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

4th
May 2017
garykidd
@garykidd
May 04 2017 00:04
hmmm...noticing some differences between MDT sensor reported on pump and BG reflected in NS...not huge...is that normal?
john36m
@john36m
May 04 2017 00:18
@JELCRAWFORD Thanks for the info. I will be in Montreal in Quebec Province and I will report back my success or lack thereof in a week or two.
PopcornPam
@PopcornPam_twitter
May 04 2017 00:26
Curious, with all the hackers here, has anyone been able to hack a medtronic pump remote? 150$ is steep, but bolusing when it is hidden under a dress is a pain
I know the riley link does that, it is possible to do that without a riley link?
Dana Lewis
@danamlewis
May 04 2017 00:31
@PopcornPam_twitter do you have the easy button set to good intervals?
PopcornPam
@PopcornPam_twitter
May 04 2017 00:31
I can't always access it based on wardrobe choices
Dana Lewis
@danamlewis
May 04 2017 00:32
yea. so, there are commands you can use to bolus - but it ups the risk of you not being able to confirm if the bolus went through (or that you accidentally do it multiple times if you think one didn’t go through but it did...)
PopcornPam
@PopcornPam_twitter
May 04 2017 00:32
Commands that can be issued from nightscout?
Dana Lewis
@danamlewis
May 04 2017 00:33
no - underlying openaps toolkit commands
PopcornPam
@PopcornPam_twitter
May 04 2017 00:33
It is mainly when I want to bolus on the sly, like at a meeting or something like that
Dana Lewis
@danamlewis
May 04 2017 00:33
I think slyly entering carbs & triggering SMB might be safest?
PopcornPam
@PopcornPam_twitter
May 04 2017 00:33
How do you tell the pump to bolus?
Is SMB ready to impliment?
Martin Haeberli
@mhaeberli
May 04 2017 00:35
@danamlewis @PopcornPam_twitter so not the safest thing - but if Pam has a trusted partner, with an ssh / reverse tunnel into the openaps instance, she could text the partner and they could issue the bolus?
PopcornPam
@PopcornPam_twitter
May 04 2017 00:35
I don't understand why it would have to be a different person?
Dana Lewis
@danamlewis
May 04 2017 00:36
@PopcornPam_twitter SMB is in dev for testing right now. So you can turn on dev, and I’d test a) your normal setup for a day and make sure everything still works before b) (if you so choose) enabling SMB and testing those components
Martin Haeberli
@mhaeberli
May 04 2017 00:36
wouldn’t but it sounds like it’s complexity you don’t want to handle when you’re being discreet ...
PopcornPam
@PopcornPam_twitter
May 04 2017 00:36
Well, I am bionic til the end of may, so moot point at the moment regardless
Dana Lewis
@danamlewis
May 04 2017 00:36
heh
how’s it feel to say that? :)
PopcornPam
@PopcornPam_twitter
May 04 2017 00:36
Honestly, the other three trials it was amazing
But I am not finding it to be as successful as my openaps rig
And my skin reacts horribly to glucagon, so I will be glad to be on the insulin only system in a week.
Dana Lewis
@danamlewis
May 04 2017 00:38
:/
PopcornPam
@PopcornPam_twitter
May 04 2017 00:38
So I am revving to get back on the homemade system
But then again, the additional daily tasks due to it being a study are also starting to wear on me
Scott Leibrand
@scottleibrand
May 04 2017 00:48
SMB is now in dev. By the time you're done being bionic and get back in your groove with oref0, you'll probably be about ready to start testing the new oref1 features one by one.
PopcornPam
@PopcornPam_twitter
May 04 2017 00:48
cool
Eric
@ecc1
May 04 2017 01:31
@PieterGit another codebase that parses calibration records: https://github.com/ecc1/dexcom/blob/master/record.go#L173
rsilvers129
@rsilvers129
May 04 2017 01:57
I set up my Nightscout on Azure. I see that the docs are recommending Heroku as being cheaper. What are the odds the free Azure site will reach a limit and shut down? I have not started uploading yet, so it is a good time to change if I should.
Dana Lewis
@danamlewis
May 04 2017 01:58
Good odds. Good idea to switch now.
jmcrawford
@jmcrawford
May 04 2017 02:05
When you switch, you will probably use the same Mlab back-end, so any data that is uploaded will be available whichever front-end you choose
rsilvers129
@rsilvers129
May 04 2017 02:25
@danamlewis Ok. I will.
Dana Lewis
@danamlewis
May 04 2017 02:26
:+1: (NS docs could probably also use some PRs if you feel up to it).
Andy Sharrow
@dramageek
May 04 2017 02:34
@cameronrenwick just seeing your post from last night on case request- That's my all-day rig too. I ordered a pack of these: https://www.amazon.com/gp/product/B007Q6PJO8/ and with a bit of Dremmel work shaved off a few of the brackets. Took one of them in half, fit the Edison/Explorer into it, and slapped it onto the top of the one holding the battery.
Gina
@gclyon
May 04 2017 02:48
@dramageek that reminds me of a cassette tape holder. Might try that out!
Andy Sharrow
@dramageek
May 04 2017 02:53
dmdfreak
@dmdfreak
May 04 2017 03:15
I needed to run the ipconfig bnep0 to get my setting to setup offline looping. It wasn't giving me a number so I thought it was because it was connecting to my wifi so I went into wpa_supplicant.conf and temp added ~ to keep it from connecting and rebooted. Still couldn't get the bnep0 number so I figured I would try to redo oref0 install because I also had a problem with my rig not talking to my pump. So the problem is even after taking out the ~ it won't connect to my wifi and therefore won't let me redo oref0 because it can't connect to GitHub. Help?
RedCyclist1
@RedCyclist1
May 04 2017 03:30
OK, stupid question, I set up a 2nd rig and it's working but the name of the rig didn't get changed. What should I do?
rsilvers129
@rsilvers129
May 04 2017 03:31
I was trying to figure out how I ended up with Azure to begin with. It was because I use the Nightscout docs to setup Nightscout rather than the OpenAPS docs. They suggest Azure here and no way to add to their docs: http://www.nightscout.info/wiki/nightscout-setup-guide
Martin Haeberli
@mhaeberli
May 04 2017 03:52
RTD - what is easy way to configure max temp basal?
dmdfreak
@dmdfreak
May 04 2017 03:52
Anybody that could help I'm desperate.
rsilvers129
@rsilvers129
May 04 2017 03:54
This page http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-1/nightscout-setup.html says for ENABLE "Enter all of the words without commas” but it doesn’t say that the choices are. What things should I enable?
Martin Haeberli
@mhaeberli
May 04 2017 04:01
bridge iob careportal basal cage sage maker openaps pump
for example
rsilvers129
@rsilvers129
May 04 2017 04:03
Thanks. http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-1/nightscout-setup.html# says "We highly recommend NOT using the Nightscout COB pill. We even recommend removing it from your Nightscout ENABLE web app settings as it causes bugs” But it doesn’t say what it is called so I am not sure how to disable it.
This message was deleted
Martin Haeberli
@mhaeberli
May 04 2017 04:23
cob is called cob …
rsilvers129
@rsilvers129
May 04 2017 04:39

@RedCyclist1 Log in as root/edison via serial console.

