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

5th
Jan 2016
Oskar Pearson
@oskarpearson
Jan 05 2016 00:38
Hey all. I'm looking at purchasing a bunch of ERF modules from https://www.wirelessthings.net/erf-0-1-pin-spaced-radio-module and writing the alternative firmware to them
They are Intel Edison compatible, work with openaps / mmeowlink, and have a range as good as the TI stick. They are also much smaller, use a lot less power, and are much cheaper
If anyone in Europe / the USA wants one, please drop me a message at oskar@deckle.co.uk (please include 'ERF' in the subject) with your address and whether your pump is a USA pump or a EU pump.
I'm attempting to get them to sell me the 915mhz versions in the UK - I think they are likely to do it if I get a reasonable size order
File 29-12-2015, 19 59 21.jpeg
File 29-12-2015, 20 00 34.jpeg
These would replace the carelink stick - it's just for relative size comparisons
When combined with an intel Edison you can get the openaps rig that size (with 13 hour battery life)
Oskar Pearson
@oskarpearson
Jan 05 2016 00:43
I'll sell them at cost + shipping fees
For the US, I'd like to post them to one person somewhere in the US that can then hand them out in person or post them locally
You can use them on an Intel Edison or a Raspberry pi
Note that you will need some soldering skills to deal with this on the Edison... it doesn't come with the header soldered in
If you're really sure you want to stick with the Raspberry pi and can't solder, and there are a few of you, I might do the same for https://www.wirelessthings.net/slice-of-radio-wireless-rf-transciever-for-the-raspberry-pi
Jason Calabrese
@jasoncalabrese
Jan 05 2016 00:51
awesome @oskarpearson
Oskar Pearson
@oskarpearson
Jan 05 2016 00:52
If you're a bit apprehensive about making a purchase now, I'll do a repeat order once a few people have them and can share their experiences etc.
I'll also be updating the docs in the next few days, so that if someone wants to order direct they can certainly do so
I'm just going to be doing this for at least one person, so figured I may as well get a batch and save other people hassle
Range seems better than the TI stick to me
Gustavo
@bustavo
Jan 05 2016 00:56
Sending email!
Oskar Pearson
@oskarpearson
Jan 05 2016 00:58
:)
Dana Lewis
@danamlewis
Jan 05 2016 00:59
Wondering if any of the Seattle Openaps crew have access to soldering gear ? @logichammer @eyim @esteward
Oskar Pearson
@oskarpearson
Jan 05 2016 01:00
Also in the email - vote for whether you'd like them pre-soldered or not
If I do solder them, I'll solder them in low-profile - and would have to add some of these: https://shop.pimoroni.com/products/jumper-jerky
I think for the ultimately smallest size, you're going to need to have someone that solders something
Those would do for connecting them to a pi at least for testing
Getting them connected to the Edison might be a bit more challenging and require soldering
I hoped that https://www.sparkfun.com/products/13040 would work, but didn't have much luck. It seems like it should be perfect, but actually 5V and it doesn't work
eyim
@eyim
Jan 05 2016 01:04
I don't have any soldering gear. I have a friend who did the xdrip hardware for me. If the instructions are clear, I might be able to get him to do them for us as well. Oskar - I'll send you an email to order a bunch for both the edison and the pi
What is the range you are getting for it? Same as carelink or better?
Rachel Sandlain
@audiefile
Jan 05 2016 01:06
you can get soldering stuff for a good price from radio shack
Oskar Pearson
@oskarpearson
Jan 05 2016 01:06
It's pretty straight forward soldering - though some of the connectors are smallish. I'll be taking photos and putting them online. Currently I'm connecting the above to https://www.sparkfun.com/products/13025 which requires soldering onto the base board.
Rachel Sandlain
@audiefile
Jan 05 2016 01:06
I think the soldering iron I have was maybe 20$?
Oskar Pearson
@oskarpearson
Jan 05 2016 01:08
NB to note that range is going to vary a bit between the US version and the EU version... so I can't really guarantee anything. Our experience is that while it's still only a few meters, we find it much much better than the Carelink. But it's possible that our Carelink is exceptionally bad.
Some others in the US and EU have said the TI stick is around the same as the Carelink for them
However, we have only one device at the bed, and we don't have drop-outs for hours like we had with the Carelink
eyim
@eyim
Jan 05 2016 01:09
Nice !
Oskar Pearson
@oskarpearson
Jan 05 2016 01:09
The main advantage is not neccessarily huge range - it's that the device can get a lot smaller
eyim
@eyim
Jan 05 2016 01:10
Great. My son will like that. He thinks the pi + Battery + carelink is too bulky to carry around so we don't get much day usage.
Oskar Pearson
@oskarpearson
Jan 05 2016 01:11
aemazaheri was going to chat to me sometime about getting the one he's already bought in the USA to work there
We can perhaps give you some anecdotal evidence once he's tested
Siobhan carried around the whole pi/battery/carelink thing for a bit, but it's only once changing to this system that we've really had success with things during the day.
The old version had to live in her bag where it never had any reception
She carries it on her now
Dana Lewis
@danamlewis
Jan 05 2016 01:18
@audiefile yup that looks great! want to PR that into the Hardware section, or if not I can.
@oskarpearson glad that’s working out better now!
Oskar Pearson
@oskarpearson
Jan 05 2016 01:21
Not as glad as me! :)
#lifechanging
Dana Lewis
@danamlewis
Jan 05 2016 01:21
:)
Oskar Pearson
@oskarpearson
Jan 05 2016 01:21
#gratitude
Also - if anyone's got an experience building 3d printed boxes or laser-cut boxes like https://shop.pimoroni.com/collections/raspberry-pi/products/raspberry-pi-2-pibow and wouldn't mind volunteering some time, it'd be great to try and coordinate
But, unfortunately, now is not quite the time
I'm off to bed
Andrew Warrington
@andrew-warrington
Jan 05 2016 01:35
This message was deleted
@oskarpearson, for when you wake up, I would like to volunteer. I have designed and 3D printed a box (going through various prototypes) for my own setup. Would love to work with you on these phenomenal miniature things you're making.
Dana Lewis
@danamlewis
Jan 05 2016 01:47
woohoo! just merged @mddub 's excellent openaps/docs#42
So we're all clear on existing PRs to the openaps docs. If anyone had some PRs in mind/in the works, make sure you've reforked master to make sure you've got the latest, we did some structural work as well recently.
But if in doubt, submit what you've got, and we can help fit it in.
Scott Leibrand
@scottleibrand
Jan 05 2016 01:49
You merged that to dev, not yet to master.
Dana Lewis
@danamlewis
Jan 05 2016 01:49
Ah.
Ok. Swap "master" for dev up there then. ^^
Scott Leibrand
@scottleibrand
Jan 05 2016 01:49
:-)
Dana Lewis
@danamlewis
Jan 05 2016 01:49
Fork dev :) for the next PR
Rachel Sandlain
@audiefile
Jan 05 2016 01:51
PR send for xdrip info. Hopefully @ochenmiller can add/check the reference to his repo
Scott Leibrand
@scottleibrand
Jan 05 2016 01:52
terminology-wise, a fork actually gets you the whole repo, including all branches, master, dev, etc. So if you delete and re-fork, you'll just want to make sure you base your changes on the dev branch. Or, if you want to be non-destructive, you can merge the dev branch from openaps/docs into your fork's dev branch to get it up to date.
@audiefile's change targets master, FYI. Not sure if that'll cause a merge conflict with dev
Dana Lewis
@danamlewis
Jan 05 2016 01:54
I was just looking into that. Reviewing now.
Scott Leibrand
@scottleibrand
Jan 05 2016 01:55
I made a second PR to dev. It's clean: openaps/docs#71
Rachel Sandlain
@audiefile
Jan 05 2016 01:58
I thought I targeted dev, sorry
Dana Lewis
@danamlewis
Jan 05 2016 01:59
no worries
Dana Lewis
@danamlewis
Jan 05 2016 02:26
cool. just did a pr from dev to master pulling in @audiefile and @mddub work, plus a few other lifestyle related notes on hardware, if anyone wants to take a look tonight (openaps/docs#73)
Dana Lewis
@danamlewis
Jan 05 2016 02:33
wondering @eyim , is this still in need of a solution? openaps/docs#48
eyim
@eyim
Jan 05 2016 02:34

