ahhhh..... can't ssh into RP3
thanks...got it.
Need a little help figuring this out
ns-upload <my NS site> <my API SECRET> devicestatus.json monitor/upload-status.json
just get .... Input file monitor/upload-status.json does not exist.
Silly question - does it?
but cant seem to create the .json
Ah.
is the API secret the plain api or the hashed api
My
upload-status.json actually lives in upload. Are you certain your report says that it’s in monitor?
was just following rtd section
let me check there
i just have clock.json and glucose.json in monitor right now
so you have a separate upload directory?
do you have
nightscout autoconfigure-device-crud?
@bewest I think that's how I set it up. Incidentally, I tried a while back to make PR to add crud method to rtd...don't thin its in there currently though.
@bewest if I just re entered
nightscout autoconfigure-device-crud it should work..right?
oh i see........says...ConfigParser.DuplicateSectionError: Section 'device "ns"' already exists
maybe that's OK?
@bewest You know how you’re encouraging everyone to use the
oref0 tools? Do you happen to have a list somewhere of the tools/devices/reports whose functionality is now provided by the oref tools instead of things like iter_glucose and the like? I’d like to go through the docs, find the old stuff, and replace it with oref0.
@Spazholio it's a little more subtle....
what's happened is that
oref0 received several updates to be easier to use
for formatting glucose data from dexcom g4, there's a new use called
oref0_glucose, it's part of the openaps/dexcom vendor
it combines sensor and glucose records same way nightscout does
so it's ideal for both oref0 and nightscout
there are several issues with notes I created to help people who are working on docs
they represent lots of suggestions I have for how to create a loop that is debuggable/easy to maintain/and easy to set up
it also assumes basics/defaults, it does not enable advanced features
for setting up aliases and one-off devices more easily: openaps/docs#136
for installing openaps and associated tools, using pip: openaps/docs#133
using
oref0 template to set up common reports, devices, and even aliases: openaps/docs#132
for integrating with nightscout, and an introduction to
lsgaps, select and friends, openaps/docs#129
basically the
ns device is now like a nightscout swiss army knife
it can do uploads, downloads, gap syncing all builtin
@garykidd so if you have eg:
monitor/device-status-for-ns-to-upload.json, you can do openaps use ns shell upload devicestatus.json monitor/device-status-for-ns-to-upload.json
that should upload the file once to
devicestatus endpoint
you can then put it in a report and add the report to alias
combined, the changes/tips/tools discussed in those four issues should make setting up openaps and maintaining it a lot simpler
Well, not what I was looking for, but definitely what I needed. Thanks. =)
so for example
a lot of the set up stuff... making all the reports...
it's not needed "if you have a medtronic device called
pump, oref0 template ... | openaps import creates all the reports with all the zones/units applied
Well, once I get my Edison set up, I’ll have a spare rig to test things on, like doing a full setup from scratch.
@benwest all I have in monitor is glucose.json and clock.json do I simply need add device report for decice-status-ns-to-upload.json ?
you need a file prepared to upload
looks like
ns-status prepares the file for devicestatus
so there's probably a report for
ns-status to create that file
hmmm...tried
openaps report add monitor/upload-status.json JSON ns-status shell monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/battery.json monitor/reservoir.json monitor/status.json
sounds good, does
invoke monitor/upload-status.json create the file?
hmmmm....wth...
pi@raspberrypi:~/APS0 $ invoke monitor/upload-status.json
-bash: invoke: command not found
-bash: invoke: command not found
You forgot the
openaps in front.
openaps report invoke
actually tried that
should be different error then
openaps report invoke monitor/upload-status.json
shoot forgot report...
but still... openaps report invoke monitor/upload-status.json
ns-status://JSON/shell/monitor/upload-status.json
Could not require: /home/pi/APS0/monitor/iob.json { [Error: Cannot find module '/home/pi/APS0/monitor/iob.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/enact/suggested.json { [Error: Cannot find module '/home/pi/APS0/enact/suggested.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/enact/enacted.json { [Error: Cannot find module '/home/pi/APS0/enact/enacted.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/monitor/clock-zoned.json { [Error: Cannot find module '/home/pi/APS0/monitor/clock-zoned.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/monitor/battery.json { [Error: Cannot find module '/home/pi/APS0/monitor/battery.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/monitor/reservoir.json { [Error: Cannot find module '/home/pi/APS0/monitor/reservoir.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/monitor/status.json { [Error: Cannot find module '/home/pi/APS0/monitor/status.json'] code: 'MODULE_NOT_FOUND' }
reporting monitor/upload-status.json
ns-status://JSON/shell/monitor/upload-status.json
Could not require: /home/pi/APS0/monitor/iob.json { [Error: Cannot find module '/home/pi/APS0/monitor/iob.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/enact/suggested.json { [Error: Cannot find module '/home/pi/APS0/enact/suggested.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/enact/enacted.json { [Error: Cannot find module '/home/pi/APS0/enact/enacted.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/monitor/clock-zoned.json { [Error: Cannot find module '/home/pi/APS0/monitor/clock-zoned.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/monitor/battery.json { [Error: Cannot find module '/home/pi/APS0/monitor/battery.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/monitor/reservoir.json { [Error: Cannot find module '/home/pi/APS0/monitor/reservoir.json'] code: 'MODULE_NOT_FOUND' }
Could not require: /home/pi/APS0/monitor/status.json { [Error: Cannot find module '/home/pi/APS0/monitor/status.json'] code: 'MODULE_NOT_FOUND' }
reporting monitor/upload-status.json
so look slike Ijust ned to invoke a bunch more reports...right?
