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

9th
Feb 2016
Jim Matheson
@jmatheson
Feb 09 2016 00:22
what is the best way to manually adjust my insulin action curve? looks like it pulls it into settings.json
i am only allowed to set it to whole hours on my pump firmware which is rediculous and a huge jump between 3 and 4
Aaron Michelson
@Aaybob
Feb 09 2016 00:29
|| ( ( mm-stick warmup 2>&1 || sudo oref0-reset-usb ) | egrep -v \"^ \"; echo FAIL; openaps get-bg; sleep 120; exit 1 )"'
this checks for carelink dongle working, if not it resets the usb's. After either command it checks that the output of either contains "^", but no matter what result it echo's FAIL and sends info to NS and pauses for two minutes and then exits OpenAPS. Is that right?
Scott Leibrand
@scottleibrand
Feb 09 2016 00:44
the egrep stuff just gets rid of stuff that starts with whitespace in the output
it's designed to let you see tracebacks without all the irrelevant details
you can skip that part if you want.
otherwise, correct
Aaron Michelson
@Aaybob
Feb 09 2016 00:54
So here's the logic: Preflight didn't get a good response to 'use pump clock' so IF its a stick problem reset the usb.
Scott Leibrand
@scottleibrand
Feb 09 2016 00:55
Exactly.
Aaron Michelson
@Aaybob
Feb 09 2016 00:56
No reason to be afraid of that.
Scott Leibrand
@scottleibrand
Feb 09 2016 00:57
:-)
Jim Matheson
@jmatheson
Feb 09 2016 01:01
I was thinking about creating an insulin_action_duration_override.json, populating it with {
"insulin_action_curve": 3.5
} and then just checking if that file exists and if so use it instead of the value in the pump. https://github.com/openaps/oref0/blob/fbe12067aef55415ecf14d30e31c846a42c25c87/lib/profile/index.js#L19
Aaron Michelson
@Aaybob
Feb 09 2016 01:03
Most of the "visualization" to NS is not documented? But all I see from others is their NS graphs.
Scott Leibrand
@scottleibrand
Feb 09 2016 01:41
@jmatheson you could do that. it would look like this: https://github.com/openaps/oref0/blob/dev/bin/oref0-get-profile.js#L58
Jim Matheson
@jmatheson
Feb 09 2016 01:42
thx, will give it a try locally
Jason Curry
@mccgm
Feb 09 2016 03:11
I have not yet been able to see the basal visualization show up on my NS graph. I see my OpenAPS temps and pump data, but I don't know where my latest treatment is going. Any suggesttions?
Untitled7.png
Jim Matheson
@jmatheson
Feb 09 2016 03:15
turn on basal
did you turn on the basal plugin?
actually, sorry see your image now. looks like it is turned on
Scott Leibrand
@scottleibrand
Feb 09 2016 03:18
what do you mean by "I see my OpenAPS temps"?
are you talking about the OpenAPS pill (5m ago)?
Jason Curry
@mccgm
Feb 09 2016 03:18
When I mouseover my OpenAPS pill, I see the popup
Jim Matheson
@jmatheson
Feb 09 2016 03:19
yea, it looks to me like you are just rolling with your standard basal and maybe no changes being applied right now
since BG is great :)
Scott Leibrand
@scottleibrand
Feb 09 2016 03:19
more likely your treatments upload is broken
and only the status upload is working
Jim Matheson
@jmatheson
Feb 09 2016 03:19
or that :)
Jason Curry
@mccgm
Feb 09 2016 03:19
@scottleibrand It's definitely that.
I have the aliases in my ini
pi@raspberrypi:~/ayeletssail $ openaps upload-recent-treatments
bash: upload/latest-treatments.json: No such file or directory
No recent treatments to upload
So I'm missing a report? One that saves my latest treatments for upload
Scott Leibrand
@scottleibrand
Feb 09 2016 03:22
your upload-recent-treatments alias needs to call the upload/latest-treatments.json report or alias first
Jason Curry
@mccgm
Feb 09 2016 03:28
So, I should be calling the latest treatments from a local file, not from NS?