Not sure how to do a PR on this one but on this page: docs/getting-started/hardware.md It states:
Your OpenAPS implementation can also pull CGM data from a Nightscout site in addition to pulling from the CGM directly.

+* More documentation is coming soon about pulling CGM data from Nightscout. - I would change this part to: You can find more documentation on this at: https://openaps.gitbooks.io/building-an-open-artificial-pancreas-system/content/docs/Log-clean-analyze-with-openaps-tools/using.html

Dana Lewis
@danamlewis
Jan 05 2016 02:35
AHhhh
ok, I was trying to figure out how I had pasted that link by accident :)
makes sense now. will add that to my #73. Thanks!
eyim
@eyim
Jan 05 2016 02:37
@danamlewis Not sure about openaps/docs#48 as I have a clone of my SD card and install from that these days. Hopefully I'll have time tonight and I'll wipe one of my SD cards and start from scratch. I'll let you know if it fails.
Dana Lewis
@danamlewis
Jan 05 2016 02:42
Ok, openaps/docs#73 now includes the Nightscout link and I'll close that issue, too
Easy when it's already done :) good find @eyim
For future reference: what I did was go back to the dev branch and make that edit and committed it to dev, and then checked that it was automatically pulled into the open PR.
Jason Calabrese
@jasoncalabrese
Jan 05 2016 02:54
docs are looking great
Dana Lewis
@danamlewis
Jan 05 2016 03:00
:+1:
Tony Zarro
@tzarro
Jan 05 2016 03:56
Having a weird problem, hoping someone can help...
about a half hour ago, my loop runs have been getting stuck...
I've got a killall run setup in my crontab as well, and its running...
but my loops aren't completing
when I attempt a manual loop, it gets stuck at the first task (monitor/glucose.json) and never moves forward
I've rebooted a couple of times, to no avail...any ideas where to start troubleshooting?
Scott Leibrand
@scottleibrand
Jan 05 2016 04:00
Run the steps manually
Tony Zarro
@tzarro
Jan 05 2016 04:00
ok
Scott Leibrand
@scottleibrand
Jan 05 2016 04:00
run each alias, figure out which one is getting stuck, then run the subsidiary reports
or jump straight to the report you think is failing.
Tony Zarro
@tzarro
Jan 05 2016 04:10
weird...I shut down and left it off for a bit longer this time, and it started working again...back to normal...
Dana Lewis
@danamlewis
Jan 05 2016 04:11
Sometimes rebooting is magical.
Sometimes it's the fifth reboot.
:)
Tony Zarro
@tzarro
Jan 05 2016 04:12
I think it's a sign...I'm now going to reboot my brain...'night all!
Dana Lewis
@danamlewis
Jan 05 2016 04:13
:+1: night!
Pete Schwamb
@ps2
Jan 05 2016 05:28
@oskarpearson As for range/reliability, you might be interested in the latest subg_rfspy settings. I used this script: https://github.com/ps2/minimed_rf/blob/master/bin/mmtune, and discovered some changes that improve things significantly. Here are the changes: ps2/subg_rfspy@688fe30
eyim
@eyim
Jan 05 2016 05:30