Type/edit the following:

myedisonhostname=<thehostname-you-want> #Do not type the <>
And then paste the following to rename your Edison accordingly:

echo $myedisonhostname > /etc/hostname
sed -r -i"" "s/localhost( jubilinux)?$/localhost $myedisonhostname/" /etc/hosts

katie disimone
@Kdisimone
May 04 2017 05:03
@dmdfreak have you tried ifup wlan0
dmdfreak
@dmdfreak
May 04 2017 05:10
I have not. honestly i am in the middle of completely redoing everything now. A complete reflash and install. Super bummed. I'm on the re flash part ATM
I've been without the rig all day and got desperate when I couldn't even get it back on wifi
katie disimone
@Kdisimone
May 04 2017 05:12

@rsilvers129 just as a way of overall NS explanation...there are default "plugins" that the site uses that you don't have to specifically turn on. There are other plugins that you specifically have to turn on. The phrase that NS uses for "turn on" is enable. And the way you enable plugins is by adding the name of the particular plugin to the variable line setting called ENABLE in heroku.

So the plugins such as openaps bridge maker iob sage and others are "extras" to the default, simple Nightscout site.

The official NS website has pages that can explain the various plugins and what they do. The plugin called cob is for carbs-on-board displayed on your NS. There's two places that NS can get your cob value. One is by calculating cob by using any carbs it recorded as a "treatment" and the values in your NS profile for carb decay. The second source of cob is directly from a "device upload" such as OpenAPS or Loop. Unfortunately, the fact that cob values can draw from two sources seems to cause problems for the display and heroku's smooth operations. Something needs to be improved on the NS code to fix that (I'd guess) but until that time, better to simply leave cob off the ENABLE line so that the problems aren't started.

rsilvers129
@rsilvers129
May 04 2017 05:13
Ok. I will turn off cob.
katie disimone
@Kdisimone
May 04 2017 05:17

@rsilvers129 also the NS page in the OpenAPS docs are missing the cell borders....I thought I added them PR this morning, but that looks like they still didn't display properly. So maybe some of your confusion is that the cells still are clearly defined enough 😜. I'll get that fixed eventually. If you look here, the cells are a little easier to separate out.

https://github.com/openaps/docs/blob/master/docs/docs/walkthrough/phase-1/nightscout-setup.md

*still are not clearly defined enough
rsilvers129
@rsilvers129
May 04 2017 05:23
I noticed that, but I figured that out. I loved that page. It was easy to follow. A list of all of the optional plugins would be good though to have right there though.
dmdfreak
@dmdfreak
May 04 2017 05:42
@Kdisimone ifup wlan0 returned the following:
root@timsedison:~# ifup wlan0
wpa_supplicant: cannot read contents of /etc/wpa_supplicant.conf
run-parts: /etc/network/if-pre-up.d/wpasupplicant exited with return code 1
Failed to bring up wlan0.
Martin Haeberli
@mhaeberli
May 04 2017 06:57
@dmdfreak suggests that the syntax of the file is wrong...
dmdfreak
@dmdfreak
May 04 2017 07:24
total redo worked. I'm back up and looping :)
now if I can figure out offline looping I'm golden LOL
Matthias
@ceben80
May 04 2017 09:39

@/all linux Q

export NIGHTSCOUT_HOST=https://xxxxxxxxxxxxxx.herokuapp.com
export API_SECRET=xxxxxxxxxxxxxxxxxxxxxxxxx

has been set in .profile

but echo doesn't show a content

root@eddie-eb:~# echo $API_SECRET

root@eddie-eb:~# echo $NIGHTSCOUT

root@eddie-eb:~#
any idea :question:
and yes, I have re-login again so settings in .profile should be effective, but doesn't work :worried:
Marco
@CaptainBalou
May 04 2017 11:46
@ceben80 Welche shell ist für den User gesetzt. Bei Linux ja oftmals bash - die liest standardmäßig .bash_profile bei anderen Distributionen. Habe leider kein Jubilinux wo ich schauen könnte.
@ceben80 Sorry for talking German: What shell is it? bash? Taht shell normally reads .bash_profile in other distributions. Have no Jubilinux were I could look into.
Matthias
@ceben80
May 04 2017 11:52
@CaptainBalou AFAIK bash, but this was working since I have re-flashed to the newest jubilinux version, that's really weird....
Marco
@CaptainBalou
May 04 2017 11:54
@ceben80 Oh ok, so I don't know. Cannot test anything in Jubilinux. To see if it is bash you can use ps. But that doesn't helps us because it worked before...
garykidd
@garykidd
May 04 2017 12:21
Everything ran fine last night but when I left house....rig did connect via BT but now I'm getting Comms error. Cannot get subg rfspy state or version. Hard reset fails. Any quick ideas to check. Difficult to resolve on the go.
dmdfreak
@dmdfreak
May 04 2017 13:05
When turning on SMB exactly what should the"flash" look like? I've tried --microbolus and -microbolus but both say unknown
*"flag"
garykidd
@garykidd
May 04 2017 13:34
can you use -h?
Martin Haeberli
@mhaeberli
May 04 2017 13:43
so running openaps on a test pump (no patient at the end) with real BG data … two problems:
-occasionally even when BG data is live, openaps doesn’t seem to be able to fetch it as current (‘too old’), even when a manual fetch works fine.
then, big mystery, overnight, even loop cron seemed somehow not to run regularly anymore. recent excerpt:
pumphistory-24h refreshed
Completed pump-loop at Thu May 4 05:28:04 PDT 2017

