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

6th
Jul 2017
John Benjamin
@Bender1061
Jul 06 2017 00:11
@scottleibrand I was trying to get that in before you merged mmtune-more, but there are a couple errors on the MDT stuff, only echo related, but i put in #525 to fix them.
Scott Leibrand
@scottleibrand
Jul 06 2017 00:22
thanks. that was all down to my misunderstanding how it works and trying to "fix" it. thanks for correcting that.
Martin Haeberli
@mhaeberli
Jul 06 2017 02:05
@scottleibrand - thx - i’ll see if I can extract and revert...
wharper11
@wharper11
Jul 06 2017 02:15
how do I know if I'm running master or dev? openaps with xdrip and 722 medtronic.
Martin Haeberli
@mhaeberli
Jul 06 2017 02:16
on your Edison; cd /src/oref0<tab> to get to your oref0 directory
then git branch -a
should tell you your current branch
katie disimone
@Kdisimone
Jul 06 2017 02:18
@applehat how’s the urchin write up going? Anywhere I can peek at it? I tried to follow on gitter but am not having much luck
wharper11
@wharper11
Jul 06 2017 02:22
@mhaeberli - sorry - cd /src/oref0<tab>? I don't understand?
Martin Haeberli
@mhaeberli
Jul 06 2017 02:22
hang on a sec
i meant: cd ~/src/oref0
which should change you into the oref0 source directory
and then git branch -a should show you a list of branches including the currently select branch
hope that helps
wharper11
@wharper11
Jul 06 2017 02:25
*dev! thanks @mhaeberli . someone asked me yesterday and I had no clue.
Martin Haeberli
@mhaeberli
Jul 06 2017 02:25
:+1:
Cameron Chunn
@applehat
Jul 06 2017 02:37
@Kdisimone someone else ended up tackling the pancreabble stuff - But im not sure if he's gotten it completely documented. Last I looked it only really worked for Dexcom plugged into rig setups, because it was using the dex-glucose.json file
John Benjamin
@Bender1061
Jul 06 2017 02:50
@applehat @Kdisimone Pancreabble is in oref0-setup, Yeah it looks liek some of the commands are not in the newer version of docs, but you should be able to get it to work by running cd && ~/src/oref0/bin/oref0-setup.sh --btpeb=AA:BB:CC:DD:EE:FF with the btpeb being the mac address of your pebble. and then adding the pebble to your edison, just like adding a phone for BT. I haven't done much playing with it since Jan, but it should still work.
katie disimone
@Kdisimone
Jul 06 2017 02:51
The directions aren’t working for me. theres still a lot of holes. Like systemctl status bluetooth.service isn’t a valid option. And, I’m not certain…but I’d love to confirm…it would appear that setting this up is basically for a “I KNOW I’m going camping” situtation. Because my rig can’t maintain a pairing with both the iphone and the pebble. So, if I pair the pebble…that iPhone tether isn’t going to work.
@Bender1061 tried that…doesn’t work at all
Cameron Chunn
@applehat
Jul 06 2017 02:52
@Bender1061 yeah, they don't. Its super broken. =)
John Benjamin
@Bender1061
Jul 06 2017 02:52
looks like a weekend project
katie disimone
@Kdisimone
Jul 06 2017 02:52
my biggest problem seems to be this bluetooth.service
I’d really appreciate some help on it. @logichammer I’ve been going through the directions and it’s just not working for me.
Cameron Chunn
@applehat
Jul 06 2017 02:53
I had no issue getting my pebble talking to my edison, but stopped bothering with it before writing the stuff to make the data actually go to the pebble
katie disimone
@Kdisimone
Jul 06 2017 03:02
I haven’t been close to wanting to chuck something out the window and run over it with my car yet in OpenAPS, but this might be it. The instructions just aren’t there yet.