@danamlewis Hi Dana - looks like installing from source still does not work. You still have to run the install from package manager first curl -s https://raw.githubusercontent.com/openaps/docs/master/scripts/quick-packages.sh | bash -

Not sure if that is the expected behavior. I thought I used to be able to run the installing from source by itself

openaps/docs#48
Scott Leibrand
@scottleibrand
Jan 05 2016 05:32
It should not be necessary. Can you try running the source install repeatedly, running a single command out of the package install script in between each run, until the source install succeeds? Need to narrow down what the dependency is.
eyim
@eyim
Jan 05 2016 05:39
yep will try now
eyim
@eyim
Jan 05 2016 05:52
Failed at first command sudo apt-get update. Ran it a second time and it completed. Let me know if you prefer a private message
pi@raspberrypi ~ $ sudo apt-get update
Get:1 http://archive.raspberrypi.org jessie InRelease [13.3 kB]
Get:2 http://mirrordirector.raspbian.org jessie InRelease [15.0 kB]
Get:3 http://archive.raspberrypi.org jessie/main Sources [32.5 kB]
Get:4 http://mirrordirector.raspbian.org jessie/main armhf Packages [8,963 kB]
Get:5 http://archive.raspberrypi.org jessie/ui Sources [5,197 B]
Get:6 http://archive.raspberrypi.org jessie/main armhf Packages [101 kB]
Get:7 http://archive.raspberrypi.org jessie/ui armhf Packages [7,639 B]
Ign http://archive.raspberrypi.org jessie/main Translation-en_GB
Ign http://archive.raspberrypi.org jessie/main Translation-en
Ign http://archive.raspberrypi.org jessie/ui Translation-en_GB
Ign http://archive.raspberrypi.org jessie/ui Translation-en
Get:8 http://mirrordirector.raspbian.org jessie/contrib armhf Packages [37.5 kB]
Get:9 http://mirrordirector.raspbian.org jessie/non-free armhf Packages [70.2 kB]
Get:10 http://mirrordirector.raspbian.org jessie/rpi armhf Packages [1,356 B]
Err http://mirrordirector.raspbian.org jessie/contrib Translation-en_GB
  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]
Err http://mirrordirector.raspbian.org jessie/contrib Translation-en
  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]
Err http://mirrordirector.raspbian.org jessie/main Translation-en_GB
  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]
Err http://mirrordirector.raspbian.org jessie/main Translation-en
  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]
Err http://mirrordirector.raspbian.org jessie/non-free Translation-en_GB
  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]
Err http://mirrordirector.raspbian.org jessie/non-free Translation-en
  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]
Err http://mirrordirector.raspbian.org jessie/rpi Translation-en_GB
  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]
Err http://mirrordirector.raspbian.org jessie/rpi Translation-en
  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]
Fetched 9,246 kB in 35s (261 kB/s)
W: Failed to fetch http://mirrordirector.raspbian.org/raspbian/dists/jessie/contrib/i18n/Translation-en_GB  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]

W: Failed to fetch http://mirrordirector.raspbian.org/raspbian/dists/jessie/contrib/i18n/Translation-en  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]

W: Failed to fetch http://mirrordirector.raspbian.org/raspbian/dists/jessie/main/i18n/Translation-en_GB  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]

W: Failed to fetch http://mirrordirector.raspbian.org/raspbian/dists/jessie/main/i18n/Translation-en  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]

W: Failed to fetch http://mirrordirector.raspbian.org/raspbian/dists/jessie/non-free/i18n/Translation-en_GB  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]

W: Failed to fetch http://mirrordirector.raspbian.org/raspbian/dists/jessie/non-free/i18n/Translation-en  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]

W: Failed to fetch http://mirrordirector.raspbian.org/raspbian/dists/jessie/rpi/i18n/Translation-en_GB  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]

W: Failed to fetch http://mirrordirector.raspbian.org/raspbian/dists/jessie/rpi/i18n/Translation-en  Unable to connect to mirrordirector.raspbian.org:http: [IP: 2001:41c9:1:3ce::11 80]

E: Some index files failed to download. They have been ignored, or old ones used instead.
testing the rest now
Previously have run the script multiple times and it failed at the same place. This is the first time it worked when I ran it again
Scott Leibrand
@scottleibrand
Jan 05 2016 06:09
If apt-get update fails, that's a problem with Rasbian's mirrordirector service. You could modify your sources.lst to specify a known-good mirror. Google for Raspbian mirrors to find out how to do that.
We might want to add something to the script to automatically add another mirror if update fails...
eyim
@eyim
Jan 05 2016 06:10
The rest of the commands ran fine
eyim
@eyim
Jan 05 2016 06:16
I am not worried about me. I've figured out how to get around this issue. But for someone new trying to install it, they may not know how to get around it if they get this error - especially if they aren't really tech savvy.
Scott Leibrand
@scottleibrand
Jan 05 2016 06:16
So was that only running source? Never had to run packages?
Yes, agreed.
eyim
@eyim
Jan 05 2016 06:17
Correct. If I run packages first then I get no error
Scott Leibrand
@scottleibrand
Jan 05 2016 06:17
That's why I want to automate a fix for whatever is intermittently failing.
did you have to run packages at all this time though?
or did it succeed rerunning src?
eyim
@eyim
Jan 05 2016 06:18
No I just ran each command separately
the first command I had to run twice before it went through
but previously I have run the script multiples times and it failed. Not idea why there is a difference
Scott Leibrand
@scottleibrand
Jan 05 2016 06:19
Were you running the individual commands from packages or from src?
eyim
@eyim
Jan 05 2016 06:19
src
I've never had an issue with packages
Scott Leibrand
@scottleibrand
Jan 05 2016 06:19
lets look at the differences between them
eyim
@eyim
Jan 05 2016 06:20
Actually - let me redo this again.
one big difference is that quick-src doesn't do the update and upgrade
eyim
@eyim
Jan 05 2016 06:22
I think I just had a brain fart and did packages rather than src
Scott Leibrand
@scottleibrand
Jan 05 2016 06:22
if there is a dependency that needs to be upgraded, then that might account for it.
See if you can reproduce the failure with src, and then whether just running update and upgrade fixes it.
eyim
@eyim
Jan 05 2016 06:23
k
Scott Leibrand
@scottleibrand
Jan 05 2016 06:23
If so, you can add that line to quick-src and PR the fix.
I'm off to bed, but there are plenty of people here who can help with this kind of troubleshooting.
eyim
@eyim
Jan 05 2016 06:24
Good night
eyim
@eyim
Jan 05 2016 07:56