Starting pump-loop at Thu May 4 06:28:05 PDT 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Old pumphistory: Refreshed pumphistory
{"carbs":0,"boluses":0,"mealCOB":0}
{"ratio":0.99}
BG data is too old (it's probably this), or clock set incorrectly.  The last BG data was read at Thu May 04 2017 06:16:58 GMT-0700 (PDT) but your system time currently is Thu May 04 2017 06:29:21 GMT-0700 (PDT)
determine-basal://text/shell/enact/suggested.json
reporting enact/suggested.json
{"reason":"BG data is too old (it's probably this), or clock set incorrectly.  The last BG data was read at Thu May 04 2017 06:16:58 GMT-0700 (PDT) but your system time currently is Thu May 04 2017 06:29:21 GMT-0700 (PDT)"}
{"reason":"BG data is too old (it's probably this), or clock set incorrectly.  The last BG data was read at Thu May 04 2017 06:16:58 GMT-0700 (PDT) but your system time currently is Thu May 04 2017 06:29:21 GMT-0700 (PDT)"}
Waiting for ntpd to synchronize... OK!
{"bg":154,"temp":"absolute","snoozeBG":144,"recieved":true,"predBGs":{"IOB":[154,153,151,150,149,148,147,147,146,146,145,145,145]},"rate":1.25,"reason":"COB: 0, Dev: -10, BGI: 0, ISF: 46, Target: 111; Eventual BG 144 >= 121, max_iob 0, 0m@-0.075 = 0.000 > 2 * req 0. Setting temp basal of 1.25U/hr","COB":0,"eventualBG":144,"timestamp":"2017-05-04T04:23:04.067443","duration":30,"tick":-1,"IOB":0}
Profile less than 60m old
temp_basal.json less than 5m old
Pumphistory less than 5m old
Settings refreshed
pumphistory-24h refreshed
Completed pump-loop at Thu May 4 06:29:56 PDT 2017
note the HOUR between loop attempts. Note that new BG data arrived at 06:16 local
rsilvers129
@rsilvers129
May 04 2017 14:00
@dmdfreak I assume you installed the dev branch?
dmdfreak
@dmdfreak
May 04 2017 14:02
yes. I can see the things to change in preferences.json
Martin Haeberli
@mhaeberli
May 04 2017 14:04
so this is log excerpts near where it seems to have gone a bit off the rails … trying to understand why
May 03 23:02:13 etghopenaps1 pump-loop.log:  Settings less than 10m old
May 03 23:02:13 etghopenaps1 pump-loop.log:  Pumphistory-24 less than 20m old
May 03 23:02:13 etghopenaps1 pump-loop.log:  Completed pump-loop at Wed May 3 23:02:13 PDT 2017
May 03 23:03:02 etghopenaps1 ns-loop.log:  Starting ns-loop at Wed May 3 23:03:02 PDT 2017:
May 03 23:03:10 etghopenaps1 ns-loop.log:  ns://JSON/shell/cgm/ns-glucose.json
May 03 23:03:10 etghopenaps1 ns-loop.log:  reporting cgm/ns-glucose.json
May 03 23:03:21 etghopenaps1 ns-loop.log:  Refreshed temptargets and meal-carbs
May 03 23:03:21 etghopenaps1 ns-loop.log:  {"batteryVoltage":3840, "battery":66}
May 03 23:03:32 etghopenaps1 ns-loop.log:  Uploaded
May 03 23:06:01 etghopenaps1 ns-loop.log:  Starting ns-loop at Wed May 3 23:06:01 PDT 2017:
May 03 23:06:10 etghopenaps1 ns-loop.log:  ns://JSON/shell/cgm/ns-glucose.json
May 03 23:06:10 etghopenaps1 ns-loop.log:  reporting cgm/ns-glucose.json
May 03 23:06:20 etghopenaps1 ns-loop.log:  Refreshed temptargets and meal-carbs
May 03 23:06:20 etghopenaps1 ns-loop.log:  {"batteryVoltage":3840, "battery":66}
May 03 23:06:31 etghopenaps1 ns-loop.log:  Uploaded
May 03 23:25:02 etghopenaps1 network.log:  Starting oref0-online.
May 03 23:25:02 etghopenaps1 network.log:  At Wed May  3 23:25:02 PDT 2017 my local IP is: 10.0.0.209
May 03 23:25:02 etghopenaps1 network.log:  At Wed May  3 23:25:02 PDT 2017, my wifi network name is HOME-60CE-5, and my public IP is: 
May 03 23:25:02 etghopenaps1 network.log:  Error, cycling networking 
May 03 23:25:19 etghopenaps1 network.log:  Stopping networking (via systemctl): networking.service.
May 03 23:25:19 etghopenaps1 pump-loop.log:  Starting pump-loop at Wed May 3 23:25:16 PDT 2017:
May 03 23:25:19 etghopenaps1 network.log:  Starting networking (via systemctl): networking.service.
May 03 23:25:19 etghopenaps1 network.log:  and getting new wlan0 IP
May 03 23:25:20 etghopenaps1 network.log:  At Wed May  3 23:25:20 PDT 2017, my wifi network name is HOME-60CE-5, and my public IP is: 
May 03 23:25:20 etghopenaps1 network.log:  At Wed May  3 23:25:20 PDT 2017 my public IP is: 
May 03 23:25:50 etghopenaps1 pump-loop.log:  Radio ok, Listening: .No pump comms detected from other rigs
May 03 23:25:57 etghopenaps1 network.log:  Error, connecting BT to 90:60:F1:B2:E1:BB, getting bnep0 IPPING 172.20.10.1 (172.20.10.1) 56(84) bytes of data.
May 03 23:25:57 etghopenaps1 network.log:  --- 172.20.10.1 ping statistics ---
May 03 23:25:57 etghopenaps1 network.log:  1 packets transmitted, 0 received, +1 errors, 100% packet loss, time 0ms
May 03 23:25:57 etghopenaps1 network.log:  At Wed May  3 23:25:57 PDT 2017, my wifi network name is HOME-60CE-5, and my public IP is: 2601:0646:c001:be3e:92b6:86ff:fe0c:5239
May 03 23:25:57 etghopenaps1 network.log:  Finished oref0-online.
May 03 23:26:31 etghopenaps1 pump-loop.log:  Old pumphistory: Refreshed pumphistory
May 03 23:26:36 etghopenaps1 pump-loop.log:  {"carbs":0,"boluses":0,"mealCOB":0}
May 03 23:26:36 etghopenaps1 pump-loop.log:  {"ratio":1}
May 03 23:26:36 etghopenaps1 pump-loop.log:  BG data is too old (it's probably this), or clock set incorrectly.  The last BG data was read at Wed May 03 2017 23:01:59 GMT-0700 (PDT) but your system time currently is Wed May 03 2017 23:26:36 GMT-0700 (PDT)
May 03 23:26:37 etghopenaps1 pump-loop.log:  determine-basal://text/shell/enact/suggested.json
May 03 23:26:37 etghopenaps1 pump-loop.log:  reporting enact/suggested.json
May 03 23:26:37 etghopenaps1 pump-loop.log:  {"reason":"BG data is too old (it's probably this), or clock set incorrectly.  The last BG data was read at Wed May 03 2017 23:01:59 GMT-0700 (PDT) but your system time currently is Wed May 03 2017 23:26:36 GMT-0700 (PDT)"}
May 03 23:26:37 etghopenaps1 pump-loop.log:  {"reason":"BG data is too old (it's probably this), or clock set incorrectly.  The last BG data was read at Wed May 03 2017 23:01:59 GMT-0700 (PDT) but your system time currently is Wed May 03 2017
dmdfreak
@dmdfreak
May 04 2017 14:04
I've got another weird issue though. The only thing I've changed this morning is the flags in preferences.json and my rig stopped talking to my pump an hour ago. When I was trying to turn on the smb flag I never went through with an install so nothing changed there. Before yesterday, I've never had a problem with it talking to the pump
Scott Leibrand
@scottleibrand
May 04 2017 14:16
@mhaeberli looks like your rig had local connectivity but couldn't reach the Internet.