Scott Leibrand
@scottleibrand
Feb 09 2016 03:32
not sure what that means
Jason Curry
@mccgm
Feb 09 2016 03:33
Yeah, me neither. Sorry
Scott Leibrand
@scottleibrand
Feb 09 2016 03:37
probably time to start walking through your aliases, figure out what calls what, and paste ones you have questions about
Jason Curry
@mccgm
Feb 09 2016 03:40
Well, this one is
latest-ns-treatment-time = ! bash -c "nightscout latest-openaps-treatment $NIGHTSCOUT_HOST | json created_at"
Which I thought would grab my latest treatment, but I can't tell
Scott Leibrand
@scottleibrand
Feb 09 2016 03:43
that grabs the time of the last treatment that NS knows about
that is used to determine which treatments are newer than that, and should be included in upload/latest-treatments.json
Jason Curry
@mccgm
Feb 09 2016 03:45
Which seems to be working, I don't know how to check and verify
Scott Leibrand
@scottleibrand
Feb 09 2016 03:45
what do you have that creates upload/latest-treatments.json?
Jason Curry
@mccgm
Feb 09 2016 03:49
format-latest-nightscout-treatments = ! bash -c "nightscout cull-latest-openaps-treatments monitor/pump-history-zoned.json monitor/model.json $(openaps latest-ns-treatment-time) > upload/latest-treatments.json"
Scott Leibrand
@scottleibrand
Feb 09 2016 03:50
ok. and does anything run openaps format-latest-nightscout-treatments?
openaps upload-recent-treatments was complaining that hadn't been done (or had failed) and upload/latest-treatments.json didn't exist
Jason Curry
@mccgm
Feb 09 2016 03:53
Yeah, it doesthanks. That helps me begin tracing back through to the right things.
Aaron Michelson
@Aaybob
Feb 09 2016 03:53
Shoot I think I have visualization working but how do I fix this?
blob
blob
Scott Leibrand
@scottleibrand
Feb 09 2016 03:57
@Aaybob I don't get that on your site. click Certificate information and see what it's complaining about
oh, it's a 2017 sha-1 cert. you must have a newer version of chrome than I do
Microsoft needs to get their cert cross-signed by a stronger root anchor: the one they're using now is being deprecated by Chrome.
you can just ignore it for now: Microsoft and Google will probably hash it out soon.
Jason Calabrese
@jasoncalabrese
Feb 09 2016 04:02
Ha-ha
Scott Leibrand
@scottleibrand
Feb 09 2016 04:02
;-)
Jason Curry
@mccgm
Feb 09 2016 04:02
You may not want to post your hostname
Scott Leibrand
@scottleibrand
Feb 09 2016 04:02
yeah, that might be more of a security risk than sha-1 :)
(not that anything is very secure here when we're relying on not sharing URLs)
Aaron Michelson
@Aaybob
Feb 09 2016 04:05
I'd like to ignore it. How do I get passed the lock?
@mccgm what security risk, what hostname?
Jason Curry
@mccgm
Feb 09 2016 04:07
<hostname>.azurewebsites.net
Aaron Michelson
@Aaybob
Feb 09 2016 04:08
can anybody go there and change my settings?
and isn't that public?
Dana Lewis
@danamlewis
Feb 09 2016 04:09
no, your website is not public unless you make it so.
Aaron Michelson
@Aaybob
Feb 09 2016 04:09
actually anybody but me can go there now
Jason Curry
@mccgm
Feb 09 2016 04:09
It is public, and like Scott said, relying on "not sharing urls" is not anything super secure, but if you're good with it...
Aaron Michelson
@Aaybob
Feb 09 2016 04:09
I can't get passed the lock
should I download a different browser?
blob
should I do this process?
Scott Leibrand
@scottleibrand
Feb 09 2016 04:12
there should be a link to ignore and continue. bottom left of the warning screen?
Aaron Michelson
@Aaybob
Feb 09 2016 04:24
blob
There's no warning window. But if I click on the lock I get the information shown.
