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

24th
Oct 2018
Niels Hartvig
@hartvig
Oct 24 2018 00:09
Just installed on a fresh Pi tonight. Using @mhaeberli tip with removing nodejs after rebooting from the piBakery image did the trick - thank you :pray:
And it's successfully looping now. Is there anyway to sync autotune from one rig to the other?
Niels Hartvig
@hartvig
Oct 24 2018 00:15
@danamlewis You've thought of everything <3 - I swear I did search the docs : o )
Dana Lewis
@danamlewis
Oct 24 2018 00:15
😅👍🏼
so, while I'm waiting for new one, does carelink stick works with latest dev oref0? Hat works on few centimeters distance.
Daniel
@rofid12_twitter
Oct 24 2018 09:01
I've managed to resolve the SD-issue, I just shouldn't touch the partitions. But now I can't for the life of me manage to connect to the pi. I keep getting an error that the host doesn't exist. I've given it a good 5-10 minutes to boot up. I haven't attached the hat board yet though, I'm waiting for the installation to complete before I do that. https://imgur.com/a/4mdaM0q has 3 screenshots in them with my setup in PiBakery, my putty setup and the error.
Daniel
@rofid12_twitter
Oct 24 2018 09:20
Also might be of importance: when I format the sd card and try to write to it using PiBakery it gives me an error that it can't find the SD card when the writing is at like 99%. But after I try it again it works. But when I look in disk manager only 44Mb of the SD card is actually filled. Is this right or is something going wrong?
My device manager: https://imgur.com/a/YIRz27x
Scott Leibrand
@scottleibrand
Oct 24 2018 14:49
@isuran no, you’ll need to use 0.6.x for Carelink.
tzachi-dar
@tzachi-dar
Oct 24 2018 16:01
I have a question regarding basal:
I have set it manualy on my rpii (x12 pump). but it seems that this does not have any effect on what nightscout is showing. (NS is showing it's own values). Should I set the values on NS manually?
Eric
@ecc1
Oct 24 2018 16:09
@tzachi-dar I think NS uses whatever profile you've set there, and it's up to you to keep it in sync with your pump
This is mentioned in the NS setup section of the openaps docs
tzachi-dar
@tzachi-dar
Oct 24 2018 16:16
Thanks Eric.
Daniel
@rofid12_twitter
Oct 24 2018 18:21
Does anyone have experience with this error? Got this while running "npm run global-install"
https://imgur.com/a/iAkrFy2
Eric
@ecc1
Oct 24 2018 19:30
Unpopular opinion: requiring PWDs to be embedded Linux sysadmins is not a scalable or sustainable approach to closed looping
Tore Bjørndalen
@BjoTore_twitter
Oct 24 2018 19:33
@ecc1 Totally disagree. "Learning by doing" forces you to be a "GOOD" sysadmin :)
Or.. did I totally miss your point...??
Eric
@ecc1
Oct 24 2018 19:39
Emphasis on the "scalable" part. I think it fundamentally limits the addressable community. I'm not arguing we should be doing anything differently right now, except perhaps do a better job of setting expectations than "anyone can do it"
Raymond Richmond
@PedanticAvenger
Oct 24 2018 19:45
I agree with @ecc1 . While it isn't the most high-end technical requirements it is certainly not reasonable for a significant portion of the PWD community.
I recently did a talk at a local gathering put on my my endos office and there was at least 30 to 40% of the audience who completely glazed over explaining how the pieces work together. Some still can't quite come to terms with carb counting.
Martin Haeberli
@mhaeberli
Oct 24 2018 20:44
asking again - getting frequent NewTimeSet / ChangeTimes on our 723. but not clear that the pump time was off by much, or at all; almost as if the base time used for comparison is one from a few minutes earlier, not the current pump time. I think I remember some posting here in ITB about this and about some oref0 dev or other branch that may fix this (likely running on Edisons now on master) - suggestions welcome
@rofid12_twitter yes - your error is frustratingly familiar - upstream a few weeks ago I commented on this and recently someone else acknowledged my ‘contribution’ - I say ‘contribution’ because I explained (but have since forgotten) how to remove node and then re-install.
@ecc1 - I agree - while linux sysadmin - ish is fun for some, it’s way unduly hard for many.
Daniel
@rofid12_twitter
Oct 24 2018 20:48
@mhaeberli Rerunning the script fixed it somehow, I got past the error. So I'm guessing it'll be okay.
Martin Haeberli
@mhaeberli
Oct 24 2018 20:48
@rofid12_twitter glad to hear it!
Dana Lewis
@danamlewis
Oct 24 2018 20:48