Okay sorry Scott - sort of a brain fart from earlier. I was testing package before and got the error I mentioned above that corrected itself after running it a couple of times.

In terms of src, the failure is with npm that is found in the first line. Everything installs correctly until you get to that part. So when you run sudo apt-get install -y npm, you get this error

pi@raspberrypi ~ $ sudo apt-get install -y npm
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following extra packages will be installed:
  gyp libc-ares-dev libjs-node-uuid libjs-underscore libssl-dev libssl-doc libv8-3.14-dev node-abbrev node-ansi node-ansi-color-table node-archy node-async
  node-block-stream node-combined-stream node-cookie-jar node-delayed-stream node-forever-agent node-form-data node-fstream node-fstream-ignore
  node-github-url-from-git node-glob node-graceful-fs node-gyp node-inherits node-ini node-json-stringify-safe node-lockfile node-lru-cache node-mime
  node-minimatch node-mkdirp node-mute-stream node-node-uuid node-nopt node-normalize-package-data node-npmlog node-once node-osenv node-qs node-read
  node-read-package-json node-request node-retry node-rimraf node-semver node-sha node-sigmund node-slide node-tar node-tunnel-agent node-underscore
  node-which nodejs-dev
Suggested packages:
  node-hawk node-aws-sign node-oauth-sign node-http-signature
The following NEW packages will be installed:
  gyp libc-ares-dev libjs-node-uuid libjs-underscore libssl-dev libssl-doc libv8-3.14-dev node-abbrev node-ansi node-ansi-color-table node-archy node-async
  node-block-stream node-combined-stream node-cookie-jar node-delayed-stream node-forever-agent node-form-data node-fstream node-fstream-ignore
  node-github-url-from-git node-glob node-graceful-fs node-gyp node-inherits node-ini node-json-stringify-safe node-lockfile node-lru-cache node-mime
  node-minimatch node-mkdirp node-mute-stream node-node-uuid node-nopt node-normalize-package-data node-npmlog node-once node-osenv node-qs node-read
  node-read-package-json node-request node-retry node-rimraf node-semver node-sha node-sigmund node-slide node-tar node-tunnel-agent node-underscore
  node-which nodejs-dev npm
0 upgraded, 55 newly installed, 0 to remove and 0 not upgraded.
Need to get 2,223 kB/3,710 kB of archives.
After this operation, 13.3 MB of additional disk space will be used.
Err http://mirrordirector.raspbian.org/raspbian/ jessie/main libssl-dev armhf 1.0.1k-3+deb8u1
  404  Not Found [IP: 5.153.225.207 80]
Err http://mirrordirector.raspbian.org/raspbian/ jessie/main libssl-doc all 1.0.1k-3+deb8u1
  404  Not Found [IP: 5.153.225.207 80]
E: Failed to fetch http://mirrordirector.raspbian.org/raspbian/pool/main/o/openssl/libssl-dev_1.0.1k-3+deb8u1_armhf.deb  404  Not Found [IP: 5.153.225.207 80]

E: Failed to fetch http://mirrordirector.raspbian.org/raspbian/pool/main/o/openssl/libssl-doc_1.0.1k-3+deb8u1_all.deb  404  Not Found [IP: 5.153.225.207 80]

E: Unable to fetch some archives, maybe run apt-get update or try with --fix-missing?

that doesn't clear no matter how many times I run it. Even running the whole line still fails

sudo apt-get install -y git python python-dev python-software-properties python-numpy python-pip nodejs-legacy npm watchdog