Martin Haeberli
@mhaeberli
May 04 2017 14:16
@scottleibrand thx
Scott Leibrand
@scottleibrand
May 04 2017 14:16
@dmdfreak you might want to switch back to master until you have a second rig to play with...
dmdfreak
@dmdfreak
May 04 2017 14:17
probably a good idea @scottleibrand
Scott Leibrand
@scottleibrand
May 04 2017 14:18
Looks like one of your input files is messed up so it can't calculate IOB.
Martin Haeberli
@mhaeberli
May 04 2017 14:18
@scottleibrand but net connectivity is up now, yes BG data through nightscout is 15 minutes old, but OpenAPS pill showing / still showing “unknown"
papertrail logs are now live
dmdfreak
@dmdfreak
May 04 2017 14:19
So weird. I did a complete reflash and install last night and all was good all night
Scott Leibrand
@scottleibrand
May 04 2017 14:19
@dmdfreak easiest fix, since you just did it, might be to re-run oref0-setup.
Martin Haeberli
@mhaeberli
May 04 2017 14:19
(i wasn’t watching them live at midnight)
dmdfreak
@dmdfreak
May 04 2017 14:19
ok I'll do that first
Scott Leibrand
@scottleibrand
May 04 2017 14:19
@mhaeberli I wouldn't worry about OpenAPS pill until you get current BGs.
Martin Haeberli
@mhaeberli
May 04 2017 14:19
k
dmdfreak
@dmdfreak
May 04 2017 14:20
@scottleibrand what is the flag to add for SMB?
Scott Leibrand
@scottleibrand
May 04 2017 14:28
Take a look at the oref0-setup source code. We've purposefully not documented that yet to make sure that people understand what they're doing before enabling it.
dmdfreak
@dmdfreak
May 04 2017 14:35
ok. I'll wait a little while if I get this up and running before I take a run at it then.
also, re-run fixed the issue. Thanks!
garykidd
@garykidd
May 04 2017 14:38
@scottleibrand any thing you might suggest troublshooting with continual COuld not get subg_rfspy state or version errors?
Scott Leibrand
@scottleibrand
May 04 2017 14:49
What are the two LEDs next to the microUSBs doing?
They should flash briefly when it mmtunes and then stay off.
You can try reseating the Edison on the Explorer board.
garykidd
@garykidd
May 04 2017 14:52
the two really small LEDs next to the UART are remaining lighted.
the larger LED between the two microUSB ports is also lighted and remains so...
Scott Leibrand
@scottleibrand
May 04 2017 14:57
If those two LEDs are on, nothing will work. Try a reseat, and if that doesn't fix it, use ccprog to reflash the radio chip firmware per the docs. Also, please PR the docs troubleshooting section to better cover your situation: it's not an uncommon one.
You should be able to search the docs and find a troubleshooting section for your issue. If that's not yet true, please PR in a fix to make it true for the next person.
garykidd
@garykidd
May 04 2017 15:00
I didnt see something but will look again....if not there will PR once I figure out exactly how to fix it.
Scott Leibrand
@scottleibrand
May 04 2017 15:01
There is a section for this already that details how to ccprog. It may need more info on what situations to try that in, so people like you can find it.
(People in your situation I mean.)
garykidd
@garykidd
May 04 2017 15:11
I will closely review and see if I can augment accordingly...thanks....
can you confirm that the correct port should be.... /dev/spidev5.1 ?
Scott Leibrand
@scottleibrand
May 04 2017 15:22
For Explorer board, correct.
RedCyclist1
@RedCyclist1
May 04 2017 15:25
@Kdisimone So I took COB off the ENABLE line in the var configs in Heroku. The COB pill is now not showing on my NS site. Is that correct?
garykidd
@garykidd
May 04 2017 15:26
@scottleibrand thx
katie disimone
@Kdisimone
May 04 2017 15:27
@RedCyclist1 yes. That's how it should look. Use your OpenAPS pill in order to see cob when you need it
RedCyclist1
@RedCyclist1
May 04 2017 15:29
@Kdisimone Oh OK, Thanks!
garykidd
@garykidd
May 04 2017 15:31
so I'm in my openaps directory and cant run 'oref0-runagain' Am I not in the correct directory?
garykidd
@garykidd
May 04 2017 15:44
wget https://github.com/EnhancedRadioDevices/subg_rfspy/releases/download/v0.8-explorer/spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
worksed fine.
ccprog -p 19,7,36 erase.....no such luck...different directory?
got -bash: ccprog: command not found...
Rob Kresha
@rkresha
May 04 2017 16:03
@scottleibrand @danamlewis I've seen the pictures from Dana's Urchin watchface, can someone clue me in how you are populating the status lines . I'm playing around with the options that are delivered with it and can't get anywhere close. Thanks in advance!
garykidd
@garykidd
May 04 2017 16:07
aggghhhh
rsilvers129
@rsilvers129
May 04 2017 16:13
Do an ls and see if you see it. Make sure it has execute permissions. Do a chmod 755 on it if needed. If current directory is not in your path you could put a ./ before it.
garykidd
@garykidd
May 04 2017 16:14
would using sudo circumvent any of this potenitally?
Eric
@ecc1
May 04 2017 16:16
No, sudo uses the same PATH as the user. You're not getting permission denied (which sudo might address), you're getting program not found.
jmcrawford
@jmcrawford
May 04 2017 16:16
I think the docs say to execute it using "bash oref0-runagain"
@ceben80 A couple of things: to find out what shell you're in, do
echo $SHELL
garykidd
@garykidd
May 04 2017 16:17
Thanks @ecc1 what's making this difficult is trying toi type this on an iphone too.
jmcrawford
@jmcrawford
May 04 2017 16:17
You could also try to load that .profile manually:
. ~/.profile
echo $API_SECRET
echo $NIGHTSCOUT_HOST
jmcrawford
@jmcrawford
May 04 2017 16:22
It likely is as @CaptainBalou suggested, the system is loading .bash_profile instead.
garykidd
@garykidd
May 04 2017 16:28
sorry guys I know thi sis simple---I know I'm probably not i nth eright directory but can someone walk me through these 4 lines and smack me awake here...please
Matthias
@ceben80
May 04 2017 16:30
@jmcrawford did this already, no output
Savek-CC
@savek-cc
May 04 2017 16:31
@ceben80 does a single set produce any output?
Matthias
@ceben80
May 04 2017 16:32
Yes
Savek-CC
@savek-cc
May 04 2017 16:35
set|grep API aber nicht?
jmcrawford
@jmcrawford
May 04 2017 16:35
When you logged in, did you log in as edison, then become root, or did you log in as root directly?
garykidd
@garykidd
May 04 2017 16:38
@jmcrawford was tha directed at me...root
jmcrawford
@jmcrawford
May 04 2017 16:39
@ceben80 If you are edison and run "su root" it doesn't load .profile. If you run "su - root", it does load .profile.