John Benjamin
@Bender1061
Jul 06 2017 03:03
hmm. my rig is still making all of the urchin upload stuff, I'm charging my offline peb now, I'll have to see if it starts working in the morning.
so unless something has changed in the Bluetooth on the edison, which that stuff is still same version. I don't know why it would not be working.
oh wait, they did change the firmware of the pebbles
katie disimone
@Kdisimone
Jul 06 2017 03:05
@Bender1061 give it a whirl…:)
John Benjamin
@Bender1061
Jul 06 2017 03:06
when they gave us the keys
Dana Lewis
@danamlewis
Jul 06 2017 03:06
@Kdisimone afaik, @logichammer was using the instructions in the Pancreabble readme, and has not updated the OpenAPS side docs
katie disimone
@Kdisimone
Jul 06 2017 03:07
@danamlewis I’m on those ones…still no dice. But, I’ll try a bit more
Dana Lewis
@danamlewis
Jul 06 2017 03:07
K. Sorry to not be more help. Haven't tried a Pancreabble install in over a year.
katie disimone
@Kdisimone
Jul 06 2017 03:08
no worries, thanks. I’ll keep trying. obviously it’s possible for others, just feels like docs are still missing.
Cameron Chunn
@applehat
Jul 06 2017 03:09
@logichammer has his own fork of it where he posted more instructions
katie disimone
@Kdisimone
Jul 06 2017 03:09
yup…looked there too
I’ve looked at four places, tried the various options in all four. hence the hair pulling
Dana Lewis
@danamlewis
Jul 06 2017 03:09
They totally are missing, been needing updates since oref0-setup.sh was created in October ish
John Benjamin
@Bender1061
Jul 06 2017 03:10
yeah the docs have a mix of old and new instructions for the oref0-setup
Cameron Chunn
@applehat
Jul 06 2017 03:11
oref0-setup could probably do a bit more, as well. If I ever get around to making it work for me, I'll update them. Thats if tho. I've been busy lately =/
Cameron Chunn
@applehat
Jul 06 2017 03:11
oref0-setup could probably do a bit more, as well. If I ever get around to making it work for me, I'll update them. Thats if tho. I've been busy lately =/
katie disimone
@Kdisimone
Jul 06 2017 03:12
@Bender1061 “should just have to” famous last words…go for it
John Benjamin
@Bender1061
Jul 06 2017 03:12
Im the one that put it in oref0-setup
katie disimone
@Kdisimone
Jul 06 2017 03:12
because i’ve already “just"
John Benjamin
@Bender1061
Jul 06 2017 03:13
i used it for a week cruise, worked great, but yes I'm charging the pebble now. have to do it tomorrow.
katie disimone
@Kdisimone
Jul 06 2017 03:17

These commands alone hang up my rig

# this installs bluez 5.44
killall bluetoothd &>/dev/null
sudo apt-get update
sudo apt-get install -y libusb-dev libdbus-1-dev libglib2.0-dev libudev-dev libical-dev libreadline-dev --fix-missing
cd $HOME/src/ && wget https://www.kernel.org/pub/linux/bluetooth/bluez-5.44.tar.gz && tar xvfz bluez-5.44.tar.gz || die "Couldn't download bluez"
cd $HOME/src/bluez-5.44 && ./configure --enable-experimental --disable-systemd &&  make && sudo make install && sudo cp ./src/bluetoothd /usr/local/bin/ || die "Couldn't make bluez"

# this installs libpebble2
sudo pip install libpebble2

#reboot the computer to make sure the correct bluetoothd is loaded...I needed to
sudo reboot

https://github.com/mddub/pancreabble/tree/e032d054c3c62e31387de6cb8d97d1245a410453

IN particular this one four times in a row
sudo apt-get install -y libusb-dev libdbus-1-dev libglib2.0-dev libudev-dev libical-dev libreadline-dev --fix-missing
gregtcm
@gregtcm
Jul 06 2017 03:19
I think we'll be seeing new Pancreabble docs and videos from @Kdisimone before the month is out! :)
katie disimone
@Kdisimone
Jul 06 2017 03:19
If I’m able to speak on video without swearing…LOL
gregtcm
@gregtcm
Jul 06 2017 03:20
Both a G and R rated version, maybe??