JaysonEwer
@JaysonEwer
Feb 09 2016 04:24
I have my first endo appt tomorrow, since I learned about and started using both Nightscout and OpenAPS! I'm very excited. My last A1C with my endo in Nov was the worst it has ever been. 8.9. Nightscout is estimating I'm right around 7 trending downward. I actually had a 7.1 A1C at another doctor appointment about a month ago which was also pre OpenAPS. I'm hopeful that my A1C will be the best it's ever been and will only continue to get better as I get the meal assist and the dev branch installed. I am looking to get one of those Official Dr excuses for my OpenAPS. :-D
Scott Leibrand
@scottleibrand
Feb 09 2016 04:25
If it's just changing the icon, that's not an issue. Just ignore it.
Aaron Michelson
@Aaybob
Feb 09 2016 04:26
Does it look like visualization is working on my website? I'm eager to see it.
JaysonEwer
@JaysonEwer
Feb 09 2016 04:26
Thanks again for all the help with this endeavor, everyone!! I really appreciate all you have already done for me and others!!
Jim Matheson
@jmatheson
Feb 09 2016 04:27
:thumbsup:
Aaron Michelson
@Aaybob
Feb 09 2016 04:28
Again,, I would love to ignore it. A locked door is easy to ignore if there is another entrance.
Scott Leibrand
@scottleibrand
Feb 09 2016 04:28
The lock icon is not your problem. What isn't working?
Aaron Michelson
@Aaybob
Feb 09 2016 04:28
I can't see the website
Scott Leibrand
@scottleibrand
Feb 09 2016 04:29
I could
Aaron Michelson
@Aaybob
Feb 09 2016 04:29
Can you see a working website with all the visualization working?
Because that's what I'm expecting to see.
Scott Leibrand
@scottleibrand
Feb 09 2016 04:30
No, just standard NS.
Jim Matheson
@jmatheson
Feb 09 2016 04:30
I can see the site too and I checked and I am on the current version of chrome
Aaron Michelson
@Aaybob
Feb 09 2016 04:31
This is what I see.
blob
Looks like I have to troubleshoot visualization. What a disapointment. But if I can troubleshoot the internet I can do it.
Scott Leibrand
@scottleibrand
Feb 09 2016 04:37
Did you try another browser?
Aaron Michelson
@Aaybob
Feb 09 2016 04:37
NO
Dana Lewis
@danamlewis
Feb 09 2016 04:37
@JaysonEwer :) :) :+1:
Aaron Michelson
@Aaybob
Feb 09 2016 04:38
I don't know how
JaysonEwer
@JaysonEwer
Feb 09 2016 04:38
@danamlewis did you have a dr excuse form letter put together yet? ;-)
was it @jasoncalabrese that mentioned the medical necessity note for his son, or was that someone else's son?
Dana Lewis
@danamlewis
Feb 09 2016 04:42
I have no letters :)
I think a couple of ppl have gotten notes from docs but don't remember who.
Aaron Michelson
@Aaybob
Feb 09 2016 04:44
maybe reboot, googled, went to microsoft, said I already had explorer, "swipe in from the left"? but explorer can't even find the site
That swipe works 5% of the time
Dragan Maksimovic
@dm61
Feb 09 2016 04:50
@danamlewis , @JaysonEwer, notes, letters, ... what for?
Aaron Michelson
@Aaybob
Feb 09 2016 04:51
I can't establish an azure ticket because I don't know what service...
Dana Lewis
@danamlewis
Feb 09 2016 04:53
@dm61 Ppl get them for their kids 504 plans at school. Some adults like to have a letter detailing their medical equipment for air travel, etc. I've never had the need for a letter, personally.
But if you wanted to , you could ask your doc for one, and include artificial pancreas on it
Dragan Maksimovic
@dm61
Feb 09 2016 04:55
I see. I've never had any letters or any problems with going through security. Just went with APS in my hand luggage - they do not care about RPI or anything like that.
Ben West
@bewest
Feb 09 2016 04:56
also has to do with tapping into PCORI/patient-centered stuff
Dragan Maksimovic
@dm61
Feb 09 2016 04:56
Well, I am yet to tell my doctor about APS - expect will be fun to see the face when she gets to see I am changing basal every five minutes :smile:
Ben West
@bewest
Feb 09 2016 04:57
@channemann got a note from his dr that said "patient not using an authorized device; doing well" or similar
the clinicians are another stake-holder in this whole game, making it clear that patients and clinicians are united is a big win; (others are industry, regulators, payors)
Dana Lewis
@danamlewis
Feb 09 2016 04:59
I think that was a note in his record
Ben West
@bewest
Feb 09 2016 04:59
yeah, a note
little scribble or whatever
Dana Lewis
@danamlewis
Feb 09 2016 04:59
@dm61 Hopefully your doc will be more engaged/interested :) than mine was.