@garykidd No, was a different thread
@garykidd Try this command:
locate ccprog
Matthias
@ceben80
May 04 2017 16:39
@jmcrawford no root, Explorer Board :wink:
jmcrawford
@jmcrawford
May 04 2017 16:40
It'll tell you if the program is installed. It's not on my rig, so it must not be installed by default.
garykidd
@garykidd
May 04 2017 16:40
hmmm didnt see any overt reponse other than another command line after that---I swear I saw the file earlier though...
Install ccprog tools on your Edison: cd ~/src; git clone https://github.com/ps2/ccprog.git
If you saw it, it would be under ~/src somewhere
garykidd
@garykidd
May 04 2017 16:42
yup did that...but f'it...I'll do it all again.
now it says ccprog directory exists...I imagine thats OK...right
I do see ccprog under ~/src
ccprog appears ot be a subdirectory...shall I enter ~/src and then attempt to run the ccprog erase and replace commands...
Rob Kresha
@rkresha
May 04 2017 16:51
if you are in directory do ./ccprog ............. I'm guessing your PATH variable needs to be updated with the location of ccprog
but the above should get you by in the meantime
jmcrawford
@jmcrawford
May 04 2017 16:52
If you've just downloaded the ccprog, it looks like it needs to be compiled before it'll run: https://github.com/ps2/ccprog
Matthias
@ceben80
May 04 2017 16:53
@jmcrawford forget the .profile stuff, had an issue with xdrip as cgm source and found a bug in oref0-setup.sh
FYI @scottleibrand - @colinlennon is going to do an issue in dev (or PR)
garykidd
@garykidd
May 04 2017 16:53
@rkresha when you say in directory ...do you mean in ~/src or ~/src/ccprog?
jmcrawford
@jmcrawford
May 04 2017 16:53
~/src/ccprog
garykidd
@garykidd
May 04 2017 16:53
ok; stanby
Rob Kresha
@rkresha
May 04 2017 16:54
yeah or give it the full path like @jmcrawford is saying ~/src/ccprog/ccprog <blah>
garykidd
@garykidd
May 04 2017 16:55
1st way = NG
Rob Kresha
@rkresha
May 04 2017 16:59
what is it saying
can you paste some context?
garykidd
@garykidd
May 04 2017 16:59
No such file but I'm trying it again
jmcrawford
@jmcrawford
May 04 2017 17:00
It needs to be compiled. try this:
cd ~/src/ccprog && make
garykidd
@garykidd
May 04 2017 17:09
that did seem to do somethin gbut re-running ccprog still getting me nothing WTF....command not found
Matthias
@ceben80
May 04 2017 17:10
@jmcrawford thanks for investigation :wink:
jmcrawford
@jmcrawford
May 04 2017 17:20
@garykidd not sure what you're seeing - can you show us the output from "ls -l ~/src/ccprog/"
garykidd
@garykidd
May 04 2017 17:42
defintely just back from a quick bite to eat---going low...ooops...
ip address changed just got back into rig...
: Thu May  4 12:38:23 2017 from gkphone.local root@white_dove_host:~# ls -l ~/src/ccprog/ total 56 -rwxr-xr-x 1 root root 16344 May  4 13:02 ccprog
-rw-r--r-- 1 root root 24177 May  4 11:39 ccprog.c
-rw-r--r-- 1 root root    68 May  4 11:39 Makefile
-rw-r--r-- 1 root root  3919 May  4 11:39 readhex.c
-rw-r--r-- 1 root root  1556 May  4 11:39 readhex.h
-rw-r--r-- 1 root root   795 May  4 11:39 README.md
root@white_dove_host:~#
jmcrawford
@jmcrawford
May 04 2017 17:50
So the program compiled ok, and has the executable bit set. Does it work if you fully path it:
~/src/ccprog/ccprog
garykidd
@garykidd
May 04 2017 17:53
would that mean ~src/ccprog/ccprog -p 19,7,36 erase
jmcrawford
@jmcrawford
May 04 2017 17:54
Need another slash in there:
~/src/ccprog/ccprog -p 19,7,36 erase
garykidd
@garykidd
May 04 2017 17:54
right...brb
Dana Lewis
@danamlewis
May 04 2017 17:55
@rkresha I've piped json status content to a web page, and so the json from the web URL is what's displaying (I just give urchin the webpage)
Martin Haeberli
@mhaeberli
May 04 2017 17:57
@garykidd you’re right on target
I have been planning a PR having gotten lost twice doing this in the last week or two
sorry I haven’t done it yet
indeed, you have to make ccprog after cloning it from github
and then the path to ccprog has to be correct or you can’t invoke it
@scottleibrand thx for commenting on logs this morning - what is the obvious thing to look for to confirm there isn’t / wasn’t internet connectivity? Also, why would my system wait for an hour between attempts to start ns-loop? (same answer? )
garykidd
@garykidd
May 04 2017 18:00
woot!!
Martin Haeberli
@mhaeberli
May 04 2017 18:00
@garykidd I would have chimed in earlier but was offline in the interim
garykidd
@garykidd
May 04 2017 18:00
chip erased
Martin Haeberli
@mhaeberli
May 04 2017 18:01
yes, great next step!
garykidd
@garykidd
May 04 2017 18:01
on to last step...so difficult with the BT connection and working off an iphone
Martin Haeberli
@mhaeberli
May 04 2017 18:01
@garykidd fwiw, if you have a mac handy you can also BT tether it to the same iPhone and then ssh from the mac to the edison. in theory that should work on a windows machine, too, but...
garykidd
@garykidd
May 04 2017 18:02
gotta try it ---I do have a surface Pro 4 handy
Martin Haeberli
@mhaeberli
May 04 2017 18:03
no direct experience with that but should work, too, if you have putty or some other ssh client on the surface
garykidd
@garykidd
May 04 2017 18:03
nothing set up yet but gotta work that
Martin Haeberli
@mhaeberli
May 04 2017 18:06
:+1:
Martin Haeberli
@mhaeberli
May 04 2017 18:15
[94523.349665] Bluetooth: hci0 command 0x0c14 tx timeout
[94525.349388] Bluetooth: hci0 command 0x0c14 tx timeout
[94527.349100] Bluetooth: hci0 command 0x0c14 tx timeout
[94529.348936] Bluetooth: hci0 command 0x0c13 tx timeout
more generally, auto-roving between Bluetooth hotspot on iPhone and a work and home wifi are not so reliable. time for more RTD
ejoe132
@ejoe132
May 04 2017 18:32
Im having a problem with my edison out of the blue stop communicating or when switching from wifi to hotspot itll connect to the hotspot but then not communicate. Is there a watchdog like software that can be installed to fix this as currently I shut it off then turn it back on and it will normally start again
Scott Leibrand
@scottleibrand
May 04 2017 18:34
@mhaeberli I just looked at oref0-online, and saw that it failed to get your public IP, which means no Internet connectivity (to the server it uses)
garykidd
@garykidd
May 04 2017 18:36
root@white_dove_host:~# ~/src/ccprog/ccprog -p 19,7,36 erase Using pins: DC=19, DD=7, RESET=36 Erasing chip.
root@white_dove_host:~# ~/src/ccprog/ccprog root@white_dove_host:~# ~/src/ccprog/ccprog -p 19,7,36 write spil_alt2_EDISON_EXPLORER_US_STDLOC.hex
