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

21st
Aug 2016
garykidd
@garykidd
Aug 21 2016 02:01
so what's the recommended firmware software image for intel edison for APS instance?
Martin Haeberli
@mhaeberli
Aug 21 2016 02:37
@garykidd :+1: on your question - I don't know either...
garykidd
@garykidd
Aug 21 2016 02:39
psyched...just got edison and starting to configure now
Eric Jensen
@elnjensen
Aug 21 2016 02:45
@garykidd I just got one today and am sitting here configuring it right now! :smile: Amazingly small - looking forward to having this be my main rig.
Other people have been using mainly ubilinux (I gather) but I'm going to try yocto, since it seems better-supported in the long run. Following the steps here to get started: https://github.com/oskarpearson/mmeowlink/wiki/Prepare-Edison-for-OpenAPS-with-Yocto
garykidd
@garykidd
Aug 21 2016 02:47
@elnjensen so cool right? thinking sam ething. I just got mine to access wifi and configured ssh...but cant seem to ssh into it...do you know how?
Eric Jensen
@elnjensen
Aug 21 2016 03:05
You need to figure out its IP address. If you are logged in via serial console, then do ifconfig and look under the wlan0 entry to find IP.
Then you can do ssh root@1.2.3.4 (putting in your IP address there.
garykidd
@garykidd
Aug 21 2016 03:15
I was using Intels board configuration tool via the micro usb cable attached tomy imac
Eric Jensen
@elnjensen
Aug 21 2016 03:21
Right. Once you've completed that, then you can find directions that let you connect to it via a second cable to the other port. But, if SSH is really running and you can figure out the IP address some other way, you can go straight to the method above. Can you see what devices are connected to your home router? If so, then see if you can find the IP address that way.
jmcrawford
@jmcrawford
Aug 21 2016 03:22
@elnjensen On a phone or tablet, an app called Fing will scan your network and identify the devices. Other apps work as well, but this is the one I use.
garykidd
@garykidd
Aug 21 2016 03:24
hmmm...i have the IP but it wouldnt let me ssh using.... '''ssh root@1.2.3.4'''
Eric Jensen
@elnjensen
Aug 21 2016 03:24
Cool, thanks @jmcrawford
@garykidd You need to find your own ip address and put it in there - the 1.2.3.4 was just a made-up placeholder.
might be something like 192.168.x.x.
garykidd
@garykidd
Aug 21 2016 03:25
yup...did do that with right IP.
got a could not resolve hostname error
jmcrawford
@jmcrawford
Aug 21 2016 03:27
Some systems won't allow root to access via ssh - it's a security risk. There should be a default user/password, but each distro will have a different setup - what are you using - yocto, ubilinux?
garykidd
@garykidd
Aug 21 2016 03:27
ssh root@192.168.xxx.xxx
I literally installed latest firmware available via intel's board configuration utility
I did se tup a p/w...but figured it would just ask me for it after trying to ssh into it..?
Eric Jensen
@elnjensen
Aug 21 2016 03:30
Hmm. OK, try connecting with a cable - follow directions here and you can connect via cable. Then perhaps you can see what's happening: https://software.intel.com/en-us/setting-up-serial-terminal-intel-edison-board
jmcrawford
@jmcrawford
Aug 21 2016 03:30
depends on what you're ssh'ing from. On windows, if you're using Putty, it will ask. If you are connecting from another unixish (linux, OSX, etc) it'll assume the user you are logged in as, unless you specify the user as you did (root@...)
try ssh edison@192...
garykidd
@garykidd
Aug 21 2016 03:40
hmmm.....still not sure yet. iMac:~ g$ ssh root.192.168.1.172
ssh: Could not resolve hostname root.192.168.1.172: nodename nor servname provided, or not known
oops @
Jason Calabrese
@jasoncalabrese
Aug 21 2016 03:42
@jyaw I'm not using this same setup, but @johnmales has a nice gist
garykidd
@garykidd
Aug 21 2016 03:43
iMac:~ g$ ssh root@192.168.1.172
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@ WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED! @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
IT IS POSSIBLE THAT SOMEONE IS DOING SOMETHING NASTY!
Someone could be eavesdropping on you right now (man-in-the-middle attack)!
It is also possible that a host key has just been changed.
The fingerprint for the ECDSA key sent by the remote host is
Zzzzzzzhhhhhhhhhdddhdhahdkjabfsjsbdivbsjvbjasbdkvbnk.
Please contact your system administrator.
Add correct host key in /Users/g/.ssh/known_hosts to get rid of this message.
Offending ECDSA key in /Users/g/.ssh/known_hosts:4
ECDSA host key for 192.168.1.172 has changed and you have requested strict checking.
Host key verification failed.