Correct.
admitedly i have left my work here a number of times only to re-engage after memory fades...but.
in re-engaging this time. it certainly seem like rtd jumps into needed reports for NS integration...before it covers anything about adding invoking reports etc.
openaps report invoke monitor/iob.json
usage: openaps-report invoke [-h]
{bg_targets.json,carb_ratios.json,current_basal_profile.json,enact/suggested.json,insulin_sensitivities.json,last2bgs.json,last_four_cgm_hours.json,last_four_pump_hours.json,lastbg.json,monitor/clock-zoned.json,monitor/clock.json,monitor/closck.josn,monitor/glucose.json,monitor/temp_basal,monitor/upload-status.json,pump_settings.json,pump_status.json,read_Battery_status.json,read_clock.json,recent_pump_data.json,settings/basal_profile.json,settings/bg_targets.json,settings/insulin_sensitivities.json,settings/profile.json,settings/settings.json}
[{bg_targets.json,carb_ratios.json,current_basal_profile.json,enact/suggested.json,insulin_sensitivities.json,last2bgs.json,last_four_cgm_hours.json,last_four_pump_hours.json,lastbg.json,monitor/clock-zoned.json,monitor/clock.json,monitor/closck.josn,monitor/glucose.json,monitor/temp_basal,monitor/upload-status.json,pump_settings.json,pump_status.json,read_Battery_status.json,read_clock.json,recent_pump_data.json,settings/basal_profile.json,settings/bg_targets.json,settings/insulin_sensitivities.json,settings/profile.json,settings/settings.json} ...]
openaps-report invoke: error: argument report: invalid choice: 'monitor/iob.json' (choose from u'bg_targets.json', u'carb_ratios.json', u'current_basal_profile.json', u'enact/suggested.json', u'insulin_sensitivities.json', u'last2bgs.json', u'last_four_cgm_hours.json', u'last_four_pump_hours.json', u'lastbg.json', u'monitor/clock-zoned.json', u'monitor/clock.json', u'monitor/closck.josn', u'monitor/glucose.json', u'monitor/temp_basal', u'monitor/upload-status.json', u'pump_settings.json', u'pump_status.json', u'read_Battery_status.json', u'read_clock.json', u'recent_pump_data.json', u'settings/basal_profile.json', u'settings/bg_targets.json', u'settings/insulin_sensitivities.json', u'settings/profile.json', u'settings/settings.json')
usage: openaps-report invoke [-h]
{bg_targets.json,carb_ratios.json,current_basal_profile.json,enact/suggested.json,insulin_sensitivities.json,last2bgs.json,last_four_cgm_hours.json,last_four_pump_hours.json,lastbg.json,monitor/clock-zoned.json,monitor/clock.json,monitor/closck.josn,monitor/glucose.json,monitor/temp_basal,monitor/upload-status.json,pump_settings.json,pump_status.json,read_Battery_status.json,read_clock.json,recent_pump_data.json,settings/basal_profile.json,settings/bg_targets.json,settings/insulin_sensitivities.json,settings/profile.json,settings/settings.json}
[{bg_targets.json,carb_ratios.json,current_basal_profile.json,enact/suggested.json,insulin_sensitivities.json,last2bgs.json,last_four_cgm_hours.json,last_four_pump_hours.json,lastbg.json,monitor/clock-zoned.json,monitor/clock.json,monitor/closck.josn,monitor/glucose.json,monitor/temp_basal,monitor/upload-status.json,pump_settings.json,pump_status.json,read_Battery_status.json,read_clock.json,recent_pump_data.json,settings/basal_profile.json,settings/bg_targets.json,settings/insulin_sensitivities.json,settings/profile.json,settings/settings.json} ...]