Using pins: DC=19, DD=7, RESET=36
Intel Hex file: spil_alt2_EDISON_EXPLORER_US_STDLOC.hex
Unable to open file "spil_alt2_EDISON_EXPLORER_US_STDLOC.hex" for reading.
root@white_dove_host:~#
Scott Leibrand
@scottleibrand
May 04 2017 18:42
you might want to do a quick linux tutorial to figure out how to reference relative vs. absolute paths
you need to specify either the full path to the hex file, or the relative path from your current working directory
garykidd
@garykidd
May 04 2017 18:43
hmmm....within the path it didnt work the absolute path worked and only th absolute path for some reason...I'm sure I did something wrong but stange it worked for the first rease command but not the program command...no?
Scott Leibrand
@scottleibrand
May 04 2017 18:44
you fixed the path to ccprog
now you have to do the same for the .hex file
the erase command didn't require an input filename
garykidd
@garykidd
May 04 2017 18:44
copy that
defintely think a PR is in order for this section...
Rob Kresha
@rkresha
May 04 2017 18:52
@danamlewis Can you talk more specifically on what and how you are piping... I'm struggling here :(. I'd like to do more playing but I hit dead ends like api-docs.html
garykidd
@garykidd
May 04 2017 19:20
~/src/ccprog/ccprog root@white_dove_host:~# ~/src/ccprog/ccprog -p 19,7,36 write spil_alt2_EDISON_EXPLORER_US_STDLOC.hex
Using pins: DC=19, DD=7, RESET=36
Intel Hex file: spil_alt2_EDISON_EXPLORER_US_STDLOC.hex
Unable to open file "spil_alt2_EDISON_EXPLORER_US_STDLOC.hex" for reading.
root@white_dove_host:~#
~/src/ccprog/ccprog -p 19,7,36 write ~/src/spil_alt2_EDISON_EXPLORER_US_STDLOC.hex
Martin Haeberli
@mhaeberli
May 04 2017 19:29
@scottleibrand thx - more reading!
garykidd
@garykidd
May 04 2017 20:23
does this look right?
~/src/ccprog/ccprog -p 19,7,36 write ~/src/spil_alt2_EDISON_EXPLORER_US_STDLOC.hex
Martin Haeberli
@mhaeberli
May 04 2017 20:30
@garykidd looks right but fyi I did it as two separate commands; I have no direct knowledge of whether combining works.
also it is reassuring to see the erased confirmation before flashing
garykidd
@garykidd
May 04 2017 20:31
hmmm...I am attempting to do it as 2 different commands as well.....first erase...then write........same for you right?
Martin Haeberli
@mhaeberli
May 04 2017 20:33
@garykidd indeed does look right - as the second command - my cockpit error on not seeing it correctly
you’re good!
garykidd
@garykidd
May 04 2017 20:47
```
```
root@white_dove_host:~/src# ~/src/ccprog/ccprog -p 19,7,36 write ~/src/spil_alt2_EDISON_EXPLORER_US_STDLOC.hex
Using pins: DC=19, DD=7, RESET=36
Intel Hex file: /root/src/spil_alt2_EDISON_EXPLORER_US_STDLOC.hex
Unable to open file "/root/src/spil_alt2_EDISON_EXPLORER_US_STDLOC.hex" for reading.
root@white_dove_host:~/src#
I know I 'm not referencing something correctly but cant see it.
Martin Haeberli
@mhaeberli
May 04 2017 20:48
try ls -1
also try updatedb (to update the locate database) then locate spil_alt2_EDISON_EXPLORER_US_STDLOC.hex
to find where the file is
garykidd
@garykidd
May 04 2017 20:53
root@white_dove_host:~/src# locate spil_alt2_EDISON_EXPLORER_US_STDLOC.hex
root@white_dove_host:~/src#
root@white_dove_host:~# ls-1
-bash: ls-1: command not found
root@white_dove_host:~# cd src
root@white_dove_host:~/src# ls -1
bluez-5.37
bluez-5.37.tar.gz
ccprog
EdisonVoltage
mraa
oref0
spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex.1
spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex.2
root@white_dove_host:~/src# updatedb
Martin Haeberli
@mhaeberli
May 04 2017 20:55
@garykidd - try:
wget https://github.com/EnhancedRadioDevices/subg_rfspy/releases/download/v0.8-explorer/spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
ok so you have the hex file;
garykidd
@garykidd
May 04 2017 20:55
looks like I have 3...
Martin Haeberli
@mhaeberli
May 04 2017 20:55
try copying and pasting the name
it is spi 1 (one) not spi l (letter L)
you can even autocomplete
like
ccprog/ccprog (parameters) spi1<tab>
garykidd
@garykidd
May 04 2017 20:56
copy that.....it's my set up today...PITA
Martin Haeberli
@mhaeberli
May 04 2017 20:57
where parameters are as you showed earlier
ccprog/ccprog -p 19,7,36 write spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
might work
garykidd
@garykidd
May 04 2017 21:00
OMFG YES..thx
Martin Haeberli
@mhaeberli
May 04 2017 21:00
glad it worked out
I’m still mystified by a bunch of stuff from the level of quarks to life, the universe, and everything, even in the microcosm of openaps
garykidd
@garykidd
May 04 2017 21:02
it was the 1 vs. L F'in Ehh Man!!!
Martin Haeberli
@mhaeberli
May 04 2017 21:02
:+1:
garykidd
@garykidd
May 04 2017 21:02
Just when you think you're getting more confident....life tells you you'er a 1L.
Martin Haeberli
@mhaeberli
May 04 2017 21:02
IMHO, serial numbers and other things should BAN the use of visual homophones - I l 1 i ; O D 0, etc
especially as my eyes get worse and the print keeps on getting smaller on the objects I own as they / I age
garykidd
@garykidd
May 04 2017 21:03
amen brotha
amen
htanks to everyone for helpin gthis knucklehead
And just so you know...looping now.......
Martin Haeberli
@mhaeberli
May 04 2017 21:04
:+!:
garykidd
@garykidd
May 04 2017 21:05
@scottleibrand I'm imagining there could be a way to automoate this resolution....almost like if mmtune fails a few times...and a forced restart doesnt rectify....implement these 4 lines of code..... below...
thoughts?
JELCRAWFORD
@JELCRAWFORD
May 04 2017 21:09
A
garykidd
@garykidd
May 04 2017 21:16
oref0-runagain
cd ~/src; git clone https://github.com/ps2/ccprog.git
wget https://github.com/EnhancedRadioDevices/subg_rfspy/releases/download/v0.8-explorer/spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
ccprog -p 19,7,36 erase
ccprog -p 19,7,36 write spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
garykidd
@garykidd
May 04 2017 21:23
Not sure what caused that state in the first place...2 things I can see that need a PR...
  1. I think ccprog may need to be compiled in order to run; and
  2. I thought I had followed from the correct directories---but maybe if we had absolute references in each of the above commands so they could be administered/ or reference the wrong/invalid directories.
    If you think this makes sense to PR, I'd be happy to.
Scott Leibrand
@scottleibrand
May 04 2017 21:31
Yes please. make compiles it.
Frank Best
@efbest
May 04 2017 21:36