which then causes this command to fail

sudo npm install -g json
As I mentioned before running package first then src removes this error which is how I get around it
Ben West
@bewest
Jan 05 2016 08:00
looks like the mirror is flaky
Oskar Pearson
@oskarpearson
Jan 05 2016 10:03
That sometimes happens if you've not recently run an 'apt-get update' before the apt-get install
It's probably that libssl 1.0.1k-3+deb8u1 has security issues, so they have upgraded it and replaced it
If you 'apt-get update' first, it'll probably pull down a later version of libssl
Scott Leibrand
@scottleibrand
Jan 05 2016 12:01
@eyim can you try that? Run just sudo apt-get update and retry quick-src?
Tony Zarro
@tzarro
Jan 05 2016 14:52
Good morning all!
Had a great second night on OpenAPS...I did have one hypo, but not severe...I had good connectivity from OpenAPS to the pump most of the night, but I did have a recurrence of an issue I have had several times before...one that only a reboot seems to cure. When I woke up with low blood sugar (and after treating), I immediately checked to see what happened and realized I needed to reboot...once I did that all went back to normal. I am planning to add a periodic reboot to my crontab just in case I keep getting the issue and can't find any other way to fix it.
Tony Zarro
@tzarro
Jan 05 2016 14:59
I can't find the issue with "cat loop.log", perhaps its gone now, but it was something like "ACK=0", from some of the other text (which I can't remember) it seemed like an issue with connectivity
Jason Calabrese
@jasoncalabrese
Jan 05 2016 15:01
are you running Jessie on your pi? I was seeing stuff like that a few month ago before getting my loop going
Tony Zarro
@tzarro
Jan 05 2016 15:02
Yes, Jessie
Scott Leibrand
@scottleibrand
Jan 05 2016 15:30
Do you have oref0-reset-usb in your loop/cron?
openaps alias add preflight '! bash -c "rm -f monitor/clock.json && echo -n \"PREFLIGHT \" && openaps report invoke monitor/clock.json 2>/dev/null >/dev/null && grep -q T monitor/clock.json && echo OK || ( mm-stick warmup || sudo oref0-reset-usb; echo FAIL; sleep 120; exit 1 )"'
that should fix it would a reboot.
Tony Zarro
@tzarro
Jan 05 2016 15:34
Yes, I had that in my crontab
Scott Leibrand
@scottleibrand
Jan 05 2016 15:37
Hmm, strange. I wonder if it's not working with your hardware. Which Pi?
Tony Zarro
@tzarro
Jan 05 2016 15:37
Pi 2
Scott Leibrand
@scottleibrand
Jan 05 2016 15:38
Same. Can you run it manually and see what it does? Then run it next time you see the frozen loop and see what it does then.
Tony Zarro
@tzarro
Jan 05 2016 15:40
run oref0-reset-usb? Yep
the last bit of text in my preflight alias is "|| sudo 0-reset-usb; echo FAIL; openaps get-bg; sleep 120; exit 1 )"
the inclusion of "openaps get-bg" is different from yours.
going to run oref0-reset-usb now
Scott Leibrand
@scottleibrand
Jan 05 2016 15:46
sudo 0-reset-usb? Is the typo in the alias or just here?
get-bg is a recent addition you don't need.
Tony Zarro
@tzarro
Jan 05 2016 15:47
ok, I'll remove it
sorry, a typo here, not in my crontab
Scott Leibrand
@scottleibrand
Jan 05 2016 15:47
Doesn't hurt. It's for my system of rsyncing BG between Pi's
Tony Zarro
@tzarro
Jan 05 2016 15:47
how long does the power cycling usually take?
ah, that's right, it kills wifi as well
Scott Leibrand
@scottleibrand
Jan 05 2016 15:48
it's done in seconds, but you might lose wifi. Ya.
Tony Zarro
@tzarro
Jan 05 2016 15:48
ok, it seemed to work fine
Scott Leibrand
@scottleibrand
Jan 05 2016 15:48
K. Strange that isn't sufficient. Definitely run that the next time you notice it lock up.
Tony Zarro
@tzarro
Jan 05 2016 15:50
ok
eyim
@eyim
Jan 05 2016 17:15
@oskarpearson @scottleibrand Success - Yes running sudo apt-get update first before the src script removing the errors. I can put in a PR that simply says to run that command first before running the script or someone else can fix the script itself. Let me know what you would like to do. Thank you Oskar for finding the solution.
Dana Lewis
@danamlewis
Jan 05 2016 17:20
Go ahead and PR that so its fixed for the next person? Thanks!
Jason Calabrese
@jasoncalabrese
Jan 05 2016 17:20
nice, glad you found and verified that
Dana Lewis
@danamlewis
Jan 05 2016 17:20
Actually, pr that into the script itself
eyim
@eyim
Jan 05 2016 17:27
@danamlewis Okay I'll try that. I didn't realize the scripts were in there but just found them.
Dana Lewis
@danamlewis
Jan 05 2016 17:32
:)
Scott Leibrand
@scottleibrand
Jan 05 2016 17:37
You can also download the script, modify it locally, and test.
Or locally clone your fork of the docs repo, test locally, and then git push it back up to your repo to PR
eyim
@eyim
Jan 05 2016 17:45
PR sent - targeting dev branch which is how I think you guys want it. Also added a bit about what to look for to know that the install was successful. And also to retry packages multiple times if it fails like it did for me last night. Feel free to exclude the extra parts if you don't like it
Dana Lewis
@danamlewis
Jan 05 2016 17:46
Cool. Thanks for targeting dev! Just merged.
Gustavo
@bustavo
Jan 05 2016 18:18
Hi all, I’ve been trying to implement —auto-init, with no consistent success on decocare ( sometimes it works, sometimes it doesn’t)… is there something special I must consider?
Ben West
@bewest
Jan 05 2016 18:29
hmm, autoinit was an experiment... I don't think it works very well