John Benjamin
@Bender1061
Jul 06 2017 03:20
what's wrong with using the bluez 5.37 we are using for phone bt tethering?
katie disimone
@Kdisimone
Jul 06 2017 03:20
:clap:
@Bender1061 That assumes “we” are all on the same build. I’m trying to do this as ANY JOE USER would. Following docs. New rig installation…how will it go. So far, horrible.
John Benjamin
@Bender1061
Jul 06 2017 03:21
i don't think the newer version of Bluez needed the experimental flag
katie disimone
@Kdisimone
Jul 06 2017 03:21
@Bender1061 Give it all a whirl
John Benjamin
@Bender1061
Jul 06 2017 03:22
I'm using latest and greatest oref0? and it installed everything, I've been defaulty installing the btmac since january, even though i don't have a pebble on or near the rig
katie disimone
@Kdisimone
Jul 06 2017 03:23
Then you have 5.44
if you’re using the latest and greatest…it’s not 5.37 anymore
John Benjamin
@Bender1061
Jul 06 2017 03:23
echo Checking for BT Mac, BT Peb or Shareble
if [[ ! -z "$BT_PEB" || ! -z "$BT_MAC" || ${CGM,,} =~ "shareble" ]]; then
    # Install Bluez for BT Tethering
    echo Checking bluez installation
    if ! bluetoothd --version | grep -q 5.37 2>/dev/null; then
        cd $HOME/src/ && wget https://www.kernel.org/pub/linux/bluetooth/bluez-5.37.tar.gz && tar xvfz bluez-5.37.tar.gz || die "Couldn't download bluez"
        cd $HOME/src/bluez-5.37 && ./configure --enable-experimental --disable-systemd && \
        make && sudo make install && sudo cp ./src/bluetoothd /usr/local/bin/ || die "Couldn't make bluez"
        oref0-bluetoothup
    else
        echo bluez v 5.37 already installed
    fi
fi
```
katie disimone
@Kdisimone
Jul 06 2017 03:24
That’s master? Dev is 5.44
John Benjamin
@Bender1061
Jul 06 2017 03:24
my bad, you are correct
I'm still running 5.37 due to I already had it installed on the edison
So yeah I'm starting to see the problem here
so I'm betting its the configure parameters will need to be changed, for the newer version of bluez that you are running
katie disimone
@Kdisimone
Jul 06 2017 03:28
root@edison1:~# sudo apt-get install -y libusb-dev libdbus-1-dev libglib2.0-dev libudev-dev libical-dev libreadline-dev
Reading package lists... Done
Building dependency tree       
Reading state information... Done
libdbus-1-dev is already the newest version.
libical-dev is already the newest version.
libusb-dev is already the newest version.
libreadline-dev is already the newest version.
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 libglib2.0-dev : Depends: libglib2.0-bin (= 2.48.0-1~bpo8+1) but it is not going to be installed
 udev : Breaks: ifupdown (< 0.8.5~)
        Breaks: systemd (< 224-2) but 215-17+deb8u7 is to be installed
E: Error, pkgProblemResolver::Resolve generated breaks, this may be caused by held packages.
root@edison1:~# packet_write_wait: Connection to fe80::fedb:b3ff:fe9a:2e77%en0 port 22: Broken pipe
Kathryns-iMac:~ iMac4K$
if I keep the --fix-missingit crashes rig every single time
John Benjamin
@Bender1061
Jul 06 2017 03:30
or some of the other dependencies won't work with the newer Bluez. Why did we go to the newer version of bluez to begin with?
Cameron Chunn
@applehat
Jul 06 2017 03:30
Removed the need for the --experimental
John Benjamin
@Bender1061
Jul 06 2017 03:31
but if it breaks the libglib2.0, does not help much
Cameron Chunn
@applehat
Jul 06 2017 03:31
Doesn't break it on mine
wait. let me go backup
and figure out why its being installed
John Benjamin
@Bender1061
Jul 06 2017 03:33
when I got all the to work, I'm def no bluetooth expert, I did a lot of digging to get the Bluetooth hotspot working with the phones (android and iphone) and to get this working. But def not a linux expert. Google is my friend!