Trying to upgrade oref0 to oref1 I get the following error:

root@edi2:~/src/oref0# git pull remote: Counting objects: 44, done. remote: Compressing objects: 100% (25/25), done. remote: Total 44 (delta 25), reused 13 (delta 13), pack-reused 6 Unpacking objects: 100% (44/44), done. From git://github.com/openaps/oref0 0fc8151..143879f dev -> origin/dev 3858b43..1962a56 pushover -> origin/pushover c5d2edf..b9a547f uam-cob -> origin/uam-cob Updating 2c512ee..143879f error: Your local changes to the following files would be overwritten by merge: bin/oref0_init_pump_comms.py bin/oref0_subg_ww_radio_parameters.py Please, commit your changes or stash them before you can merge. Aborting

what does it mean???

Scott Leibrand
@scottleibrand
May 04 2017 21:49
Do what it suggests and git stash from ~/src/oref0 first.
Maybe also add a note to that effect in the PR?
fjungman
@fjungman
May 04 2017 22:28
@scottleibrand or @danamlewis Just to be certain, since my kid doesn't bolus, I should NOT upgrade to oref1, right? He is supposed to bolus, but he suffers from being 16 years old.
Frank Best
@efbest
May 04 2017 22:31
@scottleibrand worked fine so far, but:
/root/src/subg_rfspy/ already exists; pulling latest
fatal: unable to access 'https://github.com/ps2/subg_rfspy/': Could not resolve host: github.com
Couldn't pull latest subg_rfspy
fjungman
@fjungman
May 04 2017 22:35
Looks like the device is not connected to the internet, or DNS isn't working.
Frank Best
@efbest
May 04 2017 22:35
👍
Dana Lewis
@danamlewis
May 04 2017 22:43
@fjungman actually, some of the features in oref1 might be helpful. Can't tell you what's right for you, but if you decide to test I would test one thing at a time when you can observe how oref1 behavior differs from oref0. But it's up to you to decide if you want to give it the ability to SMB.
Martin Haeberli
@mhaeberli
May 04 2017 22:55
ok so a dumb suggestion (yes, I need to write it myself, but I definitely won’t get to this anytime soon) - on submarines, they have a green / red board called the “christmas tree” that tells you whether the submarine is clear to dive. - so the idea is - in addition to the “l” shortcut for logs, maybe add a “t” shortcut which shows you an ever-updated multiline status board on what is green, yellow, red - pump data (how recent), BG data (how recent), internet (how recent), … that would help people figure out what to fix so openaps comes back online if it’s not...
ok, also pump battery, openaps battery, reservoir status, sensor age, cannula age, ...
Martin Haeberli
@mhaeberli
May 04 2017 23:01
curl -v https://<xxx>.herokuapp.com/api/v1/entries.json?count=1 --header "api-secret:yyy
returns a reasonable value
but openaps says bgs are 10 hours old
?
i mean, this returns a reasonable value ON the Edison in question...
so what report should i / can i run standalone to test?
{"reason":"BG data is too old (it's probably this), or clock set incorrectly.  The last BG data was read at Thu May 04 2017 06:31:58 GMT-0700 (PDT) but your system time currently is Thu May 04 2017 15:52:53 GMT-0700 (PDT)”}
Scott Leibrand
@scottleibrand
May 04 2017 23:09
which CGM configuration are you using?
fjungman
@fjungman
May 04 2017 23:09
Thanks, @danamlewis
Martin Haeberli
@mhaeberli
May 04 2017 23:09
@scottleibrand feeding from NightScout
(at least trying to) ; and it worked off and on over the last week...
Scott Leibrand
@scottleibrand
May 04 2017 23:10
no, what did you configure in oref0-setup?
you had to choose a CGM type
regardless of what you choose, it will populate monitor/glucose.json with whatever it retrieves. you'll need to know how it's configured to know where that comes from, though, unless it's obvious from the data
Martin Haeberli
@mhaeberli
May 04 2017 23:11
--cgm=G4 --ns-host=https://xxx.herokuapp.com...
tried G4 type as G4 last week at someone’s suggestion here - probably yours, I think ...
is there a report I can try? I was trying to look at RTD and found the thing about do-everything ...
and openaps gather-clean-data
but at least the latter doesn’t seem to run here.
so the curl works and gets data
but at the moment openaps doesn’t seem to see it
Scott Leibrand
@scottleibrand
May 04 2017 23:13
find the section in the docs about running openaps aliases and reports manually for troubleshooting
and then start with looking at crontab -l | grep cgm or similar
Martin Haeberli
@mhaeberli
May 04 2017 23:13
k thx
so cgm/glucose.json and ns-glucose.json are less than 30 minutes old (but not current…)
Martin Haeberli
@mhaeberli
May 04 2017 23:19
openaps get-bg
AttributeError: 'NoneType' object has no attribute 'iter_records'
grep: cgm/cgm-glucose.json: No such file or directory
so openaps state is “waiting” - how to “un-wait” it?
Scott Leibrand
@scottleibrand
May 04 2017 23:21
that alias gets data from a microUSB-connected CGM
perhaps you want to start with the crontab -l | grep ns-loop instead
Martin Haeberli
@mhaeberli
May 04 2017 23:21
k thx looking
Kevin Marshall
@ruess
May 04 2017 23:22
Hi all - getting a BG data too old message - closed loop not looping issue
NS shows very recent BG data
Scott Leibrand
@scottleibrand
May 04 2017 23:22
@ruess work with @mhaeberli to do the same troubleshooting he's doing
Kevin Marshall
@ruess
May 04 2017 23:22
and I was able to successfully login to my rig with ssh
Martin Haeberli
@mhaeberli
May 04 2017 23:23
@ruess having exactly same problem i think and debugging now - @scottleibrand understood
Kevin Marshall
@ruess
May 04 2017 23:23
haha - nice @scottleibrand
so i’m assuming since I can login to the rig, it’s connected
i can restart, just wanted to see if there was something else that can be done to prevent this that i may be doing wrong beforehand
Martin Haeberli
@mhaeberli
May 04 2017 23:23
@ruess also you can check ping www.apple.com or ping 8.8.8.7 to check outside connectivity
you can also TRY to ping your nightscout site from the ssh / rig login
if that fails, that doesn’t prove anything; if that works, that’s good
Scott Leibrand
@scottleibrand
May 04 2017 23:24
@ruess don't assume you have Internet just because you have local connectivity. you can ping google.com to check
Kevin Marshall
@ruess
May 04 2017 23:24
interesting - google checks fine, but apple.com doesn’t - hmmm
Martin Haeberli
@mhaeberli
May 04 2017 23:24
@scottleibrand :+1:
@ruess try also 8.8.8.8
Scott Leibrand
@scottleibrand
May 04 2017 23:24
apple.com might block pings
Kevin Marshall
@ruess