Scott Leibrand
@scottleibrand
Jan 05 2016 18:30
Openaps keeps track of last init expire time to decide when to init again.
Gustavo
@bustavo
Jan 05 2016 18:30
ok, sticking to —init
@scottleibrand how long does the session last for?
Ben West
@bewest
Jan 05 2016 18:31
10 minutes, defaul, in decocare, you can set --rf-minutes, in openaps you can also set it
Gustavo
@bustavo
Jan 05 2016 18:32
so if the session lasts 10 minutes, and then the loop runs every 5 min, does that mean that by the third iteration I need to use —init again? or does the session time reset with each connection?
Ben West
@bewest
Jan 05 2016 18:32
best to change it
I use minutes = 3
Gustavo
@bustavo
Jan 05 2016 18:33
ah, that sounds good @bewest !
Ben West
@bewest
Jan 05 2016 18:33
yes, you can get into a phase on 3rd/second time where it doesn't work as well
Gustavo
@bustavo
Jan 05 2016 18:35
does changing it to 3 min increase battery life by a lot?
Paul Andrel
@stavlor
Jan 05 2016 19:05
hmm
[pandrel@paul-desktop ~]$ openaps use pump model
Traceback (most recent call last):
  File "/usr/bin/openaps-use", line 63, in <module>
    app( )
  File "/usr/lib/python2.7/site-packages/openaps/cli/__init__.py", line 44, in __call__
    self.run(self.args)
  File "/usr/bin/openaps-use", line 57, in run
    output = app(args, self)
  File "/usr/lib/python2.7/site-packages/openaps/uses/__init__.py", line 92, in __call__
    return self.method.selected(args)(args, app)
  File "/usr/lib/python2.7/site-packages/openaps/uses/__init__.py", line 31, in __call__
    return self.method(args, app)
  File "/usr/lib/python2.7/site-packages/openaps/uses/use.py", line 44, in __call__
    self.before_main(args, app)
  File "/usr/lib/python2.7/site-packages/openaps/vendors/medtronic.py", line 60, in before_main
    self.setup_medtronic( )
  File "/usr/lib/python2.7/site-packages/openaps/vendors/medtronic.py", line 137, in setup_medtronic
    self.uart.open( )
  File "/usr/lib/python2.7/site-packages/decocare/stick.py", line 869, in open
    log.info('%s' % self.product_info( ))
  File "/usr/lib/python2.7/site-packages/decocare/stick.py", line 510, in product_info
    return self.query(ProductInfo)
  File "/usr/lib/python2.7/site-packages/decocare/stick.py", line 504, in query
    return self.process( )
  File "/usr/lib/python2.7/site-packages/decocare/stick.py", line 485, in process
    raw = self.send_force_read( )
  File "/usr/lib/python2.7/site-packages/decocare/stick.py", line 601, in send_force_read
    self.link.write(reader.format( ))
  File "/usr/lib/python2.7/site-packages/decocare/link.py", line 45, in write
    r = self.serial.write( string )
  File "/usr/lib/python2.7/site-packages/serial/serialposix.py", line 517, in write
    raise portNotOpenError
serial.serialutil.SerialException: Attempting to use a port that is not open
fun
Bus 001 Device 008: ID 0a21:8001 Medtronic Physio Control Corp. MMT-7305WW [Medtronic Minimed CareLink]
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               2.00
  bDeviceClass            0 
  bDeviceSubClass         0 
  bDeviceProtocol         0 
  bMaxPacketSize0        64
  idVendor           0x0a21 Medtronic Physio Control Corp.
  idProduct          0x8001 MMT-7305WW [Medtronic Minimed CareLink]
  bcdDevice            1.10
  iManufacturer           0 
  iProduct                0 
  iSerial                 0 
  bNumConfigurations      1
  Configuration Descriptor:
    bLength                 9
    bDescriptorType         2
    wTotalLength           32
    bNumInterfaces          1
    bConfigurationValue     1
    iConfiguration          0 
    bmAttributes         0x80
      (Bus Powered)
    MaxPower              100mA
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        0
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass       255 Vendor Specific Class
      bInterfaceSubClass    255 Vendor Specific Subclass
      bInterfaceProtocol    255 Vendor Specific Protocol
      iInterface              0 
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x01  EP 1 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0040  1x 64 bytes
        bInterval               1
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x82  EP 2 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0040  1x 64 bytes
        bInterval               1
can't get device qualifier: Resource temporarily unavailable
can't get debug descriptor: Resource temporarily unavailable
Device Status:     0x0000
  (Bus Powered)
Jason Calabrese
@jasoncalabrese
Jan 05 2016 19:11
running on a linux desktop?
Paul Andrel
@stavlor
Jan 05 2016 19:12
yep
Ben West
@bewest
Jan 05 2016 19:15
check persionsmis
permissions
try sudo openaps-install-udev-rules and try again
oh interesting
never seen that before
Paul Andrel
@stavlor
Jan 05 2016 19:22
hmm wonder if i'm missing things or something
[pandrel@paul-desktop ~]$ sudo openaps-install-udev-rules 
[sudo] password for pandrel: 
Traceback (most recent call last):
  File "<stdin>", line 2, in <module>
  File "/usr/lib/python2.7/site-packages/pkg_resources/__init__.py", line 1173, in resource_string
    self, resource_name
  File "/usr/lib/python2.7/site-packages/pkg_resources/__init__.py", line 1605, in get_resource_string
    return self._get(self._fn(self.module_path, resource_name))
  File "/usr/lib/python2.7/site-packages/pkg_resources/__init__.py", line 1728, in _get