openaps-report invoke: error: argument report: invalid choice: 'monitor/iob.json' (choose from u'bg_targets.json', u'carb_ratios.json', u'current_basal_profile.json', u'enact/suggested.json', u'insulin_sensitivities.json', u'last2bgs.json', u'last_four_cgm_hours.json', u'last_four_pump_hours.json', u'lastbg.json', u'monitor/clock-zoned.json', u'monitor/clock.json', u'monitor/closck.josn', u'monitor/glucose.json', u'monitor/temp_basal', u'monitor/upload-status.json', u'pump_settings.json', u'pump_status.json', u'read_Battery_status.json', u'read_clock.json', u'recent_pump_data.json', u'settings/basal_profile.json', u'settings/bg_targets.json', u'settings/insulin_sensitivities.json', u'settings/profile.json', u'settings/settings.json')
openaps device show
medtronic://GK715
dexcom://GKDEX
process://ns/nightscout/ns <> <>
process://oref0/oref0/
process://get-profile/oref0/get-profile
process://calculate-iob/oref0/calculate-iob
process://determine-basal/oref0/determine-basal
process://ns-status/ns-status/
timezones://tz
medtronic://GK715
dexcom://GKDEX
process://ns/nightscout/ns <> <>
process://oref0/oref0/
process://get-profile/oref0/get-profile
process://calculate-iob/oref0/calculate-iob
process://determine-basal/oref0/determine-basal
process://ns-status/ns-status/
timezones://tz
does the process//ns-status/ns-status/ look right?
@danamlewis Hello Dana, just sent you the form to registrate! Yabba Dabba Doo, I had my three nights in a row!
That's great :star2: ......the German openAPS community grows :+1:
:)
@Missdazzle yeah
Dana might run out of numbers ;-)
I think I have some expertise on that (running out of numbers, like IPv4 addresses), and I can say that will not be a problem. ;-)
(I'm right now at a conference where the main topic for the last decade has been the fact that we're running out of 32-bit numbers (IPv4) and trying to get everyone to switch to a new protocol (IPv6) that uses 128-bit numbers.)
2^128 = 340 undecillion. you know, million billion trillion quadrillion quintillion sextillion septillion octillion nonillion decillion undecillion. ;-)
good to hear :-)
...but we are running out of pumps :worried:
I haven't seen any evidence of that yet, at least in the US.
the grassroots pressure on the market to provide pumps that do what we want cannot be understated
that is a real issue, in Germany it is really difficult getting old medtronic pumps on ebay
vendors respond to market demands :-)
has anyone started working on an openaps driver for the Dana R?
I know people are looping with custom phone apps that haven't been open sourced yet AFAIK, so that could be an alternative if/when someone writes the code
@bewest if they are not too scared :-(
there seem to be a lot of concerns about liability
but to be honest, currently I would only see to options when going for a new pump, either medtronic 640 to get the supplies for my old 715 or Dana R
Morning
I am struggling to find the Google Docs to register multiple days of loop running (!!!), any tips?
@C-Ville There is also work ongoing with the 640G to loop with it, using microboluses for high-temps and relying on Smartguard PLGM to turn off insulin delivery if BG is predicted to go low.
Sweet thanks!
I still have not gotten the Blood GLucose to Nightscout working, relying on xDrip for this.
If anybody is up for helping me troubleshoot, I would love it
@scottleibrand i've an app with mealassist ver of openaps ... but it's android only. history is not usable like on medtronic pump so porting to driver for oref0 is wasting of time
bummer. is there enough data from the pump (perhaps reservoir status) to build our own insulin dosing history if we keep track of state over time?
bolusing history is fine but there is no basal history (events like we upload to ns)
ok, but you can see what temp basal is currently running, right?
so openaps could create its own temp basal history file, creating temp basal records when it sees the pump reporting it has started a new temp basal, and creating cancel-temp records when it sees the pump reporting that a previously running temp has ended early
yes, i create temp basal history manually from pump status
ok. no reason we couldn't do that with a new openaps
use as well
i'm pretty happy how it works on android. one less device because you need phone for tethering anyway. and another pros you can use xdrip. share is not available in europe
how close do you think we are to being able to open-source the android app in some fashion?
@MilosKozak did you get a hold of a Dana running the new firmware to check if this still loops?
not yet
Wonder if I should try to have someone get me a DanaR to use for testing so we can figure out how to open-source an app like this safely. Would probably need to make it start with a conservative algorithm (like low glucose suspend only) and gradually allow users to enable more features as they get experience with it and demonstrate some level of understanding of what it's doing and why.
:+1:
(ditto. luckily, @scottleibrand & I share the same mailing address so we can collaborate even more on this ;))
we might get name overload if we try to get Dana a Dana, though. :-p
Dana^2.
Joking aside, that’s why I ask people to describe the pump as DanaR, otherwise it’s realllllly confusing :)
kind there is some "stolen" code because it is the only way :(
morning
morning!
How's London? I won't be able to make it for lunch (at work until 1330)
My colleagues are awesome (and it turns out one of my colleagues knows @andrew-warrington! it’s a fantastically small world) so mostly been at work, but all around great :)
can’t wait to see a few people at lunch today and more for dinner tonight!
I'm looking forward to this evening :)
:)
and it is not good to open source code you didn't write :(
@LorelaiL are you talking about @MilosKozak' DanaR app or something else?