Cameron Chunn
@applehat
Jul 06 2017 03:33
I think the error @Kdisimone is showing is unrelated.
at least unrelated to Bluez 5.44
John Benjamin
@Bender1061
Jul 06 2017 03:34
ok
and we did discover, some people had to do reflashes of jublinux on their edisons in order to get the bluetooth to work correctly, that ended up being the issue @danamlewis had.
Dana Lewis
@danamlewis
Jul 06 2017 03:38
Yep. But shouldn't be the issue here since everyone's being happy Bluetooth tethering ever since
John Benjamin
@Bender1061
Jul 06 2017 03:39
it has been nice!
<as I pat myself on the back>
:smile:
Dana Lewis
@danamlewis
Jul 06 2017 03:44
:+1:
Cameron Chunn
@applehat
Jul 06 2017 03:45
I assume that Bluez 5.44 isn't an issue either since people have been using it for a while now with no issues as well
This message was deleted
(im probably just defensive tho because I totally made the change to oref0 that updated to 5.44, and have no clue if its causing issues or not)
;D
FWIW tho @Kdisimone - my issues with the --btpebstuff was it trying to build urchin-status off of another "report" that didn't exist for me.
something about "unmarshled-glucose" if I remember correctly.
katie disimone
@Kdisimone
Jul 06 2017 03:51
yes…that’s what I get monitor/unzoned-glucose.json
yes, BT tether is way higher priority for me to preserve than getting urchin working
Cameron Chunn
@applehat
Jul 06 2017 03:53
unzoned** thats right
katie disimone
@Kdisimone
Jul 06 2017 03:53
I just want to get the docs in order more than anything. There’s about four different spots with Pancreabble instructions right now.
Cameron Chunn
@applehat
Jul 06 2017 03:54
I actually understand more about how openaps works now - so I think I could probably repair the unzoned-glucose issue - but I'll gladly let @Bender1061 take a shot at it before me, since he did it in the first place =)
(well, docs in order AND working ;))
John Benjamin
@Bender1061
Jul 06 2017 03:58
Oh please don't think I'll get offended, I've been spending far to much time working on this the past week, my wife is starting to complain about being an OpenAPS widower again.
katie disimone
@Kdisimone
Jul 06 2017 03:59
She can hangout with my husband. LOL
Cameron Chunn
@applehat
Jul 06 2017 04:00
@Bender1061 do you happen to remember what unzoned-glucose.json is supposed to be created by?
I assume its put in place by one of the CGM methods that I am not using
John Benjamin
@Bender1061
Jul 06 2017 04:01
I would have to look it up. But it's bedtime for me on the east coast
Cameron Chunn
@applehat
Jul 06 2017 04:01
Its "stay up to late" time for me on the east coast =D
Night!
katie disimone
@Kdisimone
Jul 06 2017 04:05
function mdt_get_bg {
    openaps report invoke monitor/cgm-mm-glucosetrend.json 2>&1 >/dev/null \
    && openaps report invoke cgm/cgm-glucose.json 2>&1 >/dev/null \
    && grep -q glucose cgm/cgm-glucose.json \
    && echo MDT CGM data retrieved \
    && cp -pu cgm/cgm-glucose.json cgm/glucose.json \
    && cp -pu cgm/glucose.json monitor/glucose-unzoned.json \
    && echo -n MDT New cgm data reformat \
    && openaps report invoke monitor/glucose.json 2>&1 >/dev/null \
    && openaps report invoke nightscout/glucose.json 2>&1 >/dev/null \
    && echo ted
oref0-pump-loop.sh
Cameron Chunn
@applehat
Jul 06 2017 04:06
medtronic sensor bg then
katie disimone
@Kdisimone
Jul 06 2017 04:06
somehow it’s not skipping that for g4-upload?