But yes, I also haven't had any problems in US or abroad with any of my rigs, even 4 of them at once in my carry on.
Glad you've had the same experience
Aaron Michelson
@Aaybob
Feb 09 2016 05:00
Hi could someone help me with visualization?
Nightscout says OpenAPS is unknown
Ben West
@bewest
Feb 09 2016 05:01
well, I wrote the tools to get data into NS, but I'm not very happy with them because they are complicated to setup/use
well, the treatment stuff at least
dunno about openaps status
are you getting treatments into NS?
Aaron Michelson
@Aaybob
Feb 09 2016 05:01
no, I haven't gotten to that step
Dragan Maksimovic
@dm61
Feb 09 2016 05:02
@danamlewis I doubt it. Last time she was mildly impressed by my pebble showing bg, but asked zero questions about it, and then went on to hand me some 1980's sheets I should use to record fasting bg, 2-hour bg, etc.
Aaron Michelson
@Aaybob
Feb 09 2016 05:02
@dm61 that's so funny
Dragan Maksimovic
@dm61
Feb 09 2016 05:03
I am not kidding - photocopies of sheets with time, bg, etc.
Ben West
@bewest
Feb 09 2016 05:03
oy
Dana Lewis
@danamlewis
Feb 09 2016 05:03
Hehe
Ben West
@bewest
Feb 09 2016 05:03
at this point that would offend me, I think
Dana Lewis
@danamlewis
Feb 09 2016 05:04
I need to send you the pic of how I got my data to my doc one time
Aaron Michelson
@Aaybob
Feb 09 2016 05:04
From the mimeograph?
Ben West
@bewest
Feb 09 2016 05:04
you can import NS data direct to google docs/excel
Dana Lewis
@danamlewis
Feb 09 2016 05:04
iPhone pics of 24hr CGM graph pasted into excel because he wanted a week of data, and I couldn't get it (3 years ago) off my receiver the . And it didn't even phase him
Dragan Maksimovic
@dm61
Feb 09 2016 05:05
:smile: :smile:
Dana Lewis
@danamlewis
Feb 09 2016 05:05
And the industry wonders sometimes why #WeAreNotWaiting :)
Aaron Michelson
@Aaybob
Feb 09 2016 05:06
I can only see my NS website on my iphone safari because of a security problem
Ben West
@bewest
Feb 09 2016 05:06
let's set that one aside
NS is up and running
how is it getting glucose data?
M$ will fix their SSL certs with or without your ticket ;-)
Aaron Michelson
@Aaybob
Feb 09 2016 05:08
from a g4 without share usb otg to android
Ben West
@bewest
Feb 09 2016 05:08
ah ok, it's getting and showing glucose already
Aaron Michelson
@Aaybob
Feb 09 2016 05:09
yes
with plugins : Insulin-on-Board, Care Portal, OpenAPS
Ben West
@bewest
Feb 09 2016 05:10
I have basal bwp iob cob cage careportal on mine
I also have TREATMENTS_AUTH=on, SHOW_RAWBG=always, SHOW_PLUGINS=yes, BASAL_RENDER=icicle, as well as all alarms turned off
ok, so on to openaps sending data into NS
Aaron Michelson
@Aaybob
Feb 09 2016 05:13
I've been stepping through the documentation and someone on facebook helped me get what I needed to start a profile, and then I went to git dev, and then the openaps documentation
Ben West
@bewest
Feb 09 2016 05:13
ah, I uploaded a version of openaps profile
ah yes
mm-format-ns-profile
Aaron Michelson
@Aaybob
Feb 09 2016 05:14
I'm just saying the noted bug that one had to set up a profile when on the master release is what i did
is that a command to run somewhere?
Ben West
@bewest
Feb 09 2016 05:16
hmm, it's a tool
it takes a bunch of the previous oref0/mm reports and compiles a profile for NS
need more tools for this to make it easier
what version of openaps are you using openaps --version?