About the same code
so what's the story there?
Low level comm is from decompiled their own mobile app. I did many changes there but to make it looking different would need more work
ah, ok.
that seems like a reasonable starting place. obviously rewriting it ourselves would be better, but glad we're not talking about something we got from a disgruntled former employee or something
what language is it written in?
Java
how much code is it?
I killed my openAPS installation yesterday. so I set up a new installation and tried to copy all of my old .ini files into the new instance, now I am getting Could not parse carbratio_data. Feature Meal Assist enabled but cannot find required carb_ratios. and the suggest will not work is there an easy way to turn meal assist off, I do not even know how this turned on
hey @danamlewis @timomer, just waiting for a train at Gatwick now
flight was delayed a little
hoping to be there around 1pm
colinlennon: where did you come from?
Sounds good @colinlennon ! See you soon.
Glasgow
A little sunnier down here :smile:
Hey all, I'm at the bower, just sitting to the right of reception. Just ping me on here by name if needed!
@C-Ville are you passing meal.json to determine-basal?
If so, you can remove that. Or you can add a report to get the carb ratios from the pump and pass them to get-profile.
@danamlewis @scottleibrand @bewest
Yesterday I got a package. I was very surprised because I didn't know who was sending me a apron.
My germanLOOP friends gave me this wonderful gift as compliment in terms of motivation to improve the A1C.
Our goal is to archive a 5,x A1C (see the „5“ club label on the apron).
Furthermore my loop number is printed on the apron – yeah.
That’s me on this photo.
pretty much all code in DanaAps/app/src/main/java/info/nightscout/danar/ and most of DanaAps/app/src/main/java/info/nightscout/danaaps/ needs to deleted and rewritten from scratch , just make sure it doesn't even look like you ever seen any of the code
scott, I do not see any report in my openaps.ini calling for meal.json
:D
@ceben80 I can´t wait to see that on our summer meeting
@ceben80 so awesome! :)
@C-Ville need to reproduce it on the command-line then. Run oref0-determine-basal with the same inputs called for by your report and see if you get the error?
@LorelaiL there is nothing wrong with using their decompiled code for guidance. Only real argument would be a copyright one, and I'm not sure that would even apply to decompiled code (vs. their actual source). So a basic rewrite should be fine. Not like they can't read the logs here and know what we did.
Love to see monitoring moved to the front of the openaps setup docs. :)
Still need some PRs to put dependencies up there too.
@scottleibrand it shouldn't be copy paste style "guidance"
Right now lots of people are getting stuck because things are out of order.
@LorelaiL agreed.
Having some errors installing openaps via easy_install on yocto, but pip seems to do the job fine. Any reason for using easy_install instead of pip? I’ve recently been doing a lot of python for my day job, and it seems like pip is becoming the de-facto standard,
Only because you haven't updated the script yet. ;-)
Wasn’t using the script; but I’m guessing an equivalent response applies. ;)
so to let a person know, we are not paying 10K GBP for acc u che k combo? any final word?
@LorelaiL why would you have to pay that much?
@LorelaiL it really depends on whether this person is willing to be public about it and explain what exactly they have and how they got it. if they insist on privacy, I don't think the OpenAPS community can work with that: we have to do everything openly and avoid doing anything that might even appear to be shady, or we'll find ourselves in all kinds of trouble.
And - personally speaking - it feels like it would fly in the face of the whole “Open” aspect of “OpenAPS”.
:+1:
@scottleibrand sounds good, thanks
Everyone at the London lunch meet up, lovely meeting you all, sorry I cannot make it tonight, have fun!
@mddub Question for your OSX menubar. First off, thanks - I’m digging it quite a bit. Would it be possible to add a menu item to the drop down that links to the NS install?
Hi wonderful people! It’s time for me to ask the obligatory cron question. I can see openaps run in a cron job, but the second that I try to add an alias parameter to the openaps invocation, it says ‘bash: openaps: command not found’. Here’s my crontab:
*/1 * * * * cd ~/clive-openaps-1 && (/usr/local/bin/openaps > hello.log 2>&1)
*/2 * * * * cd ~/clive-openaps-1 && (/usr/local/bin/openaps retry-loop > retry-loop.log 2>&1)The first line puts what you would expect into hello.log. retry-loop.log ends up containing ‘bash: openaps: command not found’.
Any help is greatly appreciated! Thank you!
use a PATH instead of specifying the full path to every command
Where can I specify that so that cron can use it?
paste PATH=whatever at the top of the crontab
ok
but I don’t quote PATH anywhere?
echo $PATH to see your current one
Now they’re not running. “cat /var/log/syslog | grep cron” shows the crontab editing and reload, but now the commands aren’t running.