Jason Calabrese
@jasoncalabrese
Aug 21 2016 03:43
I'm using the $10/month digitalocean instance, but with a single site you could use the $5/month size
Eric Jensen
@elnjensen
Aug 21 2016 03:44
Ah, you're almost there. That just means that in the past you connected to a different host on your network that had that IP address.
jmcrawford
@jmcrawford
Aug 21 2016 03:44
edit the file /Users/g/.ssh/known_hosts and delete line 4, then try the ssh command again
Eric Jensen
@elnjensen
Aug 21 2016 03:44
And it's warning you that this isn't that same host.
So do what @jmcrawford just suggested and you should be set.
garykidd
@garykidd
Aug 21 2016 03:45
sorry so slow here...how exactly to I edit the file
Eric Jensen
@elnjensen
Aug 21 2016 03:46
What kind of computer?
garykidd
@garykidd
Aug 21 2016 03:46
imac
Eric Jensen
@elnjensen
Aug 21 2016 03:47
OK, in Terminal:
open /Users/g/.ssh/known_hosts
That should open TextEdit with the file, then you can delete that line and save.
OK, I need to head to bed here. Hope you get it figured out!
garykidd
@garykidd
Aug 21 2016 03:50
@elnjensen rest easy. thanks for help.
bout to pack it in myself
so after open /Users/g//.ssh/known_hosts
I get
192.168.1.119 ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBCU9YmnsfJs5GzErGEI2O18RU5bYcqcehxWjXkghIzIAJypiHperJ22iTaknoeluiq3yY8qoLUFm+ZahOfadyrU=
raspberrypi.local ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBCU9YmnsfJs5GzErGEI2O18RU5bYcqcehxWjXkghIzIAJypiHperJ22iTaknoeluiq3yY8qoLUFm+ZahOfadyrU=
fe80::ba27:ebff:fe71:2393%en1 ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBCU9YmnsfJs5GzErGEI2O18RU5bYcqcehxWjXkghIzIAJypiHperJ22iTaknoeluiq3yY8qoLUFm+ZahOfadyrU=
aps1_edison.local,192.168.1.172 ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBFErKPpiY1Bg4AcK6DUGF3uvzcTcuFuvTne9S9x4oM8QQttTSooUsKGRRHh97z1aWJCKGdB2XM/9/0hA37G+r2E=
fe80::86ff:fe42:7f97%en6 ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBFErKPpiY1Bg4AcK6DUGF3uvzcTcuFuvTne9S9x4oM8QQttTSooUsKGRRHh97z1aWJCKGdB2XM/9/0hA37G+r2E=
aps_1.local ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBBUiTojRTYh3a0WlHDB19UxfSK3y3/xTb+LPWbEoQX3HmRKjDS15V2Qd77uZRYB22tBGKPYQqAmhsRkoZvsZcaU=
jmcrawford
@jmcrawford
Aug 21 2016 03:55
delete this line:
aps1_edison.local,192.168.1.172 ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBFErKPpiY1Bg4AcK6DUGF3uvzcTcuFuvTne9S9x4oM8QQttTSooUsKGRRHh97z1aWJCKGdB2XM/9/0hA37G+r2E=
then save the file
garykidd
@garykidd
Aug 21 2016 03:57
got it thanks. forgot how to configure so I dont need to enter p/w constantly.
jmcrawford
@jmcrawford
Aug 21 2016 04:02
Glad you got it. This is the one area that I'm strong in, it's the rest of the process that I'm working at understanding
garykidd
@garykidd
Aug 21 2016 04:09
well got ssh'd in but not much further.

go to part on .... Install git and cronie

cronie not included on base Yocto 3.0 follow the steps bellow:

vi /etc/opkg/base-feeds.conf
Add the following:

src/gz all http://repo.opkg.net/edison/repo/all
src/gz edison http://repo.opkg.net/edison/repo/edison
src/gz core2-32 http://repo.opkg.net/edison/repo/core2-32
src all http://iotdk.intel.com/repos/2.0/iotdk/all
src x86 http://iotdk.intel.com/repos/2.0/iotdk/x86
src i586 http://iotdk.intel.com/repos/2.0/iotdk/i586

not sure how you exactly ...."Add the following:"
jmcrawford
@jmcrawford
Aug 21 2016 04:16
The 'vi' command is a text editor to open the file. So enter
vi /etc/opkg/base-feeds.conf
use the arrow keys to move to the bottom of the file, then press o (the small letter)
copy/paste all those src lines, then to save the file, press the escape key, then type
:wq
to save the file. vi is not the most user friendly tool.
jmcrawford
@jmcrawford
Aug 21 2016 04:21
you could check to see if you've got nano installed, it's a bit more friendly.
garykidd
@garykidd
Aug 21 2016 04:27
I cant seem to figure out how to copy/paste....wth
Scott Leibrand
@scottleibrand
Aug 21 2016 04:30
Right-click? Or shift-insert if you're using putty.
jmcrawford
@jmcrawford
Aug 21 2016 04:32
If you're working in Terminal on your mac, just use copy/paste from the edit menu. But you need to have put vi into edit mode (using o) before you paste.