    with open(path, 'rb') as stream:
IOError: [Errno 2] No such file or directory: '/usr/lib/python2.7/site-packages/decocare/etc/80-medtronic-carelink.rules'
failed /etc/udev/rules.d/80-medtronic-carelink.rules
Traceback (most recent call last):
  File "<stdin>", line 2, in <module>
  File "/usr/lib/python2.7/site-packages/pkg_resources/__init__.py", line 1173, in resource_string
    self, resource_name
  File "/usr/lib/python2.7/site-packages/pkg_resources/__init__.py", line 1605, in get_resource_string
    return self._get(self._fn(self.module_path, resource_name))
  File "/usr/lib/python2.7/site-packages/pkg_resources/__init__.py", line 1728, in _get
    with open(path, 'rb') as stream:
IOError: [Errno 2] No such file or directory: '/usr/lib/python2.7/site-packages/dexcom_reader/etc/udev/rules.d/80-dexcom.rules'
failed /etc/udev/rules.d/80-dexcom.rules
[pandrel@paul-desktop ~]$
Toby Canning
@TC2013
Jan 05 2016 19:23
Has anyone tried the Intel Compute Stick?
Paul Andrel
@stavlor
Jan 05 2016 19:23
@TC2013 i have one sitting a few feet from me but it still has windows 10 on it
Toby Canning
@TC2013
Jan 05 2016 19:24
Do you have any free time to add linux and try out OpenAPS on it?
Paul Andrel
@stavlor
Jan 05 2016 19:28
@TC2013 its on the list of things i've been meaning to do just need to grab my usb key with ubuntu 15.10 on it
Jason Calabrese
@jasoncalabrese
Jan 05 2016 19:29
@stavlor did you find a compatible pump?
or going to start with read-only?
Paul Andrel
@stavlor
Jan 05 2016 19:34
@jasoncalabrese I have a 722 i can test with the 751 pretty much ignores anything useful other than reads
Jason Calabrese
@jasoncalabrese
Jan 05 2016 19:36
great, saw you said you had the 751 the other day, betting that 722 gets some use soon
Tony Zarro
@tzarro
Jan 05 2016 19:56
Thinking ahead: does minimed offer out-of-warranty pump repair for a fee?
Ben West
@bewest
Jan 05 2016 19:57
by default, most new devices get created with root permissions
Dana Lewis
@danamlewis
Jan 05 2016 19:58
@tzarro Maybe but they'd probably also replace the firmware. So I wouldn't send an openAPS compatible pump to them
Jason Calabrese
@jasoncalabrese
Jan 05 2016 19:58
these pumps are like classic cars
Ben West
@bewest
Jan 05 2016 19:58
hehe, they don't make them like this anymore :-)
Paul Andrel
@stavlor
Jan 05 2016 20:01
heh i know the last time i did a pump swap/change out with mm my device went from fw 1.0D to 3.0D
Ben West
@bewest
Jan 05 2016 20:07
has anyone tried the BLE share openaps driver?
Martin Haeberli
@mhaeberli
Jan 05 2016 20:07
off topic, probably, but has anyone designed a 3-d printed case for a Raspberry Pi-based OpenAPS system?
Scott Leibrand
@scottleibrand
Jan 05 2016 20:09
Thinking about switching our second receiver to be Share so we can do that on one and Dexcom's app on the other.
Probably will focus on getting Bluetooth tethering working first though.
@mhaeberli yes, most recently @andrew-warrington
Ben West
@bewest
Jan 05 2016 20:10
why use dexcom's and require internet ;-)
download adafruit/Adafruit_Python_BluefruitLE#8 from source/git and install
openaps vendor add openxshareble :-)
Scott Leibrand
@scottleibrand
Jan 05 2016 20:12
Because their app has alarms that don't depend on our loop.
John Males
@johnmales
Jan 05 2016 20:12
sorry for the offtopic question - how long do the dexcomm cgm sensors typically last? ie how often do you need to insert a new sensor?
Ben West
@bewest
Jan 05 2016 20:12
it varies
I had one that didn't survive insertion
had one that lasted 40 days
Paul Andrel
@stavlor
Jan 05 2016 20:13
reliably it can very
Scott Leibrand
@scottleibrand
Jan 05 2016 20:13
If loop dies, that's the most likely thing to wake us up.
Can't have the Pi be SPOF for our backup monitoring too.
Typically 2 weeks.
Ben West
@bewest
Jan 05 2016 20:14
I've finally started to reset some of my sensitivity to alarms
Scott Leibrand
@scottleibrand
Jan 05 2016 20:14
:-)
Ben West
@bewest
Jan 05 2016 20:14
still need more though
it got abused a bit over holidays
Scott Leibrand
@scottleibrand
Jan 05 2016 20:14
having a lot fewer lows and highs really helps
John Males
@johnmales
Jan 05 2016 20:14
Much better than the 6 days for the medtronic by the sounds of it.
Ben West
@bewest
Jan 05 2016 20:14
slept through at least 2
Martin Haeberli
@mhaeberli
Jan 05 2016 20:14
@scottleibrand - thanks for pointer to @andrew-warrington ; in the crawl - walk - run space, andrew’s design looks to depend on a mechanically hacked RPi for compactness. I was looking for something that doesn’t yet require RPi to be hacked.
Ben West
@bewest
Jan 05 2016 20:15
yeah, usually get a solid week
usually solid 2 weeks
after that it's iffy for all kinds of reasons
so depends on personal tolerance to variety of things
Milos Kozak
@MilosKozak
Jan 05 2016 20:15
we've usualy 1 week :(
Paul Andrel
@stavlor
Jan 05 2016 20:15
past 2 weeks is where i start to notice the occasional jumpyness/oddness
Jason Calabrese
@jasoncalabrese
Jan 05 2016 20:16
@bewest openxshareble doesn't pull sensor records yet right?
John Males
@johnmales
Jan 05 2016 20:16
is the sensor life determined by dexcomm? Or can you keep using it until it fails?
Ben West
@bewest
Jan 05 2016 20:16
why not?
Jason Calabrese
@jasoncalabrese
Jan 05 2016 20:17
thought you weren't doing that yet, you get the cal and sensor records?
for some reason it's commented out
Paul Andrel
@stavlor
Jan 05 2016 20:17