Cameron Chunn
@applehat
Jul 06 2017 04:06
so it looks like cgm-glucose is the "unzoned" glucose.
katie disimone
@Kdisimone
Jul 06 2017 04:06
ah
Cameron Chunn
@applehat
Jul 06 2017 04:07
Let me login to my rig and see what files I have
Cameron Chunn
@applehat
Jul 06 2017 04:22
anyone using mdt cgm wanna post up what their cgm-glucose.json format looks like?
Realisticly I think the piece that is relying on unzoned-glucose.json needs to just work with glucose.json
since everyone should have a glucose.json thats formatted the same.
PieterGit
@PieterGit
Jul 06 2017 05:26
@scottleibrand According to https://github.com/openaps/oref0/releases/ 0.2.0 is our latest release. Do you have the github right to change the "latest release" indicator to 0.4.2? It would be a good practice to see the major new features and bugs fixes in the changelog there.
LilDucky
@LilDucky
Jul 06 2017 05:36
Hi all, i have been researching OpenAPS for a good friend and am really looking forward to getting things running. The current pump is a Medtronics 640G with enlite CGM. From everything i have seen, this pump is being investigated but there is nothing really in existence for integration. He also has an older medtronic pump that is compatible with OpenAPS but no CGM transmitter for the enlite sensor.
Tim Street
@tim2000s
Jul 06 2017 05:40
@LilDucky Those transmitters can be bought from Medtronic, or they can invest in Dexcom.
LilDucky
@LilDucky
Jul 06 2017 05:41
I have some questions - is there anything on the horizon for 640G support in OpenAPS? Does the edison / explorer setup support the 754 pump as the CGM source if we can locate a compatible enlite transmitter? Thanks in advance and I hope that I can add value in the future.
Thanks @tim2000s, we are in rural australia and am looking for the transmitter.
Tim Street
@tim2000s
Jul 06 2017 05:43
640 - no
x54 is usable
And there are plenty of people looping of the Medtronic CGM set up. In the setup process you can choose it as your CGM source
LilDucky
@LilDucky
Jul 06 2017 05:46
I have seen some work by Lennart Goedhart that looks promising but nothing yet. Looks like the enlite transmitter is the best way forward at this time.
Tim Street
@tim2000s
Jul 06 2017 05:47
Lennart has done some good stuff yes, but the comms required don't work and it looks like something like RileyLink for 640G is needed
ps2 @ps2 hides
LilDucky
@LilDucky
Jul 06 2017 06:01
Thanks for the speedy replies that confirm my thoughts.
No current support, but we're working on trying to unlock the secret sauce needed for direct OTA communication with the 640g
No current support, but we're working on trying to unlock the secret sauce needed for direct OTA communication with the 640g
@LilDucky a really convoluted and stupid setup using both the 640g as a CGM receiver (with an android phone uploading it to Nightscout) and the older medtronic pump could be made to work, but man would it be a LOT of stuff to carry with you =)
LilDucky
@LilDucky
Jul 06 2017 07:13
@applehat thanks. I looked at the "hamburger with the lot" method you described but discounted it as being way too convoluted. Who is doing the secret sauce research on the 640G? Edited - Saw the link and reading eagerly.
LilDucky
@LilDucky
Jul 06 2017 07:31
@applehat let me know if there is anything that i can do to assist - I have some experience in the programming and reverse engineering space.
Cameron Chunn
@applehat
Jul 06 2017 07:49
About the only thing that would really be helpful that I can think of on my end is having another CNL24 to test with, but I plan to get one from my Endo next appointment.
LilDucky
@LilDucky
Jul 06 2017 07:51
Cool. Will be watching with interest.
Cameron Chunn
@applehat
Jul 06 2017 07:53
I ramble about stuff in the gitter channel linked in the docs, If you wanna join over there as well.