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
*/1 * * * * cd ~/clive-openaps-1 && (openaps > hello.log 2>&1)
*/2 * * * * cd ~/clive-openaps-1 && (openaps retry-loop > retry-loop.log 2>&1)
oh wait, no, I’m wrong
looks like it’s working but for some reason not going to that syslog now. Okay, will verify, but thanks!
np
So yeah, it works Scott! The weird thing is that all of my loop works except the upload to nightscout. monitor/upload-status.json gets generated, but the upload doesn’t go through. Has anyone run into this before?
You need to do the same with your NS ENV vars
NIGHTSCOUT_HOST and API_SECRET
Scott, thanks for your support, I ended up setting it up basically from scratch, now I do get the standard basal rate set as temp, but I can live with that for now.
There is a new option to disable that if you don't like it.
If this is intended I am fine with it, had a fairly old installation from back in march and was not aware of these changes
are y’all talking about dev or master?
Yeah, we basically have to do that to support autosens and meal assist, and it is nice for seeing on NS when the loop was working or not.
Oref0 dev and master are the same for now.
cool. i’m looking forward to learning about the other features soon
He was using the old master
advanced meal assist will be coming to dev soonish
@danamlewis : made some progress regarding extracting BG values from a publicly available NS website! Keep you guys updated when I figure out how to get the data correctly! :-)
:+1:
@bewest Remember when I was having different data coming back from NS versus the connected Dexcom? Do you know if that was something that got fixed with the
nightscout device? And even more improbably, do you recall which field you were saying that it needed to have in it? There were a couple date fields, but from NS apparently wasn’t returning it...
@mm22dl awesome
well i’m still not done. i’ve done some googling and it seems that cron jobs are supposed to run even when a user isn’t logged in, but if I close all my ssh terminals into the OpenAPS device, the job doesn’t run. i created the crontab as user, not sudo/root.
(and am on raspberry pi)
Is this another obvious question? :)
@morsen does the command in cron work from the command line?
If you run
sudo raspi-config and choose option 3, I think you want to either choose B2 or B4. That way, the pi user is always logged in.
If it does, then your time spec is wrong in cron.
Cron will run even if you are logged out.
Duh. @eluhrs is right.
What does your cron look like again?
This message was deleted
*/5 * * * * cd ~/clive-openaps-1 && (openaps retry-loop > retry-loop.log 2>&1)
yes, i can run that from the command line
Your PATH is set correctly in cron as well?
I set it the same as in my command line shell, except i removed the two ‘games’ paths
crontab:
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
cmd line:
$ echo $PATH
/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/games:/usr/games
I think it’s not even running, because the timestamp on my log file from the command doesn’t change while I’m logged out.
On another note - I know I’m only supposed to calibrate my Dexcom twice a day (I’ve heard overcalibrating can wonk it out) and we’re “supposed” to only use each sensor for precisely 7 days, but I’m finding that when I use it more than 7 days, my readings are occasionally way off. Dexcom is telling me I’m at 61, but in actuality I’m 111. That’s pretty far off.
we usually see the sensor starts to get more noisy and jumpy a few days before it really wonks out, usually sometime in the 10-14 day range
Yeah, I’ve definitely been jumping. =/ What do you mean by “noisy” though?
no matter the sensor age, you can get what we call "compression drops" from sitting or laying on the sensor
I’m couch potato-ing today. No compression issues at the moment. =)
noisy = successive readings bounce around a lot, rather than making a nice line
Ah, ok. I know there’s a “noise” field when you hover over the dots in NS, so wasn’t sure if you were referring to that.
They’re never populated in my NS, so I’m assuming that’s a good thing.
if we were uploading BG from the rig we'd get actual measurements of the noise level like you're talking, which are even more predictive
but we just use BG from Share, so we don't get them either
got carelink usb today. Is it ok to go through the mac for setting it up?
Interesting.
Mac won't work with Carelink: it's not compatible with the Mac's USB3 ports
if you have the Carelink plugged into a Raspberry Pi, though, you can use the Mac to ssh into the Pi
Great to know. So using a Win7 notebook would be better?
no, trying to use Carelink with a desktop computer is a world of pain
better to just get the Pi and focus on Nightscout etc. until it arrives
I have everything set up ; just working on the stick. Thanks. I will try ssh
@danamlewis @morsen So used a very nasty trick to get the data, now I need to find a way to convert it back to BG values. At least, it works flawless and gets all of the data available on a NS webpage. :-)
To verify: ssh, plug carelink usb directly into pi, and drivers will load to pi ok?
@mm22dl are you using the api endpoint?
Might be a bit early for the meal :P just got to angel
@danamlewis Nope... I am doing it more... brutally, using selenium driver and beautiful soup. xD
that is way more difficult
just use the already-documented method for downloading BG data from nightscout, and plug their nightscout URL in instead of your own.
Where’s the fun in that? =)
so if I entered
openaps report add monitor/clock-zoned.json JSON tz clock monitor/clock.json should that not create a clock-zoned.json file in monitor folder ?