Scott Leibrand
@scottleibrand
Aug 21 2016 04:34
i is typically used for inserting, fwiw. o inserts on a new line below the current cursor.
garykidd
@garykidd
Aug 21 2016 04:39
root@APS_1:~# opkg install cronie
Installing cronie (1.4.11-r0) on root.
Downloading http://iotdk.intel.com/repos/3.5/iotdk/edison/core2-32/cronie_1.4.11-r0_core2-32.ipk.
Installing pam-plugin-access (1.1.6-r5) on root.
Downloading http://iotdk.intel.com/repos/3.5/iotdk/edison/core2-32/pam-plugin-access_1.1.6-r5_core2-32.ipk.
Running groupadd commands...
NOTE: Performing groupadd with [ --system crontab] and 10 times of retry
Configuring pam-plugin-access.
Configuring cronie.
ln -s '/lib/systemd/system/crond.service' '/etc/systemd/system/multi-user.target.wants/crond.service'
Collected errors:
Is that OK?
jmcrawford
@jmcrawford
Aug 21 2016 04:42
I wonder if that means that one of the lines you inserted into /etc/opkg/base-feeds.conf already existed
Other than that, I'd say it looks like it ran cleanly.
Eric Luhrs
@eluhrs
Aug 21 2016 05:03
@garykidd you may find the nano editor easier than vi. It ships with yocto, so just type nano filename. It will show you the most common commands towards the bottom of the screen.
jmcrawford
@jmcrawford
Aug 21 2016 05:03
@eluhrs +1
Eric Luhrs
@eluhrs
Aug 21 2016 05:06
And that error above simply means that you have two repositories named src in your base-feeds.conf file. Just change one to src2 (or something else) and run the opkg update command again.
Jarred Yaw
@jyaw
Aug 21 2016 05:18
@jasoncalabrese @johnmales ya that is definitely not trivial! Glad I asked. Awesome gist, John! Are you saying your config is just slightly different Jason?
@jasoncalabrese ah, I read a bit further on and saw your comment on the $10/mo instance. My Azure CPU(days) counter reset, so all seems to be ok for now. I'll transition in the next week or so though, so that gist should come in super handy.
garykidd
@garykidd
Aug 21 2016 05:30
I tried to redo everything now I get...
192.168.1.119 ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBCU9YmnsfJs5GzErGEI2O18RU5bYcqcehxWjXkghIzIAJypiHperJ22iTaknoeluiq3yY8qoLUFm+ZahOfadyrU=
raspberrypi.local ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBCU9YmnsfJs5GzErGEI2O18RU5bYcqcehxWjXkghIzIAJypiHperJ22iTaknoeluiq3yY8qoLUFm+ZahOfadyrU=
fe80::ba27:ebff:fe71:2393%en1 ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBCU9YmnsfJs5GzErGEI2O18RU5bYcqcehxWjXkghIzIAJypiHperJ22iTaknoeluiq3yY8qoLUFm+ZahOfadyrU=
fe80::86ff:fe42:7f97%en6 ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBFErKPpiY1Bg4AcK6DUGF3uvzcTcuFuvTne9S9x4oM8QQttTSooUsKGRRHh97z1aWJCKGdB2XM/9/0hA37G+r2E=
aps_1.local ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBBUiTojRTYh3a0WlHDB19UxfSK3y3/xTb+LPWbEoQX3HmRKjDS15V2Qd77uZRYB22tBGKPYQqAmhsRkoZvsZcaU=
192.168.1.172 ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBBUiTojRTYh3a0WlHDB19UxfSK3y3/xTb+LPWbEoQX3HmRKjDS15V2Qd77uZRYB22tBGKPYQqAmhsRkoZvsZcaU=
......
DO I delete the fifth line now?
jmcrawford
@jmcrawford
Aug 21 2016 05:31
Did you get that same error :
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@ WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED! @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
garykidd
@garykidd
Aug 21 2016 05:31
i did...yes
jmcrawford
@jmcrawford
Aug 21 2016 05:55
Yes, you'd delete the fifth line.
Mitchell Slep
@moomoobloo
Aug 21 2016 06:07
so what's the protocol when traveling to a new timezone? I changed the time on the pump and cgm, and changed the timezone on the pi, but things are getting uploaded to nightscout in the wrong timezone.
Brandon Faloona
@bfaloona
Aug 21 2016 06:17
@moomoobloo I don't have an answer for you, but I'm eager to include the answer in the operations notes i'm writing.
Are you uploading glucose data to nightscout?
Which things are uploaded with the wrong timezone?