@ecc1 I (always respectfully) disagree. I talk to many people who fear they are non-technical and can't do it. And by non-technical, some aren't even on Facebook because that sounds too hard to them. With encouragement, they can try (if they decide to, and are confident in their diabetes management), and everyone who has the motivation to try ends up succeeding. Some take longer, sure, but that's also personality/preference on how to set up and how much time on uwhich stages , etc. (ran into someone on Sunday who was a great example of this - not on Fbk, got on Fbk, got encouragement, was up and running within a week of getting the gear).

Now: are there things that should be done to making hybrid closed looping accessible to all? Yes. But the push button get pancreas is going to be a regulated commercial solution, not necessarily a DIY one. DIY won't be right for everyone. And has its own limitations. For those who don't want to wait for such a push button solution, I don't want to discourage anyone based on past technical experience or lack thereof, because we have tons of evidence that people can be successful regardless of tech background.

@rofid12_twitter guessing you're who emailed me ? Saving an email, rerunning was the suggestion, glad it's working now!
Daniel
@rofid12_twitter
Oct 24 2018 20:49
@danamlewis Yeah it was me, before I rerun it. Thanks!
Dana Lewis
@danamlewis
Oct 24 2018 20:50
:+1:
HuffCam
@HuffCam
Oct 24 2018 20:59
Daft question, but relevant to the previous discussion I think: do all implementations of this thing require people to use linux?
because linux, in my experience, is the most horrendously user-hostile thing in modern computing
Martin Haeberli
@mhaeberli
Oct 24 2018 21:00
@HuffCam arguably AndroidAPS (based on OpenAPS) may not … ; nor does Loop - (which is similar in concept, but NOT the same thing)...
Eric
@ecc1
Oct 24 2018 21:01
@danamlewis Well said, and a nice illustration of why you're such a good evangelist for this community. Better you than me, clearly :-)
Martin Haeberli
@mhaeberli
Oct 24 2018 21:01
:+1:
HuffCam
@HuffCam
Oct 24 2018 21:02
I've got some tech background and I was asked to help someone set up openAPS, but I felt I had to refuse for what I suspect are obvious reasons.
But it did look fairly techy, and a big pile of circuit boards.
Daniel
@rofid12_twitter
Oct 24 2018 21:02
Thank god, my loop is working!
Martin Haeberli
@mhaeberli
Oct 24 2018 21:02
@rofid12_twitter :+1:
Daniel
@rofid12_twitter
Oct 24 2018 21:02
That was harder than I imagined it to be, although I got quite some errors
Thought it was going to be a walk in the park lol
Dana Lewis
@danamlewis
Oct 24 2018 21:04
@ecc1 it takes all hands on deck (and many perspectives) to make this community what it is :) been raving about you constantly for taking on radio support!
@rofid12_twitter :+1: sometimes it's seamless, sometimes it's not, especially depends on whether you're setting up dev or master, etc.
Daniel
@rofid12_twitter
Oct 24 2018 21:05
I can imagine, thanks for creating such a wonderful product!
Dana Lewis
@danamlewis
Oct 24 2018 21:08
btw related to the convo - for anyone not on Twitter, here's a link to a recently published article with some history and (my) perspective on DIY closed looping, with some of the recent global studies on OpenAPS and AndroidAPS referenced: https://twitter.com/danamlewis/status/1054819106048827392
screenshots cover the full article, but if you can't read those well, feel free to email me for the PDF
HuffCam
@HuffCam
Oct 24 2018 21:08
I guess the question is "does it work" and with type 1, we'll get the answer to that in a few decades.
n0rbertg
@n0rbertg
Oct 24 2018 21:09
I think a great achievement from the DYI community was teaching everyone how to use git :D (all the projects)
Dana Lewis
@danamlewis
Oct 24 2018 21:10
@HuffCam some data here ;) to start to answer that question https://openaps.org/outcomes/
HuffCam
@HuffCam
Oct 24 2018 21:10
Heh true. I dislike git so much that I tend to use the "folder full of text files" approach to source control, but then again I'm not usually collaborating.
@danamlewis Promising. I guess my other question would be how much of the grind it takes out of it.
Dana Lewis
@danamlewis
Oct 24 2018 21:11
@HuffCam :D see some of Sulka & Jason's data in the article i posted to help answer that, too
HuffCam
@HuffCam
Oct 24 2018 21:11
If it made no difference at all to long term outcomes, but made life noticeably easier, it'd be well worthwhiel.
Dana Lewis
@danamlewis
Oct 24 2018 21:12
yup, exactly. and it does for most people who try it.
HuffCam
@HuffCam
Oct 24 2018 21:13
Are these people who were well-controlled to begin with, though? I wonder if you're sort of selecting your test group quite carefully, since it'll be smart, motivated people who could deal with linux.
Dana Lewis
@danamlewis
Oct 24 2018 21:14
no, not all. a) no selecting of test group; people self-select in to DIY closed looping (and further self-selection of who donates their data). b) average does tend to be lower than tpical population, but we have plenty of people who have high starting a1c's with DIY closed looping c) that's why I encourage everyone to donate their data :) so we can do larger population studies
@applehat is one of the most vocally public about his starting A1C (he usually shares that on Twitter)
HuffCam
@HuffCam
Oct 24 2018 21:16
I don't mean to imply you're deliberately selecting, it's just that the people you get will probably be better-motivated and more intelligent to begin with.
I suspect to get a broader base it'd have to be a lot more plug-n-play.
Dana Lewis
@danamlewis
Oct 24 2018 21:16
definitely better motivated than the absolute average T1 population
but: sadly, motivation in T1 doesn't always correlate with results (until it's motivation to try closed looping), so there is quite a spread of starting A1c still
HuffCam
@HuffCam
Oct 24 2018 21:18
It's hard not to relate this to the only example I've personally known who had this disease, but the problem wasn't so much managing it, as the life impact of managing it.
Martin Haeberli
@mhaeberli
Oct 24 2018 21:19
a demanding wish of this already super-generous community - look at our nightscout portal and coach us on tuning stuff more. (we’re about to try to move forward with current autotune advice but maybe we need more…) DM me and I can DM link back if anyone volunteers ...
HuffCam
@HuffCam
Oct 24 2018 21:27
I'm not sure how much of the complexity I'm seeing is part of openAPS, and how much is just normal insulin pump stuff.
I think if I was reading the website right now, I'd conclude that a new user was in trouble because the first thing it says about hardware is that it works with "older medtronic pumps" and that the best controlled for it is an Intel Edison. Are either of those things actually available?
Why not
@Whynot2018_gitlab
Oct 24 2018 22:22