@flutelynne_twitter I had an all mac household and used my carelink on my work computer to get data up there so I could generate their reports for my doctor...
@garykidd My
clock-zoned.json is generated by: openaps report add monitor/clock-zoned.json tz clock --timezone "EDT" --adjust "missing" --date "None" --astimezone monitor/clock.json. I’m a little fuzzy on the tz device, but yours looks correct as well.
@Spazholio thanks. I was expecting to see a '''clock-zoned.json''' file in 'monitor' based on that entry but no. Generally having difficulty getting data uploaded to my NS site.
this is all I am getting
pi@raspberrypi:~/APS0 $ $ ns-upload https://XXXYYYZZZ.azurewebsites.net ############ devicestatus.json monitor/upload-status.json
-bash: $: command not found
-bash: $: command not found
GK...uhhhh duuuuh
hehehe
We’ve all done it. =)
I still thin kI have a prob though brb
@TomBoudr_twitter did you
invoke the newly configured clock-zoned.json report?
@scottleibrand Scott, wrong reply :) although I'm sure @Spazholio will see it. - The whole report setup then invoking thing is easy to forget to do.
pi@raspberrypi:~/APS0 $ openaps alias add status-upload '! bash -c "openaps report invoke monitor/upload-status.json && ns-upload https://xxxxxxxxx.azurewebsites.net ############ devicestatus.json monitor/upload-status.json"'
added status-upload ! bash -c "openaps report invoke monitor/upload-status.json && ns-upload https://xxxxxxxxx.azurewebsites.net ############ devicestatus.json monitor/upload-status.json"
Traceback (most recent call last):
File "/usr/local/bin/openaps-alias", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
File "/home/pi/src/openaps/bin/openaps-alias", line 37, in <module>
app( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 52, in call
self.epilog( )
File "/home/pi/src/openaps/bin/openaps-alias", line 27, in epilog
super(AliasApp, self).epilog( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 75, in epilog
self.create_git_commit( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 89, in create_git_commit
self.repo.git.add([diff.b_path], write_extension_data=False)
File "/usr/local/lib/python2.7/dist-packages/GitPython-1.0.2-py2.7.egg/git/cmd.py", line 450, in <lambda>
return lambda args, **kwargs: self._call_process(name, args, kwargs)
File "/usr/local/lib/python2.7/dist-packages/GitPython-1.0.2-py2.7.egg/git/cmd.py", line 902, in _call_process
return self.execute(make_call(), _kwargs)
File "/usr/local/lib/python2.7/dist-packages/GitPython-1.0.2-py2.7.egg/git/cmd.py", line 695, in execute
raise GitCommandError(command, status, stderr_value)
git.exc.GitCommandError: 'git add openaps.ini' returned with exit code 128
stderr: 'error: insufficient permission for adding an object to repository database .git/objects
error: openaps.ini: failed to insert into database
error: unable to index file openaps.ini
fatal: updating files failed'
added status-upload ! bash -c "openaps report invoke monitor/upload-status.json && ns-upload https://xxxxxxxxx.azurewebsites.net ############ devicestatus.json monitor/upload-status.json"
Traceback (most recent call last):
File "/usr/local/bin/openaps-alias", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
File "/home/pi/src/openaps/bin/openaps-alias", line 37, in <module>
app( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 52, in call
self.epilog( )
File "/home/pi/src/openaps/bin/openaps-alias", line 27, in epilog
super(AliasApp, self).epilog( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 75, in epilog
self.create_git_commit( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 89, in create_git_commit
self.repo.git.add([diff.b_path], write_extension_data=False)
File "/usr/local/lib/python2.7/dist-packages/GitPython-1.0.2-py2.7.egg/git/cmd.py", line 450, in <lambda>
return lambda args, **kwargs: self._call_process(name, args, kwargs)
File "/usr/local/lib/python2.7/dist-packages/GitPython-1.0.2-py2.7.egg/git/cmd.py", line 902, in _call_process
return self.execute(make_call(), _kwargs)
File "/usr/local/lib/python2.7/dist-packages/GitPython-1.0.2-py2.7.egg/git/cmd.py", line 695, in execute
raise GitCommandError(command, status, stderr_value)
git.exc.GitCommandError: 'git add openaps.ini' returned with exit code 128
stderr: 'error: insufficient permission for adding an object to repository database .git/objects
error: openaps.ini: failed to insert into database
error: unable to index file openaps.ini
fatal: updating files failed'
any ideas?
Corrupt git repo? I’ve not seen it happen to mine, so I don’t know the symptoms, but...
Yeah, try oref0-reset-git
@TomBoudr_twitter Thanks! I ended up using a Win7 Lenovo and got the stick communicating. Verified getting reports on the pi, but I still get an "unknown" on the NS site. Pump info is blank on NS site, too.