Ben West
@bewest
Aug 21 2016 06:23
if you change timezone on the pi, need to also mind the time of the device itself
Scott Leibrand
@scottleibrand
Aug 21 2016 06:30
@moomoobloo is there a timezone hardcoded in your ini file? If so, you'll want to remove it. Newer versions of openaps leave that off to allow roaming.
Mitchell Slep
@moomoobloo
Aug 21 2016 06:32
No hardcoded timezones, I assumed that (along with the Pi's RTC) would mean I didn't need to change anything on the Pi. But after changing just the pump clock, treatments were getting uploaded to Nightscout in the future (I'm traveling from west coast to east coast).
So I changed the timezone on the Pi using sudo dpkg-reconfigure tzdata and that seems to have gotten things working correctly.
I deleted all the boluses from NS that were uploaded incorrectly using Edit Mode, and was able to re-upload using the correct time zone.
there are still some temp basals in the future, but i assume they will get overwritten
@bewest what else do I need to change about the time on pi? after changing the timezone, it looks like I
I've got the correct local time and time zone
Mitchell Slep
@moomoobloo
Aug 21 2016 06:39
(oops instead of RTC I meant to say NTP)
Ben West
@bewest
Aug 21 2016 06:40
there's a tool in NS to delete future data
in the admin area
when shifting zones, both the pump/cgm and the pi need to all have the same zoning/clock info
Mitchell Slep
@moomoobloo
Aug 21 2016 06:43
right, so for pump/cgm I change the clock, and for Pi I change the zone and NTP should fix the clock?
Ben West
@bewest
Aug 21 2016 06:44
yeah, mainly the sudo dpkg-reconfigure tzdata forces the system zone to change
if the time is already accurate, changing the zone should be enough
with these minicomputers, they don't have a hardware clock, so when they get turned off, the clock often gets messed up
Mitchell Slep
@moomoobloo
Aug 21 2016 06:46
has anyone tried adding RTC to the pi?
Jason Calabrese
@jasoncalabrese
Aug 21 2016 06:48
@jyaw my config is pretty old, but since it's been working I haven't done anything with it, I'm not using pm2, just an upstart script to launch node. I run a local mongo there too, if you haven't run into mlab limits its only a matter of time
Scott Leibrand
@scottleibrand
Aug 21 2016 06:53
@moomoobloo see also clockset.sh in oref0
It can handle setting the device times for you after you set the Pi tz, or set time on Pi based on device time if you're offline with no NTP.
Martin Haeberli
@mhaeberli
Aug 21 2016 07:04
@moomoobloo I have added an RTC; its pretty cheap and easy, tho it uses some pins I suspect are probably also needed if one is using Slice of Radio. Looks like one can attach an RTC with for example I2C or SPI, too...
Brandon Faloona
@bfaloona
Aug 21 2016 07:05
@scottleibrand Can you elaborate on when and how to run clockset.sh? (The why part seems clear.)
Run it once?
As is, or modified to match my config?
What's the command to use?
John Males
@johnmales
Aug 21 2016 08:43
@jyaw thanks hopefully it helps, let me know if something doens’t make sense.
Hristina Dimova
@xpucuto
Aug 21 2016 10:16
@jmcrawford the Fing app is incredible . Thanks
andyhandy
@andyhandy
Aug 21 2016 10:40
hello all. sorry for the newbie question (and please direct me to the right place if this is the wrong forum!). i had just started setting up my pi and added devices (dexcom g5 and medtronic paradigm 723) when it crashed. it seems the 8gb card is now full:Filesystem Size Used Avail Use% Mounted on
/dev/root 6.0G 6.0G 0 100% /
devtmpfs 459M 0 459M 0% /dev
tmpfs 463M 0 463M 0% /dev/shm
tmpfs 463M 6.3M 457M 2% /run
tmpfs 5.0M 4.0K 5.0M 1% /run/lock
tmpfs 463M 0 463M 0% /sys/fs/cgroup
/dev/mmcblk0p6 63M 20M 44M 32% /boot
tmpfs 93M 0 93M 0% /run/user/1000
any idea why that might be and how i would fix it? thanks
Hristina Dimova
@xpucuto
Aug 21 2016 10:42
@andyhandy did you expand the filesystem of the SD card?
i did miss that line with my own setup ;) you can find it here https://github.com/openaps/docs/blob/master/docs/docs/walkthrough/phase-0/rpi.md and search for expand
andyhandy
@andyhandy
Aug 21 2016 10:44
you mean through rasps-config? i tried to but it wouldn't let me because i did a noobs install (cana kit)
Hristina Dimova
@xpucuto
Aug 21 2016 10:45
aha i am not familiar with that as i didnt start with cana kit . so you need help for from someone more experinced in linux than me. Sorry i couldnt help
andyhandy
@andyhandy
Aug 21 2016 10:45
no worries - thanks!!