Jim Matheson
@jmatheson
Feb 09 2016 05:17
If only my 2000mAh battery lasted more than 3 hrs i'd be fully murseless.
Aaron Michelson
@Aaybob
Feb 09 2016 05:17
I'm not a programmer but I took five or so programming classes in college in the 80s and early 90s
openaps 0.0.9
Jim Matheson
@jmatheson
Feb 09 2016 05:19
Once I figure out how to get battery life out of 2000mAh this setup will be fully murseless
Dragan Maksimovic
@dm61
Feb 09 2016 05:19
@Aaybob would you mind if I took a look at your NS site?
Aaron Michelson
@Aaybob
Feb 09 2016 05:19
no
Jim Matheson
@jmatheson
Feb 09 2016 05:19
IMG_2622.jpg
Aaron Michelson
@Aaybob
Feb 09 2016 05:19
its 2016CGM.az..
Ben West
@bewest
Feb 09 2016 05:20
is that it!?
next to the phone?
Dragan Maksimovic
@dm61
Feb 09 2016 05:20
ok, @Aaybob let me take a look at it and will let you know if I see anything
Jim Matheson
@jmatheson
Feb 09 2016 05:20
Yup
Ben West
@bewest
Feb 09 2016 05:23
no murse needed, indeed
need bigger battery I think
Dragan Maksimovic
@dm61
Feb 09 2016 05:25
@Aaybob I can see your NS site - looks ok to me. The version is 0.9.0-dev, which is the same as mine. Your basal profile shows up. I think you should be able to upload openAPS stuff.
Jim Matheson
@jmatheson
Feb 09 2016 05:25
Yea. It seems that way at least until i can get away from a usb connected ti.
Which is also along with the connector the most bulky part or the setup
Aaron Michelson
@Aaybob
Feb 09 2016 05:29
That's what I thought! Did you see the OpenAPS unknown under the 106 IOB and time?
Jason Calabrese
@jasoncalabrese
Feb 09 2016 05:30
@JaysonEwer, yes the school district wanted it in the official Dr's orders
Dragan Maksimovic
@dm61
Feb 09 2016 05:31
@Aaybob yes, your OpenAPS pill shows unknown, probably because you have not uploaded anything
Jason Calabrese
@jasoncalabrese
Feb 09 2016 05:32
they've been fighting to get me into MM's system to order a meter, I already got 1 on ebay, but getting the meter from them and some supplies will be nice once I we use up all the stuff we got from people in the local T1 group
Aaron Michelson
@Aaybob
Feb 09 2016 05:33
Okay, hmmm, maybe just wait for google and microsoft to decide on the future of the internet?
Jason Calabrese
@jasoncalabrese
Feb 09 2016 05:33
insurance in giving us a hard time on the contour strips
maybe try firefox or something?
Aaron Michelson
@Aaybob
Feb 09 2016 05:34
I'm in it on my iphone safari
We're thinking the issues are unrelated. Do some people just breeze into a working system?
Ben West
@bewest
Feb 09 2016 05:39
depending on your version of working system...
no.
I'm working fervently on making this easier
the NS upload stuff is the hardest/worst part
I wrote a lot of it, and I can barely follow it
Jason Calabrese
@jasoncalabrese
Feb 09 2016 05:40
when you put all the parts together it's a big system
lots of options and customization
Ben West
@bewest
Feb 09 2016 05:40
needs to be a tool, setup-ns-upload --type treatments | openaps import
yeah, you can get glucose from almost dozen places
Jason Calabrese
@jasoncalabrese
Feb 09 2016 05:41
I think it's also a process, need to do the steps get some parts working and iterate
Ben West
@bewest
Feb 09 2016 05:41
that's part of it
but other part is being able to reproduce it
Jason Calabrese
@jasoncalabrese
Feb 09 2016 05:42
I'm running stuff I setup 2 years ago
Ben West
@bewest
Feb 09 2016 05:42
exposition, communication all that :-D
Aaron Michelson
@Aaybob
Feb 09 2016 05:42
yeah big...I was thinking it was beyond doable when I got caught in the weeds of upload to NS stuff.