LilDucky
@LilDucky
Jul 06 2017 08:52
What is the channel? I only found this one so far
Paul Dickens
@thebookins
Jul 06 2017 10:34
This message was deleted
Tim Street
@tim2000s
Jul 06 2017 12:25
@LilDucky @applehat my first OpenAPS rig was a Pi3 with a CNL stick receiving BG data from the 640G and feeding stuff to the 522 with with the carelink stick. Aside from rubbish connectivity to the pump, I had to hack the Linux dev version of Lennart's code to generate the glucose.json file. It worked but was bulky, and was totally offline, which was nice. Used to eat batteries when I took it out though - 20,000mAh for a 12 hour period.
LilDucky
@LilDucky
Jul 06 2017 12:29
@tim2000s thanks for the insight. I am currently hacking around with a rPi1 which is almost a dead loss. I have the carelink USB talking to the pump but am now working on the CGM side. Any suggestions where to get an early model enlite transmitter? If I can get this then an Edison rig looks perfect. Do you still have the hacked CNL code anywhere? I would love to play with it to check out the rest of the NS setup.
dan lightner
@danimaniac
Jul 06 2017 12:53
Changed batteries in a dead G5 transmitter last night. I am transitioning from a G4 Platinum so I am somewhat unfamiliar with the G5 / OpenAPS integration. I have 2 questions: 1) What is the normal range for transmitter battery status as listed in xDrip+ System Status (mine is currently listed as "216 - ok"), and 2) Can I simply "runagain.sh" changing the cgm data source during setup?
Dana Lewis
@danamlewis
Jul 06 2017 12:53
2 ) yep
LilDucky
@LilDucky
Jul 06 2017 13:28
@applehat interesting paper on hacking the radio chip here http://www.ru.nl/publish/pages/769526/z_eduardo_lorente.pdf
Giuseppe
@giupo
Jul 06 2017 13:52
@scottleibrand it was a brand new battery, I was confident (already met that kind of error from my 554). What do you think about my idea: moving to an external microservice (deployed on a faster host like the one we deployed Nightscout) for autosens, helping my poor RPi Zero from those 9min of evaluation. Is it doable? Do you have hints where to start?
jmcrawford
@jmcrawford
Jul 06 2017 14:58
@giupo That would probably work, the tricky part would be how you update the inputs for autosens from your rig to the remote processing host, then update the outputs back to your rig for them to be used in your loop
Giuseppe
@giupo
Jul 06 2017 15:03
@jmcrawford help me understand what you mean (I'm not a native english speaker, maybe I miss something, sorry): Woudn't be as easy as identify the preconditions for autosens and its input parameters, then invoke a remote microservice: basically a sync HTTP call to a remote service implemented with the same business logic used to evaluate autosens locally) and then wait for the response. Since the lingua-franca of the rig is JSON, so it would be the microservice. I guess I have to lurk on the crontab to understand what's going on under the hood
Cameron Chunn
@applehat
Jul 06 2017 15:11
@LilDucky https://gitter.im/medtronic-flash/Lobby is the channel for all the reversing of medtronic stuff / general other nonsense =P
jmcrawford
@jmcrawford
Jul 06 2017 15:13
@giupo If you check openaps.ini for the autosens report, it reads a bunch of files locally (settings/basal_profile.json, and others). You'd need some way for the remote service to access the content of those files, or perhaps send the info with the remote request?
Giuseppe
@giupo
Jul 06 2017 15:14
@jmcrawford ahh! Now I see your point: there's a lot of "fragmentation" (if I can call this way), and there's a lot of stuff needed for evaluation to succeed. Actually (I'm brainstorming here) we should identify where we spend most of the time and externalize that piece of code. I don't know if I'm daydreaming ...