Hristina Dimova
@xpucuto
Aug 21 2016 10:45
:+1:
tazitoo
@tazitoo
Aug 21 2016 11:49
@andyhandy - this will free up space that is currently taken up with updates, etc - 'sudo apt-get clean'
it might give you some breathing space. Someting seems to be eating up most of your space though...you'll have to do some forensics to find it...
tazitoo
@tazitoo
Aug 21 2016 11:59
Some suggestions to find largest files and directories in this thread:
Hristina Dimova
@xpucuto
Aug 21 2016 11:59
When updating oref0 i noticed this warning , can someone tell me if i have to update something else ?
pi@malinka2:~/bulgaria $ sudo npm install -g git://github.com/openaps/oref0.git'#dev'
npm WARN engine cryptiles@2.0.5: wanted: {"node":">=0.10.40"} (current: {"node":"0.10.29","npm":"1.4.21"})
npm WARN engine boom@2.10.1: wanted: {"node":">=0.10.40"} (current: {"node":"0.10.29","npm":"1.4.21"})
npm WARN engine hoek@2.16.3: wanted: {"node":">=0.10.40"} (current: {"node":"0.10.29","npm":"1.4.21”})
Lukasz Kaniowski
@lukasz-kaniowski
Aug 21 2016 14:23
hi guys, my girlfriend recently upgraded her pump to 640g. I've read that currently there is no way of using openaps for bolusing for that model. We would be ok if we even could get the cgm data from it. We are getting error though after updating pump serial number: TypeError: Non-hexadecimal digit found
Pete Schwamb
@ps2
Aug 21 2016 14:30
530g is the last Medtronic pump to use 915mhz, I believe
Lukasz Kaniowski
@lukasz-kaniowski
Aug 21 2016 14:32
we are able to get data using the carelink usb and medtronic app
serial number for the pump has some letters and numbers, for the previous pump model there ware only numbers
Pete Schwamb
@ps2
Aug 21 2016 14:33
Hmm, interesting
Lukasz Kaniowski
@lukasz-kaniowski
Aug 21 2016 14:33
ok, scratch that, I'm probably doing it wrong
just found this documentaiton
it was some time since I've configure the last pump
will check it now
Pete Schwamb
@ps2
Aug 21 2016 14:35
That's really interesting that you can get data via carelink. I saw that the 640g uses the contour next link 2.4 and assumed that meant that it did comms on 2.4 ghz only.
Is the enlite 2 still on 915mhz maybe?
Lukasz Kaniowski
@lukasz-kaniowski
Aug 21 2016 14:38
actually she just told me that she didn't test that yet. But sales rep was assuring that carelink usb will definitaly work. We are going to test it now and update the channel in a sec.
Andrew Warrington
@andrew-warrington
Aug 21 2016 14:39
Hi all, back on August 8th or 9th I reported that my EU-pump-based-loop was not working well in US. Just to summarise again: 1) using Medtronic CGM, and the CGM/Pump communication was also bad (had to have them about 1 inch apart max or would get a weak signal), 2) I would only get a successful comm for large reports like pump history and glucose history, one time out of every six or so, when my APS was 30 cm away from the pump (neither closer nor further) and oriented in a particular direction. In short, it was a mess, and the whole time I was there I wondered whether it was just interference with the 868 mhz band the European pump uses, or another problem which I could not identify. In the end I concluded must be interference. I did not have any tools with me, like SDR, to check.
Just writing back as I got home yesterday, and since then everything is running perfectly.
So yes, I am now pretty sure this was a radio interference issue.
Eric
@ecc1
Aug 21 2016 15:14
Pretty sure there are 2.4GHz Carelink sticks for the newer pumps -- I remember coming across them in the FCC database
Black housing vs. off-white IIRC

got to Make sure your edison has a recent build installed

root@edison:~# configure_edison --version
159.devkit
If the above command errors, or shows an older version, please update your edison following the instructions here: https://software.intel.com/en-us/iot/hardware/edison/downloads

the configure_edison --version results in....
root@edison:~# configure_edison --version
Could not determine firmware version information. Quitting.
none
SO I followed and re-installed latest build per the instructions....and it still didnt work.
...Can I ignore the error?
Eric Jensen
@elnjensen
Aug 21 2016 15:43
I ran into the same problem . I had already configured the edison via Intel's gui tool, but the command-line version wouldn't see the firmware version. I tried configure_edison --setup, but it seemed like that was trying to configure things (wifi, ssh) that I had already configured. I moved on in the setup, but would be curious to hear from others ( @amazaheri , @ps2 ?) whether these instructions are still necessary.
This is following the wiki, Prepare Edison for OpenAPS with yocto . Happy to edit / update if others confirm.