Jason Calabrese
@jasoncalabrese
Feb 09 2016 05:42
couldn't do it all at once quikly
Ben West
@bewest
Feb 09 2016 05:42
the import tool will help a lot
for loops where we have a standard and know what it looks like, we can run simple script that collects just your serials and api secret/url
which style loop do you want? we have a bunch of kind of known options now
everything needs to run through tz, units, etc
Aaron Michelson
@Aaybob
Feb 09 2016 05:44
I do have a nice log file system now and improved stability
Jim Matheson
@jmatheson
Feb 09 2016 05:45
no breeze here, but a sense of accomplishment :)
Aaron Michelson
@Aaybob
Feb 09 2016 05:45
my loop is:
loop ! bash -c "( openaps preflight && openaps gather && openaps enact) || echo IF PREFLIGHT WAS OK THEN ENACT OR GATHER ERROR" ohhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh I thought I had updated that...................
Jim Matheson
@jmatheson
Feb 09 2016 05:45
and still struggling
Aaron Michelson
@Aaybob
Feb 09 2016 05:47
to:
loop ! bash -c "( (openaps preflight && openaps gather && openaps enact) || echo LOOP FAIL) && (openaps status-upload || echo STATUS-UPLOAD FAIL))"
Ben West
@bewest
Feb 09 2016 05:47
need more people to learn python :-)
Dragan Maksimovic
@dm61
Feb 09 2016 05:47
@Aaybob I struggled with NS upload a lot. I tried to record what I did in the dev version of the visualization section. I am still not sure everything is correct there, and I am pretty sure there are better ways to do it - but you can take a look
Ben West
@bewest
Feb 09 2016 05:48
or maybe compile list of needed uses... eg set_temp_basal can/should query history afterwards and report real mm-history
tool to generate ns-monitor loop
tool to generate oref0 loop, tool to generate $x-shaped loop
Aaron Michelson
@Aaybob
Feb 09 2016 05:53
I think its about to work, I just hadn't yet issued a status-upload command.
Well, how about just a unix auk grep ps to format the log files...
Dragan Maksimovic
@dm61
Feb 09 2016 05:54
First try to do it manually, check that you see it on NS, then do the cron
Aaron Michelson
@Aaybob
Feb 09 2016 06:16
Yes thanks, manually I have errors
openaps status-upload
ns-status://JSON/shell/monitor/upload-status.json
monitor/upload-status.json raised 'NoneType' object has no attribute 'split'
Traceback (most recent call last):
and the error messages are not as informative as what I'm used to
The other parts of this project had error messages that made the fix obvious
Jason Calabrese
@jasoncalabrese
Feb 09 2016 06:25
something isn't getting echo'd
Aaron Michelson
@Aaybob
Feb 09 2016 06:29
I'm just getting ahead of myself. Sorry for all the problems. I'm zeroing in now.
Aaron Michelson
@Aaybob
Feb 09 2016 06:57

This report "upload-status.json" uses a device "ns-status". I can't invoke the report. They look like this:

[report "monitor/upload-status.json"]
device = ns-status
use = shell
clock = monitor/clock-zoned.json
iob = monitor/iob.json
suggested = enact/suggested.json
enacted = enact/enacted.json
battery = monitor/read_battery_status.json
reservoir = monitor/reservoir.json
status = monitor/read_status.json
reporter = JSON

[device "ns-status"]
fields = clock iob suggested enacted battery reservoir status
cmd = ns-status
vendor = openaps.vendors.process
agrs =

all of the sub reports are present with good data
Dragan Maksimovic
@dm61
Feb 09 2016 07:08
I do not see anything wrong here. So, what happens when you try to invoke upload-status.json report?
Aaron Michelson
@Aaybob
Feb 09 2016 07:08
The error message reads: <file>.json raised 'NoneType' opject has no attribute 'split'. So there is a missing argument, yes? I don't know how the arguments get passed to the associated subroutines. Can others verify that the "upload-status.json" shown above is correct and that the process it calls is correct? Thanks for helping with this.