Hi guys. Just getting started. Q1 For a Medtronic pump, does it matter for OepenAPS if the pump is Canadian with mmol?? Also, for any pump, the firmware must be 2.4A or lower? TIA
@aabramowicz It does not matter your pump is mmol, there are unit conversions that can handle this. Yes 2.4A or lower is what you need.
This message was deleted
Awesome...thanks..Now to find one. Not many available in Canada.
medwow.com or craigslist
on medwow now....submitted some inquiries a few days ago...no replies.
@aabramowicz Same, weeks ago. It didn't go at all as I expected.
On Medwow? Someone got back to me after 2 weeks. Not saying it’s definitely gonna happen, but don’t give up hope.
It was Medwow, yes. I still have my inquiries in but I thought it would be much more eBay-like in that people would watch their listings for interested buyers a little more intently.
You would figure people would jump at $1000-$2000 bux for an old pump in their drawer...guess not.
...not that I went after those sellers, haha.
i never had luck on medwow but had luck on craigslist. Not in my area, though, so it was riskier than I’d usually do, but it worked out.
I might have a friend who is considering selling his 722. If anyone is interested... That is all the details I have as of now.
@baughaw I am definitely interested.
of course depending on price :)
@Spazholio yes, my advice is to not use
rm, mv and similar commands in alias, it's making it hard to debug/understand what's happening
I am seeing him tonight at the hockey game, so I will let you know.
I submitted several inquiries on medwow a few weeks back, and never received a single response. =\
well that's encouraging...
@eyim extremely graciously let us borrow their 722 pump!!! (THANKS!)
:thumbsup:
@baughaw Thanks!
@baughaw Definitely interested - what general area are you located in?
Home of the soon to be Stanley Cup champs.. Pittsburgh :)
anyone, i’m trying to find the right path forward now on increasing radio range from OPS to medtronic pump. The gitter rooms ps2/rileylink and mmeowlink are both related to RileyLink? What hardware is the next-recommended? Is there support for non-CareLink in master, or only in dev (or other)?
@baughaw Depending on the price your friend is willing to let it go for that's a trip I would be happy to make to buy in person.
@baughaw Nice, should be a good game 7 tonight
@jaylagorio i'll let you guys know. @all Sorry to be filling the feed with my banter. Just thought I could help someone out who wants to get their loop running.
:thumbsup:
just in case someone sees this instead of my message above: what’s the next radio hardware after carelink for attempting to extend radio range to medtronic pumps and/or where is related documentation? :) thanks again, everyone
tI stick ;) @morsen
Check out the mmeowlink wiki for details
For people at the London meet up: https://communities.intel.com/thread/60728 @danamlewis
Thanks @cjo20 :) hey @scottleibrand ^
@aabramowicz before I knew about this I would done that for sure. With my 530G my 722 was just collecting dust.
Thanks @danamlewis !
:)
Any tips on how to carry around the rig? I am currently using a Grid-IT, but would like to get size down
@danamlewis, also, is there a compatible parts list for the Edison setup? I’ll just add the MMeowlink to my Pi setup first, but I want to get prepared with hardware for the next steps. Thanks again! :)
With working rig, using a connected Dexcom G4 - once retry-loop has run, and data uploaded to Nightscout, BGM should go from oref up to Nightscout right?
I am trying to figure why it isn't working still, only solution I have is to get BGM into Nightscout via a xDrip integration
@sjolundjohn , we used this from one of the recommended parts lists, and it has been working well so far. It’s a tight fit, but has been working: http://www.amazon.com/Lowepro-Dashpoint-Camera-Attachment-Mirrorless/dp/B008OQUYDA?ie=UTF8&psc=1&redirect=true&ref_=oh_aui_search_detailpage
@morsen thanks! Battery is my limiting factor, but checking out camera bags is a good idea
@sjolundjohn I’m using this battery: http://www.amazon.com/MoKo-9000mAh-Portable-Simultaneously-Charge-discharge/dp/B00FGN1OBU?ie=UTF8&psc=1&redirect=true&ref_=oh_aui_search_detailpage
@danamlewis nevermind, I see that the mmeowlink repo documents these other setups. Thanks
I was going to say ;)
:+1:
@bewest Interesting. How do you recommend doing things like your
raw-dir concept without mving the files, or clearing out the monitor dir prior to running the loop?
I don't understand why it's needed, I guess https://github.com/bewest/openaps-example mine is very simple in order to make it easy to translate to the daemon stuff that hopefully is forthcoming
this loop uses an
oref0 copy-fresher as a report in a few places as a "null" translation
also, this is basically output of all the
oref0 template stuff
made a few tweaks maybe
Very cool. I’ll look that over.
Thanks everyone that came to the meet up. Was good fun.