Ali Mazaheri
@amazaheri
Aug 21 2016 15:48
Hi there, u can ignore the error, I need to validate the steps with latest yocto build as steps are based on 3.0 😀
Eric Jensen
@elnjensen
Aug 21 2016 15:52
Ali, should the "2.0" strings in the repositories that go into base-feeds.conf probably be replaced with "3.0" ?
(Assuming I'm running a 3.0 build, that is.)
garykidd
@garykidd
Aug 21 2016 16:11
@elnjensen thanks. Funny thing is I used intel's GUI to set up and then reverted to the link you just shared.
did you follo wthe rest of the instructions though?
Eric Jensen
@elnjensen
Aug 21 2016 16:27
I went through most of the rest of the instructions, but haven't tried setting up an OpenAPS instance yet.
Scott Leibrand
@scottleibrand
Aug 21 2016 17:21
@bfaloona run it manually after a tzdata change. And if you need the Pi to work offline after reboots, you can run it then, or upon seeing the "BG data too old" message from oref0. The latter is what I do.
@lukasz-kaniowski check out https://gitter.im/pazaan/decoding-contour-next-link for 640G CNL chat.
garykidd
@garykidd
Aug 21 2016 17:42
hmmm.... I'm clearly headed in the wrong direction here and now I can ssh....
iMac:~ g$ ssh root@192.168.1.172
ssh: connect to host 192.168.1.172 port 22: Operation timed out
Hichame Yessou
@hichameyessou
Aug 21 2016 17:47
Has anyone ever tried looping with a Pi 0?
Lukasz Kaniowski
@lukasz-kaniowski
Aug 21 2016 18:01
@scottleibrand thanks for countour link
garykidd
@garykidd
Aug 21 2016 18:13
any ideas to try to get ssh to work?
iMac:~ g$ ssh root@192.168.1.172
ssh: connect to host 192.168.1.172 port 22: Operation timed out
Martin Haeberli
@mhaeberli
Aug 21 2016 18:17
@garykidd first see if you can pint 192.168.1.172
ping
ping 192.168.1.172
if that works, then ssh is worth a try; if not, something else is going on in your network so that the devices can’t reach each other
garykidd
@garykidd
Aug 21 2016 18:18
hmmmm...
iMac:~ g$ ping 192.168.1.172
PING 192.168.1.172 (192.168.1.172): 56 data bytes
Request timeout for icmp_seq 0
Request timeout for icmp_seq 1
Request timeout for icmp_seq 2
Request timeout for icmp_seq 3
Request timeout for icmp_seq 4
Request timeout for icmp_seq 5
Request timeout for icmp_seq 6
Request timeout for icmp_seq 7
Request timeout for icmp_seq 8
Request timeout for icmp_seq 9
Request timeout for icmp_seq 10
Request timeout for icmp_seq 11
Request timeout for icmp_seq 12
Request timeout for icmp_seq 13
Request timeout for icmp_seq 14
Request timeout for icmp_seq 15
Request timeout for icmp_seq 16
Request timeout for icmp_seq 17
Request timeout for icmp_seq 18
Request timeout for icmp_seq 19
Request timeout for icmp_seq 20
Request timeout for icmp_seq 21
Request timeout for icmp_seq 22
Request timeout for icmp_seq 23
Request timeout for icmp_seq 24
Request timeout for icmp_seq 25
^C
Martin Haeberli
@mhaeberli
Aug 21 2016 18:19
so ping failes
fails
missing context - you’re trying to ssh into Rpi or Edison?
and assuming that if you ping www.apple.com from mac that succeeds
garykidd
@garykidd
Aug 21 2016 18:20
yup.
Martin Haeberli
@mhaeberli
Aug 21 2016 18:20
if Rpi, and if you haven’t disabled the display, try on the rpi proper to a) ping www.apple.com; b) ping the router ip address for your wifi
rpi ?
edison?
c) ping the mac from the rpi
ok - so Edison. how do you know the Edison is a) on your wifi; b at that IP address . Re-check with Fing, for example, if you can
Hichame Yessou
@hichameyessou
Aug 21 2016 18:23
@garykidd From your router you should be able to find your Raspberry IP address. Otherwise you can plug it with a display, keyboard and mouse and configure the network from the graphic interface
garykidd
@garykidd
Aug 21 2016 18:24
edison
the ip address of edison is the 192.168.1.172
which ping failed on
I tried to run screen by using ``screen /dev/cu.usbserial-A502OS83 -L```
Martin Haeberli
@mhaeberli
Aug 21 2016 18:26
@garykidd sorry to ask, but how do you know for sure that is the IP address. If you run Fing app, you should we hope find its IP address
good to reconfirm… I don’t have enough experiences with Edison, but in general, sometimes these addresses can change on you when you’re not looking ...
garykidd
@garykidd
Aug 21 2016 18:26
I just used router utility for linksys and its indicating that the ip address.