@jmcrawford ...or basically format an HTTP request with all those contents .. :)
Cameron Chunn
@applehat
Jul 06 2017 15:19
I'm late to this conversation. Why are we wanting autosense "in the cloud
"?
jmcrawford
@jmcrawford
Jul 06 2017 15:20
It's a good idea, for situations like yours with lower power rigs. I don't know how much data you'd need to send with that kind of request. The other thing to consider is what happens when you lose access to the remote host, you lose your autosens tuning
Giuseppe
@giupo
Jul 06 2017 15:20
@jmcrawford I would revert to local autosens :(
@applehat just brainstorming: since I've experienced extreme slowness on my RPi zero, and Someone pointed me out timing on RPi zeri, where autosense takes 9mins to compelte I was wondering if it's worth and viable the option to move the autosens alg on a remote microservice.
Cameron Chunn
@applehat
Jul 06 2017 15:22
Ahh.
jmcrawford
@jmcrawford
Jul 06 2017 15:23
settings/glucose.json and settings/pumphistory-24-zoned.json are both ~100k in size...
Cameron Chunn
@applehat
Jul 06 2017 15:23
This is why dev boards like the NanoPi Neo Air are interesting to me. Small as a Zero, but with a Quad Core (that can be clocked and cores enabled / disabled from user space). Run on a single 900MHz core until you fired something like autosense, then spin up a single 1.2GHz core to do that work (and eat battery) for a moment.
Giuseppe
@giupo
Jul 06 2017 15:23
@jmcrawford holy s...
@applehat where is it ? :)
jmcrawford
@jmcrawford
Jul 06 2017 15:25
@applehat Nice. I haven't looked at the specs for those boards, but that sounds attractive
Giuseppe
@giupo
Jul 06 2017 15:25
Haven't someone used it for OpenAPS?
Cameron Chunn
@applehat
Jul 06 2017 15:25
I've been planning to try and build a rig out of one and see what sort of performance it can achieve
And no @giupo - not that I'm aware of.
But linux is linux if you try hard enough ;)
Giuseppe
@giupo
Jul 06 2017 15:26
ahhh... I guess I could be the first :D
Cameron Chunn
@applehat
Jul 06 2017 15:26
Would be pretty easy to write a small script for it to manage what cores are turned on as well.
jmcrawford
@jmcrawford
Jul 06 2017 15:26
Take notes, let us know how it goes
Giuseppe
@giupo
Jul 06 2017 15:27
let me see it's I/O capabilities...
Cameron Chunn
@applehat
Jul 06 2017 15:27
Its only downside is the 512mb of ram, but you already have that limit on the Pi Zero W anyway
They are "Pi Compatible"
Sort of. The headers are different.
Has an onboard eMMC which is nice, but no onboard antenna for Wifi and Bluetooth (which is less nice)
Giuseppe
@giupo
Jul 06 2017 15:29
@applehat the fact that says wifi-Bluetooth is due some sort of extensions to buy?
Cameron Chunn
@applehat
Jul 06 2017 15:30
It has Wifi and Bluetooth onboard, you just have to plug up the included external antenna
Lots of interesting possibilities with it tho. With the right lipo charging circuit, you could even alert the board if its charging or not, and spin up all 4 cores if its plugged up (for overnight autotune and such)
Giuseppe
@giupo
Jul 06 2017 15:33
I'm horrible at soldering, tough :(
Those are solderless headers you could use
And this little guy, if you wanted something to easilly test with
I mainly like that board because its all priced super cheap =P
Giuseppe
@giupo
Jul 06 2017 15:44
@applehat :+1:
jmcrawford
@jmcrawford
Jul 06 2017 16:33
What to use for pumps comms, absent the Explorer? Or could we convince ERD to modify the explorer to break out the pins to something other than the hirose connector?