:+1:
my loop died again on me, when I try basics command like openaps use pump model, I get the model number and then an error
pi@APS:~/APS $ openaps use pump model
"715"Traceback (most recent call last):
File "/usr/local/bin/openaps-use", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
File "/home/pi/src/openaps/bin/openaps-use", line 63, in <module>
app( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 52, in call
self.epilog( )
File "/home/pi/src/openaps/bin/openaps-use", line 50, in epilog
super(UseToolApp, self).epilog( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 75, in epilog
self.create_git_commit( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 87, in create_git_commit
diffs = self.repo.index.diff(None)
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.3-py2.7.egg/git/index/util.py", line 69, in check_default_index
return func(self, args, *kwargs)
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.3-py2.7.egg/git/index/base.py", line 1218, in diff
return super(IndexFile, self).diff(other, paths, create_patch, kwargs)
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.3-py2.7.egg/git/diff.py", line 139, in diff
proc.wait()
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.3-py2.7.egg/git/cmd.py", line 325, in wait
raise GitCommandError(self.args, status, errstr)
git.exc.GitCommandError: 'git diff --abbrev=40 --full-index -M --raw --no-color' returned with exit code 128
stderr: 'error: object file .git/objects/b1/97d7ebbacc640aa0d75d30651ebb6589c9bfe7 is empty
fatal: unable to read b197d7ebbacc640aa0d75d30651ebb6589c9bfe7
"715"Traceback (most recent call last):
File "/usr/local/bin/openaps-use", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
File "/home/pi/src/openaps/bin/openaps-use", line 63, in <module>
app( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 52, in call
self.epilog( )
File "/home/pi/src/openaps/bin/openaps-use", line 50, in epilog
super(UseToolApp, self).epilog( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 75, in epilog
self.create_git_commit( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 87, in create_git_commit
diffs = self.repo.index.diff(None)
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.3-py2.7.egg/git/index/util.py", line 69, in check_default_index
return func(self, args, *kwargs)
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.3-py2.7.egg/git/index/base.py", line 1218, in diff
return super(IndexFile, self).diff(other, paths, create_patch, kwargs)
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.3-py2.7.egg/git/diff.py", line 139, in diff
proc.wait()
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.3-py2.7.egg/git/cmd.py", line 325, in wait
raise GitCommandError(self.args, status, errstr)
git.exc.GitCommandError: 'git diff --abbrev=40 --full-index -M --raw --no-color' returned with exit code 128
stderr: 'error: object file .git/objects/b1/97d7ebbacc640aa0d75d30651ebb6589c9bfe7 is empty
fatal: unable to read b197d7ebbacc640aa0d75d30651ebb6589c9bfe7
does anyone have an idea what could be the reason? I had the loop back running later today and then did a shutdown removed the SD card and made an image on my windows machine afterwards I started the pi again and installed wifi, then the loop would not run anymore :-(
You should be running oref0-reset-git in cron. Is that not in the docs? Someone should add it.
I had the same error last night on my Edison. Thought it was corrupted so I wiped my local repo and rebuilt it
scott, thanks will keep that in mind for next time :-)
The oref0-reset-git script calls the fix-corruption script to attempt to fix before moving .git and re-init'ing if unfixable.
And does nothing if
git status is ok. So perfectly safe to run in cron every minute.
Any volunteers to add that to the docs?
is deleting my <APS> directory and then init again enough to get back to a clean instance?
Scott, I will do that, so I will never forget that
No need to do it manually. Run the script and it should work.
It won't delete your configs.
Just move git history off to another directory if it can't recover it.
uff, spend the whole day making these configs
?
I was afraid I would need to start from scratch again
We didn't want to do that easier, so we wrote fix-it scripts. :-)
As you mention it, I think I even had this cron in my old setup that crashed yesterday and never understood what is was for, so I forgot to put it in today
Where do you think it belongs in the docs?
Doesn't much matter. Maybe the cron section?
Actually, I think I have it. One sec, will submit the PR.
I have
oref0-reset-git running every minute. I should probably prepend that to the loop - every minute is a bit overkill (even if it doesn’t hurt anything). Thoughts?
it may take me a while to get something entered, still figuring out how to do that
@Spazholio it need not be serialized with the loop
PR submitted.
as it can run independently, and usually does nothing, no reason not to check every minute.
Kudos on all the PRing
:+1:
@C-Ville try the first-pr thing to add your name to the list of loopers if you haven't already.
good point
@danamlewis I do what I can. I like being able to contribute. =)
:) :+1: :clap:
@scottleibrand @danamlewis Wow... I was clearly trying too hard! Thanks for mentioning the API endpoints... Turns out, all one's gotta do is type the following in their browser, then lightly parse: <NS URL>/api/v1/entries?count=<# of BG values wished>
:+1:
Looking at th
Looked 5 minutes at the code then got it. 😅
Trying to remember, did we have people in nyc or who would be interested in a nyc evening meetup?
@morsen no problem. Glad to see you are up and running. Let me know if I can help in any other way
Hi everyone - I just finished my third night of OpenAPS closed loop with a Medtronic pump, Dexcom G5 and Pi 3 rig. Thank you all for your vision, determination and hard work to make this project possible.