Martin Haeberli
@mhaeberli
Aug 21 2016 18:27
what linksys make and model exactly - I suggest you look that make and model up on the internet - there is a remote chance that it is blocking SSH and / or ping traffic unless otherwise configured
and try Fing on an iPhone for an independent view ...
garykidd
@garykidd
Aug 21 2016 18:30
blob
also indicates IP in the edison utility
I had been able to ssh into it yesterday though--no idea what happened
I reflashed latest firmware about 5x now too. so elementary but stuck. ahhhhh
Martin Haeberli
@mhaeberli
Aug 21 2016 18:32
maybe ping 192.168.1.1 (might be router address) from mac; if you can connect by wire to Edison, try same ping to router; confirm each can reach it. Also, if it worked yesterday, a reboot of each might be in order
what happens if you press “connect wifi” at the bottom right -
garykidd
@garykidd
Aug 21 2016 18:34
PING 192.168.1.1 (192.168.1.1): 56 data bytes
64 bytes from 192.168.1.1: icmp_seq=0 ttl=64 time=0.977 ms
64 bytes from 192.168.1.1: icmp_seq=1 ttl=64 time=1.300 ms
64 bytes from 192.168.1.1: icmp_seq=2 ttl=64 time=1.534 ms
64 bytes from 192.168.1.1: icmp_seq=3 ttl=64 time=1.444 ms
64 bytes from 192.168.1.1: icmp_seq=4 ttl=64 time=0.979 ms
64 bytes from 192.168.1.1: icmp_seq=5 ttl=64 time=0.804 ms
that seems to work
I had had usb serial connection too--but I just disconnected to try ssh ing without it
OK...getting closer agian.
again
iMac:~ g$ ssh root@192.168.1.172
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@ WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED! @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
IT IS POSSIBLE THAT SOMEONE IS DOING SOMETHING NASTY!
Someone could be eavesdropping on you right now (man-in-the-middle attack)!
It is also possible that a host key has just been changed.
The fingerprint for the ECDSA key sent by the remote host is
SHA256:zCA0JVry3OMgjy/yM2ne7Paq6X3n0LzdussvLNXPAks.
Please contact your system administrator.
Add correct host key in /Users/g/.ssh/known_hosts to get rid of this message.
Offending ECDSA key in /Users/g/.ssh/known_hosts:6
ECDSA host key for 192.168.1.172 has changed and you have requested strict checking.
Host key verification failed.
must've been the usb serial connection at the same time...does that make sense?
not sure why screen failed though using screen /dev/cu.usbserial-A502OS83 -L
garykidd
@garykidd
Aug 21 2016 18:41
still stuck.
Martin Haeberli
@mhaeberli
Aug 21 2016 18:53
one thing you can do is to rm /Users/g/.ssh/known_hosts
then it should just ask you to confirm that you’re adding an ssh key
unless you have other important stuff in known_hosts, that approach has worked well for me in the past
(rm just ‘removes’ that file)
garykidd
@garykidd
Aug 21 2016 18:54
iMac:~ g$ rm /Users/g/.ssh/known_hosts
iMac:~ g$ ssh root@192.168.1.172
The authenticity of host '192.168.1.172 (192.168.1.172)' can't be established.
ECDSA key fingerprint is SHA256:zCA0JVry3OMgjy/yM2ne7Paq6X3n0LzdussvLNXPAks.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added '192.168.1.172' (ECDSA) to the list of known hosts.
root@192.168.1.172's password:
root@APS:~#
BAMM!!
Martin Haeberli
@mhaeberli
Aug 21 2016 18:55
:+1:
yeah that other message is confusing and scary
Eric Luhrs
@eluhrs
Aug 21 2016 18:56
I think you also have to specify the speed of the serial connection with screen: screen /dev/cu.usbserial-A502OS83 115200
Martin Haeberli
@mhaeberli
Aug 21 2016 18:56
there is a lot fancier stuff you can do (use nano to delete the old / wrong key entry in known_hosts, for example), but why use a scalpel when a hammer does the trick here. the ssh known_hosts stuff, imho, is a bit complicated for people early in the linux learning curve
garykidd
@garykidd
Aug 21 2016 18:56
Thanks!
Martin Haeberli
@mhaeberli
Aug 21 2016 18:56
:+1:
Brandon Faloona
@bfaloona
Aug 21 2016 19:48
@hichameyessou We are using a Pi0 + TI Stick rig that I just built.
Hichame Yessou
@hichameyessou
Aug 21 2016 19:51
What about the battery? Do you use a small powerbank or a lipo? @bfaloona
garykidd
@garykidd
Aug 21 2016 19:55
nano has so many variables...confusing
garykidd
@garykidd
Aug 21 2016 20:06
root@APS:~# opkg update
Downloading http://repo.opkg.net/edison/repo/all/Packages.gz.
Inflating http://repo.opkg.net/edison/repo/all/Packages.gz.
