon

@thanksegon

I took a stab in the dark and changed the AtlasDB user password and that seems to have fixed the problem. Somewhere along the way either Atlas or Heroku munged it up. Not sure how that would have hapened. Very weird.

Dromore1911

@Dromore1911

Hey everyone, I just got the new G6 sensor and I figured since my WiFi wasn't working on my edison rig I oughta reflash and start fresh. Well, that didn;t work out in my favor because everytime i try and run the bootstrap script i get this error W: There is no public key available for the following key IDs:
AA8E81B4331F7F50
W: GPG error: http://archive.debian.org jessie Release: The following signatures were invalid: KEYEXPIRED 1587841717
W: Failed to fetch http://security.debian.org/dists/jessie/updates/main/binary-i386/Packages Hash Sum mismatch

W: Failed to fetch http://security.debian.org/dists/jessie/updates/main/i18n/Translation-en Hash Sum mismatch

E: Some index files failed to download. They have been ignored, or old ones used instead.
Error executing command, exiting
I was told that jessie is too old now, anyone got a way to get this going? I miss my loop

Foxy7

@Foxy7

@Dromore1911 you need to download Jubilinux.
https://openaps.readthedocs.io/en/latest/docs/Resources/Edison-Flashing/mac-flash.html?highlight=debian#preparing-flashing-the-edison-reflashing-the-edison

or PC

https://openaps.readthedocs.io/en/latest/docs/Resources/Edison-Flashing/PC-flash.html#download-jubilinux-and-dfu-util

Dromore1911

@Dromore1911

Ive got a copy, but ill redownload it and reflash again, will update later. Thank you @Foxy7

Dromore1911

@Dromore1911

Okay, got past my first brick wall, now I'm at a new one. Getting this error during oref0 setup

File "/usr/lib/python2.7/json/decoder.py", line 382, in raw_decode
raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded
Could not run nightscout autoconfigure-device-crud

Foxy7

@Foxy7

@Dromore1911
:point_up: August 9, 2020 12:16 PM

you need to run that line then run oref0 setup again

assume you're installing Master, since Dev has a fix in for this now.

Prior to running oref0-runagain.sh run the command pip install parsedatetime==2.5 which allows the setup to complete.

copied it out just in case.

Dromore1911

@Dromore1911

@Foxy7 Thank you, that worked

Foxy7

@Foxy7

:thumbsup:

Dave Acklam

@dcacklam

Something I discovered recently - the 'reboot' and 'shutdown' commands on the rpi-zero are currently borked with the 'latest' (buster) raspbian...

except for reboot --force

If oref0 tries to reboot the system, this results in it hanging indefinately, as systemd fails to reboot properly but also kills sshd - leaving a broken system spinning away, that can only be fixed by a power-cycle.

'Fix' is either (a) revert to sysvinit, or (b) never call any shutdown/init/reboot except for 'reboot --force'....

I did (a) on my rig, but (b) is probably more viable going-forward until the pi people fix raspbian/raspberrypios so that it works....

Would require removing all instances of 'shutdown' and 'reboot' from the code - and replacing with 'sleep 30 && reboot --force &; reboot' or similar - such that the 'reboot' hang is eventually beaten-out by a background 'reboot --force' on time-delay.

@milan.lakhani_gitlab It specifically checks the age of glucose data & takes-no-action if it is too old vs the current time.

Dave Acklam

@dcacklam

@Dromore1911
Edison reflashing is... complicated...
Do you know enough linux to log into your rig & play around with wpa_supplicant, NetworkManager, etc?

Foxy7

@Foxy7

@dcacklam do you know if the hanging can also happen on an Edison rig? Something happened after 21:24 last night, i noticed about 30 mins later and tried to reboot it via putty which didn't work, and eventually turned the power off and on at the socket. this is for a rig that has no battery. The central LED was flashing, but the rig was not connecting to the pump to talk and so not looping. (screenshot above)

i've had other instances of our daytime battery powered rig shutting down around 9pm, ~30 mins after going on charge, but on those occassions the battery rig has no central LED flashing and the rig has shutdown/turned off.

ArvoNiilo

@ArvoNiilo

Hi ! Is there any other fix for the following problem than reflashing the rig ? My Edison (Master 0.70) that has worked impeccably from 2018 failed yesterday due to disc space problems that I have encountered for the first time last week (after NS migration to Atlas in fact). I tried oref0-rerun.sh and then to reinstall + setup OpenAps dev (to avoid Python problems that have invalidated my other two rigs.
Both paths end up in the following message:
“Generating locales (this might take a while)...
en_IE.UTF-8... done
Generation complete.
Reading package lists... Error!
E: Unable to parse package file /var/lib/apt/lists/archive.debian.org_debian_dists_jessie_main_i18n_Translation-en (1)
E: The package lists or status file could not be parsed or opened.
Couldn't install bc etc.“

The loop fails as follows.
“Preflight OK.
Attempting to retrieve MDT CGM data from pump
MDT CGM data retrieved
Profile less than 60m old; Profile valid. touch: failed to get attributes of ‘monitor/glucose.json’: No such file or directory
Couldn't touch /tmp/pump_loop_enacted -r monitor/glucose.json
oref0-pump-loop failed. pump_loop_completed more than 15m old; waiting for 32 s silence before mmtuning
Listening for 32s: .No interfering pump comms detected from other rigs (this is a good thing!)
Continuing oref0-pump-loop at Wed Sep 30 10:31:59 EEST 2020
Listening for 32 s silence before mmtuning: Listening for 32s: .No interfering pump comms detected from other rigs (this is a good thing!)
Continuing oref0-pump-loop at Wed Sep 30 10:32:32 EEST 2020
mmtune: "868.500", 3, -80 -- "868.550", 3, -85
waiting for 50 second silence before continuing
Listening for 50s: .No interfering pump comms detected from other rigs (this is a good thing!)
Continuing oref0-pump-loop at Wed Sep 30 10:33:43 EEST 2020
Done waiting for rigs with better signal.
If pump and rig are close enough, this error usually self-resolves. Stand by for the next loop.
Unsuccessful oref0-pump-loop at Wed Sep 30 10:33:43 EEST 2020”

Foxy7

@Foxy7

@ArvoNiilo do you know what version of debian you are using? i see the word jessie above which makes me think it is too old, if that's the case you need to install jubilinux. if it is jessie, then follow the steps in the docs
:point_up: September 28, 2020 11:54 AM

if it's not, then the only other thing i've read recently, possibly on the AndroidAPS page, was that during migration to Atlas someone's database password got corrupt and after they changed this and it fixed it.

Dromore1911

@Dromore1911

@dcacklam Yes ive been playing around with this system since 2018 and its taught me how to navigate linux. I actually just got my tslimx2 yesterday though so I may be retiring my rig

Saskia

@Missdazzle

Friendly reminder: The OPEN survey on open-source loop systems is waiting for you to participate! Patient advocates could use the results to influence health policy-makers! And it would help to convince your HCPs that this is the real thing!

Where communities thrive

People

Repo info

Activity