@johnmales for the most part the only thing really controlled by dexcom is the sensor session length you can stop/start it infinitely i suppose but accuracy tends to fall off after 2 weeks
Jason Calabrese
@jasoncalabrese
Jan 05 2016 20:18
wondering what devices to focus on, seems like edison and erf could be pretty good, but have a couple pi zeros on the way too
Ben West
@bewest
Jan 05 2016 20:19
one of the big differences between g4 and g5 is g5's have an end of life built into the firmware
it stops working after some fixed period after it's been used
Tony Zarro
@tzarro
Jan 05 2016 20:19
Do the g5s have the same (or similar) share functionality?
Ben West
@bewest
Jan 05 2016 20:19
er, it's not the same
and it's not exactly similar
it's different
I think it can talk to 2 devices
the mechanism is mostly known
John Males
@johnmales
Jan 05 2016 20:20
thanks for the info @bewest @stavlor @scottleibrand. What is the EOL timeframe for the g5?
Paul Andrel
@stavlor
Jan 05 2016 20:20
@bewest no one seems to agree yet but i believe its based on actual sensor session time so you can technically leave the transmitter on the shelf for a week or two and it doesn't count according to some
Ben West
@bewest
Jan 05 2016 20:20
yes
it has to do with the logical time of a session
at end of session, it turns off and that's it
Jason Calabrese
@jasoncalabrese
Jan 05 2016 20:21
evil
Ben West
@bewest
Jan 05 2016 20:21
new transmitter required
it would be interesting to see if just maybe there was a command to perhaps change the session somehow
keep shifting start of session? ;-)
Paul Andrel
@stavlor
Jan 05 2016 20:23
heh potentially would be interesting but i made sure to get another g4 transmitter instead of taking dexcom up on the upgrade offer and potentially loosing functionallity
John Males
@johnmales
Jan 05 2016 20:24
Is there any advantage of the g5?
Paul Andrel
@stavlor
Jan 05 2016 20:24
not that my rep didn't try to talk me into the g5 heh
Ben West
@bewest
Jan 05 2016 20:24
given, the data access issues, I don't think so
right now we have raw and we have multiple ways to get it
having multiple ways makes it that much more trustworthy in some ways
easy to compare/contrast things
Paul Andrel
@stavlor
Jan 05 2016 20:25
other than the fact that if you use an iPhone you don't need the receiver no real advantages for android users and you lose alot of functionality
John Males
@johnmales
Jan 05 2016 20:26
Seems sad that years pass, and the new model is a step backwards. Not exactly unheard of in the area of medical devices. :worried:
Ben West
@bewest
Jan 05 2016 20:26
well, this is why I keep trying my conresversations with industry away from featu
dang sensitive trackpad
away from features
generally companies keep adding more features and then apologize for running out of resources
Paul Andrel
@stavlor
Jan 05 2016 20:27
@bewest indeed heh
John Males
@johnmales
Jan 05 2016 20:27
yes a lot to be said for doing the core functionality well
Jason Calabrese
@jasoncalabrese
Jan 05 2016 20:34
@bewest you're going to leave that commented out? intentional?
Ben West
@bewest
Jan 05 2016 20:41
no, not sure what happened rhere
Jason Calabrese
@jasoncalabrese
Jan 05 2016 20:46
using https://www.adafruit.com/products/1327 with it? guess I should order a couple of them
Ben West
@bewest
Jan 05 2016 20:49
those look good
I'm still evaluating different ways of doing ble
the big split is between using dbus (a linux only app, but language independent) vs native ported solution (depends on libbluetooth)
complexity is similar, the latter approach would continue to work on osx, windows
the former, only linux
oceantragic
@oceantragic
Jan 05 2016 21:13
Hi all, Dana suggested I pop in, say hello and ask who in the UK is either using our intended to use OpenAPS.
Dana Lewis
@danamlewis
Jan 05 2016 21:20
^ cc @oskarpearson
Oskar Pearson
@oskarpearson
Jan 05 2016 21:32
Hi @oceantragic - my wife is on closed loop. There are some other interested parties around too
We are in London. You?
Are you just getting started?
Oskar Pearson
@oskarpearson
Jan 05 2016 22:31
Oh, and also... what pump are you using and which model?
John Males
@johnmales
Jan 05 2016 22:31
This message was deleted
This message was deleted
oceantragic
@oceantragic
Jan 05 2016 22:55
Hi Oskar, it's not me it's my 15yo daughter. She has an Animas Vibe (unfortunately I guess), uses G4 on and off since June 2014 and xDrip for a year (this week).
@oskarpearson sorry I forgot, we're near Southampton.
Andrew Warrington
@andrew-warrington
Jan 05 2016 23:16
@mhaeberli yeah, my case requires a mechanically hacked RPi. The only thing which MUST be hacked is the outermost USB port (the rest is optional), but I admit the USB port was not a trivial matter. I have promised some documentation on how it was done which I will be delivering soon enough. I'd better set a deadline on that to make sure I get to it. :) Will be within a week.
Oskar Pearson
@oskarpearson
Jan 05 2016 23:33
@oceantragic I don't know anything about the Animas pumps, unfortunately. Or the G4 stuff, since we use the medtronic CGMS. Not entirely sure I'll be much help, but we can certainly try and coordinate things!
Dana Lewis
@danamlewis
Jan 05 2016 23:34
@oceantragic re: Animas, @marius is doing some work with the Animas Ping: https://vine.co/v/iqFEYaOx0AF
Martin Haeberli
@mhaeberli
Jan 05 2016 23:57
@andrew-warrington - thanks so much!