hi, I'm trying to install OpenAPS system on my RP3B+ as a rig with Dexcom G5 and CareLink Usbstick.

At the end of the installation procedure. I'm having this error /root/src/oref0/bin/oref0-setup.sh: line 1124: 5813 Segmentation fault oref0-log-shortcuts Add NIGHTSCOUT_HOST and API_SECRET to /root/.bash_profile sed: can't read .profile: No such file or directory sed: can't read .profile: No such file or directory

~/myopenaps# cat oref0-runagain.sh
To run again with these same options, use:

oref0-setup --dir=/root/myopenaps --serial=945761 --cgm=G5-upload --ns-host=https://XXXXXXX.herokuapp.com --api-secret=XXXXX --tty=/dev/ttyUSB.Carelink0 --max_iob=7 --enable=' autosens autotune meal ' --radio_locale='WW' --ww_ti_usb_reset='yes' --btmac='34:8A:7B:75:xx:xx'
# openaps use pump model
"754"root

# openaps use cgm iter_glucose 1 [ { "trend_arrow": "FLAT", "system_time": "2018-10-25T03:01:57", "display_time": "2018-10-24T21:01:32", "glucose": 230 } ]
With those configuration I'm not able to upload any date to NS.Rightnow Share Bridge I don't want to use it.I need to upload BG to NS and locally use it for the Openaps..
any help would be greatly appreciated.