Updated list of available packages in /var/lib/opkg/all.
Downloading http://repo.opkg.net/edison/repo/edison/Packages.gz.
Inflating http://repo.opkg.net/edison/repo/edison/Packages.gz.
Updated list of available packages in /var/lib/opkg/edison.
Downloading http://repo.opkg.net/edison/repo/core2-32/Packages.gz.
Inflating http://repo.opkg.net/edison/repo/core2-32/Packages.gz.
Updated list of available packages in /var/lib/opkg/core2-32.
Downloading http://iotdk.intel.com/repos/2.0/iotdk/x86/Packages.
Updated list of available packages in /var/lib/opkg/x86.
Downloading http://iotdk.intel.com/repos/2.0/iotdk/i586/Packages.
Updated list of available packages in /var/lib/opkg/i586.
Downloading http://iotdk.intel.com/repos/3.5/intelgalactic/opkg/i586//Packages.
Updated list of available packages in /var/lib/opkg/iotkit.
Downloading http://iotdk.intel.com/repos/3.5/iotdk/edison/all/Packages.
Updated list of available packages in /var/lib/opkg/iotdk-all.
Downloading http://iotdk.intel.com/repos/3.5/iotdk/edison/core2-32/Packages.
Updated list of available packages in /var/lib/opkg/iotdk-core2-32.
Downloading http://iotdk.intel.com/repos/3.5/iotdk/edison/edison/Packages.
Updated list of available packages in /var/lib/opkg/iotdk-edison.
Collected errors:
is "Collected errors:" something that needs to be addressed?
Brandon Faloona
@bfaloona
Aug 21 2016 20:09
@hichameyessou Our rig has a large 10k mAh Zendure battery. Heavy and slow to charge but it supports passthrough charging so it solved all the critical use cases.
Our next rig (Edison) will get a little battery.
The first page above links to Reports just fine. The second page creates a bad link (see 'pump.ini' link text)
The good generated ends in .html, the bad link ends in .md.
Brandon Faloona
@bfaloona
Aug 21 2016 20:29

However the source looks the same on the two pages:

[Reports](core/reports.md) -- works

[pump.ini](openaps-device-pump.md) -- does not work

Eric Jensen
@elnjensen
Aug 21 2016 20:30
Do you have spaces between the square brackets and text in the second case?
Not sure if that matters, but looks different.
Brandon Faloona
@bfaloona
Aug 21 2016 20:31
The broken link source contains backticks in the display text
Eric Jensen
@elnjensen
Aug 21 2016 20:32
Ah, I see. So that's a difference, then, but one that would be nice to work around, i.e. it would be nice if you could still link to backtick-formatted text...
Brandon Faloona
@bfaloona
Aug 21 2016 20:32
That's one theory. The other theory is that the link in source needs path information.
for example [text](../target.md)
I thought I'd ask here. Otherwise I'll just make a PR to dev and we'll find out!
Eric Jensen
@elnjensen
Aug 21 2016 20:35
Markdown spec shows this example for relative links:
If you’re referring to a local resource on the same server, you can use relative paths:

See my [About](/about/) page for details.
So that suggests that maybe you do need a path spec.
Brandon Faloona
@bfaloona
Aug 21 2016 20:36
Awesome! Thanks.
Based on their 'relative' example, would you suggest /page.md or ./page.md for a resource in the same directory?
Eric Jensen
@elnjensen
Aug 21 2016 20:41
Yea, that's not very clear.
The leading slash looks weird to me (from its unix meaning).
Almost looks like you need slashes on both ends? But maybe it really is at the doc root in their example. So probably I'd try either ./page.md or page.md/. But I don't really know.
Are there other working examples in the docs?
Ben West
@bewest
Aug 21 2016 20:55
it's the same as html links
if they're all relative they'll work on both sphinx and github views
the sphinx and github views use different base urls
so /about/ will go to base href /about/ you usually want to use ./about.md or similar ../../phase-4/other-step.md
Eric Jensen
@elnjensen
Aug 21 2016 20:57
Thx, Ben. The initial issue was that it wasn't working when just giving a bare filename with no path spec.
Ben West
@bewest
Aug 21 2016 20:57
ah, I see, yeah, I'd try relative to "here": ./sibling.md
Eric Jensen
@elnjensen
Aug 21 2016 20:57
Just about.md would work in HTML (file in same directory) but doesn't seem to work here.
Sounds good.
Ben West
@bewest
Aug 21 2016 21:05
that's true, without any path at all, sphinx will try to look in any of it's include directories
using the dot path relative to here tells sphinx exactly where to look
it's a subtle side-effect of using the markdown plugin for sphinx I think
Eric
@ecc1
Aug 21 2016 22:17
You can run markdown on your local machine and point your browser at the resulting file:///....html file. That's what I do to check READMEs before pushing to github.