May 04 2017 23:25
8.8.8.8 works - what’s that? google dns?
Scott Leibrand
@scottleibrand
May 04 2017 23:25
yep
Kevin Marshall
@ruess
May 04 2017 23:25
oh ya - makes sense about apple
think it may be a mild crash in code on my nightscout?
like it’s not conveying the BG data to the openaps plugin?
Martin Haeberli
@mhaeberli
May 04 2017 23:26
apple pings from here - also could be your provider blocking
Kevin Marshall
@ruess
May 04 2017 23:26
well either way, seems like i have net connectivity
Scott Leibrand
@scottleibrand
May 04 2017 23:27
@ruess unlikely. dig in to see if you can pull BG via the API like @mhaeberli did above
Martin Haeberli
@mhaeberli
May 04 2017 23:27
yes
similar to what i’m seeing
Kevin Marshall
@ruess
May 04 2017 23:27
ok let me try
Martin Haeberli
@mhaeberli
May 04 2017 23:29
openaps ns-loop
results in:
Starting ns-loop at Thu May 4 16:27:36 PDT 2017:
ns://JSON/shell/cgm/ns-glucose.json
reporting cgm/ns-glucose.json
Refreshed temptargetsWarning: Argument order has changed: please update your oref0-meal device and meal.json report to place carbhistory.json after basalprofile.json
 and meal-carbs
{"batteryVoltage":3847, "battery":67}
Uploaded
but NS pill still shows “waiting”
@ruess sorry to abandon you at this stage - i’m very interested in what you / we find … but i’m about to go “bingo power” on my laptop and no compatible charger handy, I think… - going to go look now
@scottleibrand or @all please advise whether I should worry about the warning
Scott Leibrand
@scottleibrand
May 04 2017 23:31
and ignore the pill until you finish troubleshooting on the rig.
Martin Haeberli
@mhaeberli
May 04 2017 23:31
ok re ignore pill . but is the argument order warning relevant?
Scott Leibrand
@scottleibrand
May 04 2017 23:32
take a look at the contents of cgm/ns-glucose.json after doing that
sorry, slow plane wifi.
the argument order warning is correct, but isn't something I (or anyone else) have bothered to fix yet. everyone gets that.
Martin Haeberli
@mhaeberli
May 04 2017 23:33
most recent data now 16:21 vs 16:33 local time
cameronrenwick
@cameronrenwick
May 04 2017 23:37
@mhaeberli I get similar time warnings like that as well. I've searched to see if I can narrow it but didn't have much luck. I am looping so no prob there and thus I've put the ~10min diff down to internet lagging between websites, uploaders etc.
Scott Leibrand
@scottleibrand
May 04 2017 23:39
you should be able to get CGM data from your receiver to rig via NS in 1-2 minutes.
Kevin Marshall
@ruess
May 04 2017 23:40
well i just restarted the rig - and it seems to be back working now - but wonder what happened - it probably had something to do with being thrown off when i got home
and it had to swtich wifi
Scott Leibrand
@scottleibrand
May 04 2017 23:41
you can examine /var/log/openaps/network.log to see if it was having trouble there
Martin Haeberli
@mhaeberli
May 04 2017 23:43
Back on gitter. I'll check the rig in 5 minutes w luck
Kevin Marshall
@ruess
May 04 2017 23:48
ok thanks will do
rsilvers129
@rsilvers129
May 04 2017 23:48
I have a feature idea. Wear a Bluetooth heart rate monitor. There is one wrist cuff one that is well rated. And dispense more or less basal based on your activity level. It would learn how sensitive you are based on heart rate zones. So autosense would have a multi dimensional table.
Kevin Marshall
@ruess
May 04 2017 23:49
nice idea
Martin Haeberli
@mhaeberli
May 04 2017 23:49
:+1: re hrm
Scott Leibrand
@scottleibrand
May 04 2017 23:50
the first obstacle is just getting the HRM data to the rig in real time
Kevin Marshall
@ruess
May 04 2017 23:51
souns like you’d have to have an iOS app of some sort to share the health data to NS - or the android equiv.
Scott Leibrand
@scottleibrand
May 04 2017 23:51
if you know how to solve that (or get real-time step count data), we would probably start with some simple stuff like activating activity mode automatically when you start being active
Martin Haeberli
@mhaeberli
May 04 2017 23:51
@scottleibrand thx 4 your help. Debugging deferred for a few more hours because laptop power
Scott Leibrand
@scottleibrand
May 04 2017 23:52
and then maybe on days with lots of extra activity, we could also inform autosens and have it be prepared to increase sensitivity
but first things first, we need the data, so if any of you want to work on that, that would be worthwhile
rsilvers129
@rsilvers129
May 04 2017 23:52
You could have an Explorer 3 with a 3-axis accelerometer.
Or Bluetooth.
Martin Haeberli
@mhaeberli
May 04 2017 23:53
Assuming openaps ns-loop has current data, what would next command in flow be? Yes I have RTD
rsilvers129
@rsilvers129
May 04 2017 23:53
Just have iPhone upload HR data to Nightscout. And then the APS can see it.
Martin Haeberli
@mhaeberli
May 04 2017 23:54
@rsilvers129 of course Edison has Bluetooth
rsilvers129
@rsilvers129
May 04 2017 23:54
So Edison could connect to the HR cuff.
Martin Haeberli
@mhaeberli
May 04 2017 23:54
And there is a third-party board via SparkFun that has 9-axis accel etc
But you would have to design another wonderful case!
rsilvers129
@rsilvers129
May 04 2017 23:55
Apple Watch has HR already.
Martin Haeberli
@mhaeberli
May 04 2017 23:55
Because the whole system would b thicker
rsilvers129
@rsilvers129
May 04 2017 23:55
HR is better than motion.
Motion is tricked by a bumpy car ride.
Martin Haeberli
@mhaeberli
May 04 2017 23:56
And Apple Watch has accel etc .
rsilvers129
@rsilvers129
May 04 2017 23:56
So the first step is upload Apple Health HR to Nightscout.
Martin Haeberli
@mhaeberli
May 04 2017 23:57
My understanding is that at least MotionX and maybe Apple have solved the car ride problem. As, somehow, has Basis Peak
rsilvers129
@rsilvers129
May 04 2017 23:57
They have. But if we have HR I don't see why steps matter.
Scott Leibrand
@scottleibrand
May 04 2017 23:57
wee, gitter is slow over inflight wifi.
rsilvers129
@rsilvers129
May 04 2017 23:58
Note that Apple Watch only takes a HR reading every 10 minutes unless you are in activity mode.
Scott Leibrand
@scottleibrand
May 04 2017 23:59
there are lots of things we could do. the current challenge is we need someone to step up and actually prototype something. :)
rsilvers129
@rsilvers129
May 04 2017 23:59
But that is still very useful to know.
Scott Leibrand
@scottleibrand
May 04 2017 23:59
back when I looked at it, there was no easy way to get fitbit or other data via a real-time API. maybe that has changed now?