Why not
@Whynot2018_gitlab
Oct 24 2018 22:30
2 more questions. Is there any # bash ~/myopenaps/oref0-runagain.sh debug mode? and for G5 , what should we add to DEXCOM_CGM_RECV_IDreceiver ID = Serial Number ?
Why not
@Whynot2018_gitlab
Oct 24 2018 22:36
and this error
# tail -F /var/log/openaps/pump-loop.log
oref0-pump-loop failed. pump_loop_completed more than 15m old; waiting for 40s silence before mmtuning
mmeowlink.vendors.serial_interface.AlreadyInUseException: /dev/ttyUSB.Carelink0 already in use

Radio check failed. mmeowlink.vendors.serial_interface.AlreadyInUseException: /dev/ttyUSB.Carelink0 already in use

The CC111x is located at /dev/ttyUSB.Carelink0
2018-10-25 00:21:13,008 ERROR TimeoutExpired. Killing process
Listening for 40s silence before mmtuning: ..............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Starting oref0-pump-loop at Thu Oct 25 00:29:01 CEST 2018 with 16 second wait_for_silence:
Waiting up to 4 minutes for new BG: ls: cannot access '/tmp/pump_loop_completed': No such file or directory

mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?

Radio check failed. mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?

The CC111x is located at /dev/ttyUSB.Carelink0
2018-10-25 00:33:05,734 ERROR TimeoutExpired. Killing process
Listening for 40s silence before mmtuning: ........................................................retry 0  (Do you need to run 'export RFSPY_RTSCTS=0' first?)
retry 0  (Do you need to run 'export RFSPY_RTSCTS=0' first?)
Why not
@Whynot2018_gitlab
Oct 24 2018 23:01
# tail -F /var/log/openaps/ns-loop.log 

Starting oref0-ns-loop at Thu Oct 25 00:53:01 CEST 2018:
Found recent valid BG:
{"glucose":195,"dateString":"2018-10-25T00:51:29+02:00"}
Latest NS temptargets: Merging local temptargets: Temptargets merged: Invalid profile.json.new after refresh
Why not
@Whynot2018_gitlab
Oct 24 2018 23:10
~/myopenaps# rm -rf settings/*  
~/myopenaps# openaps report add settings/profile.json text get-profile shell settings/settings.json settings/bg_targets.json settings/insulin_sensitivities.json settings/basal_profile.json max_iob.json
usage: openaps-report add report {base,text,stdout,JSON} get-profile shell
       [-h] [--not-json-default]
       settings bg_targets isf basal_profile preferences carb_ratios ...
openaps-report add report {base,text,stdout,JSON} get-profile shell: error: too few arguments
Why not
@Whynot2018_gitlab
Oct 24 2018 23:21
how do we pull all the information(Basal Bolus etc.) from the Pump.
Why not
@Whynot2018_gitlab
Oct 24 2018 23:35
oo damn it won't boot anymore
KS615
@KS615
Oct 24 2018 23:47

hi all, I'm having an error in my wifi settings that I can't seem to delete. Can anyone help me figure out how to fix this? I've already tried "recover" several times and that's not doing it. E325: ATTENTION
Found a swap file by the name "/etc/wpa_supplicant/.wpa_supplicant.conf.swp"
owned by: root dated: Tue Oct 23 20:35:24 2018
file name: /etc/wpa_supplicant/wpa_supplicant.conf
modified: YES
user name: root host name: edison1
process ID: 3814
While opening file "/etc/wpa_supplicant/wpa_supplicant.conf"
dated: Wed Oct 24 19:44:00 2018
NEWER than swap file!

(1) Another program may be editing the same file. If this is the case,
be careful not to end up with two different instances of the same
file when making changes. Quit, or continue with caution.
(2) An edit session for this file crashed.
If this is the case, use ":recover" or "vim -r /etc/wpa_supplicant/wpa_suppl
icant.conf"
to recover the changes (see ":help recovery").
If you did this already, delete the swap file "/etc/wpa_supplicant/.wpa_supp
licant.conf.swp"
to avoid this message.

Swap file "/etc/wpa_supplicant/.wpa_supplicant.conf.swp" already exists!
-- More --