Dana Lewis
@danamlewis
Jul 06 2017 16:34
EB can be modified to go with our computer of choice probably. A couple of options as contenders right now
Eric
@ecc1
Jul 06 2017 18:40
I'm in the early stages of working with a hardware engineer on a Raspberry Pi "pHat", a Pi-Zero form-factor board that will mate with a Pi Zero W (but could also be wired, less neatly, to any other embedded Linux board). It will have an RFM69 and a CC2500 on board for 915MHz and 2.4GHz proprietary comms (pump and G4, for example).
alternateal
@alternateal
Jul 06 2017 19:20
Upgraded to latest dev branch - struggling to get automatic connection to iPhone PAN using Bluetooth - used to work perfectly. Following instructions I can create connection to PAN but when dropping wifi waiting for auto connect doesn't work. Suggestions?
katie disimone
@Kdisimone
Jul 06 2017 19:58
@alternateal we have just experienced the same.
mkericson
@mkericson
Jul 06 2017 20:22
\ @Kdisimone I'm just getting started. I'm one of the "Any Joe User"'s. Read the doc's over many many time. Wanted to say Thank You! for posting your videos this past week. Very helpful. - Just got my hardware preflashed by the hamshack guys so may get to run through the setup this weekend. Have not used command lines since the DOS 3.2 days.
mkericson
@mkericson
Jul 06 2017 20:29
Visual Question: I have NS setup and running but I notice that it is flaky catching all my data even though I keep the app running. -Dexom G5. When I see posts here NS seems to show very clean data. BTW my dexcom graphs don't have any gaps just what pulls through NS.
Tim Street
@tim2000s
Jul 06 2017 20:34
@mkericson which host are you using for NS. Azure or Heroku?
mkericson
@mkericson
Jul 06 2017 20:35
@tim2000s Heroku
Tim Street
@tim2000s
Jul 06 2017 20:36
Check out the keep awake robot. You may need that. It's in the Heroku NS instructions
katie disimone
@Kdisimone
Jul 06 2017 20:36
@mkericson Do you have the cob on your ENABLE line? That caused my site to run jumpy.
Oh and definitely what @tim2000s said if you aren't looping yet
mkericson
@mkericson
Jul 06 2017 20:37
@Kdisimone this is my ENABLE: bridge openaps pump iob basal careportal sage maker
Meghan Rutledge
@mar5477
Jul 06 2017 21:15
@Kdisimone @applehat the unzoned glucose comes from using mdt as your cgm source. I went into /src/oref0/lib/oref0-setup/pancreabble.json and changed the glucose_history in the upload/urchin-looplog.json report to just use "monitor/glucose.json"
katie disimone
@Kdisimone
Jul 06 2017 21:15
👍🏻
Ah, so even though my cgm type was set to G4-upload, somehow that's not changing that...gotcha. Thanks. I'll be trying to work on this again tonight
Meghan Rutledge
@mar5477
Jul 06 2017 21:17
I don't get the unzoned error anymore, but I do get a displau_ error.
*display_time
PieterGit
@PieterGit
Jul 06 2017 21:28
Can somebody who uses Dexcom connected with USB OTG help me with openaps/oref0#527
Meghan Rutledge
@mar5477
Jul 06 2017 22:01
@Kdisimone I'm not sure it'll solve all your problems, but I think it'll get you closer. I'm not up and running with pancreabble yet either.
useraps
@useraps
Jul 06 2017 22:48
Hello everyone. My name is John. I'm dad to Braydon, a T1D who will be 6 on 7/9. He was diagnosed at 11 months and started pumping around 1.5yrs old - Animas Ping. I heard about this community on the juicebox podcast and so I had buy a rig and try and get it going. Just the thought of getting a good night sleep will be worth it. So I have the Edison preinstalled with Jubilinux and the Explorer Block. I set it up last night and got my wifi working. However, I have a question. When both USB's are connected, I do not see the Edison Device under my devices. I see a Multifunction Composite Gagdet which says RNDIS the drivers aren't installed. Under ports I have Com3 Serial Port and Com4 Serial Device. I did install the Stand alone Edison driver. Do I have anything to be concerned with or should I keep moving forward? Sorry if this has been addressed and for being so long. Thanks
Scott Leibrand
@scottleibrand
Jul 06 2017 23:08
if you're pre-flashed, you just need to know what COM port to connect to with PuTTY. Sounds like that might be COM4. If that works, you're good to go.