Oh, didn't see you'd written
Ben West
@bewest
Feb 09 2016 07:09
you can take the report apart and use the openaps use commands
Aaron Michelson
@Aaybob
Feb 09 2016 07:09
pi@raspberrypi:~/pancreas/First-commit $ openaps report invoke monitor/upload_status.json
ns-status://JSON/shell/monitor/upload_status.json
monitor/upload_status.json raised 'NoneType' object has no attribute 'split'
Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 4, in <module>
import('pkg_resources').run_script('openaps==0.0.9', 'openaps-report')
File "/usr/local/lib/python2.7/dist-packages/setuptools-19.6-py2.7.egg/pkgresources/_init.py", line 727, in run_script
File "/usr/local/lib/python2.7/dist-packages/setuptools-19.6-py2.7.egg/pkgresources/_init.py", line 1652, in run_script
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/EGG-INFO/scripts/openaps-report";, line 82, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/openaps/cli/init.py", line 44, in call
self.run(self.args)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/EGG-INFO/scripts/openaps-report";, line 75, in run
output = app(args, self)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/openaps/cli/subcommand.py";, line 50, in call
return self.method.main(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/openaps/reports/invoke.py";, line 40, in main
output = task.method(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/openaps/uses/use.py";, line 45, in call
output = self.main(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/openaps/vendors/process.py";, line 73, in main
command.extend(info.get('args').split(' '))
AttributeError: 'NoneType' object has no attribute 'split'
pi@raspberrypi:~/pancreas/First-commit $
Ben West
@bewest
Feb 09 2016 07:09
the report fields map to the parameters in the use command for the device
Aaron Michelson
@Aaybob
Feb 09 2016 07:10
sure,,,I was just thinking the error message said there is a missing attribute 'split'
Ben West
@bewest
Feb 09 2016 07:11
the reason they match up is so you can do the use command
and copy/paste the report add JSON ...
and it's the same exact thing
so you should be able to do the reverse
create the use command and try it out
I can tell from the inis all the fields have values basically
fields = clock iob suggested enacted battery reservoir status
your fields were:
clock = monitor/clock-zoned.json
iob = monitor/iob.json
suggested = enact/suggested.json
enacted = enact/enacted.json
battery = monitor/read_battery_status.json
reservoir = monitor/reservoir.json
status = monitor/read_status.json
so openaps use ns-status shell monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json is what it runs when you invoke that report
so each of those files should exist
if you discover some difference in behavior and are sure everything is otherwise the same, it's a bug
see how the field map to the field names in the use?
Ben West
@bewest
Feb 09 2016 07:16
cat monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json | json -g as a quick sanity check for all the files
should compile them all into one giant thing
or print some error if one isn't proper syntax or something
Aaron Michelson
@Aaybob
Feb 09 2016 07:17
trying the first thing:
pi@raspberrypi:~/pancreas/First-commit $ openaps use ns-status shell monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json
Traceback (most recent call last):
File "/usr/local/bin/openaps-use", line 4, in <module>
import('pkg_resources').run_script('openaps==0.0.9', 'openaps-use')
File "/usr/local/lib/python2.7/dist-packages/setuptools-19.6-py2.7.egg/pkgresources/_init.py", line 727, in run_script
File "/usr/local/lib/python2.7/dist-packages/setuptools-19.6-py2.7.egg/pkgresources/_init.py", line 1652, in run_script
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/EGG-INFO/scripts/openaps-use";, line 63, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/openaps/cli/init.py", line 44, in call
self.run(self.args)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/EGG-INFO/scripts/openaps-use";, line 57, in run
output = app(args, self)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/openaps/uses/init.py", line 92, in call
return self.method.selected(args)(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/openaps/uses/init.py", line 31, in call
return self.method(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/openaps/uses/use.py";, line 45, in call
output = self.main(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.0.9-py2.7.egg/openaps/vendors/process.py";, line 73, in main
command.extend(info.get('args').split(' '))
AttributeError: 'NoneType' object has no attribute 'split'
Ben West
@bewest
Feb 09 2016 07:19
hmmm
ok
same thing with ns-status tool without any thing else
ns-status monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json
btw
that should give a better error
btw, on my machine:
$ ns-status 
usage:  [ 'node', '/usr/local/bin/ns-status' ] <clock.json> <iob.json> <suggested.json> <enacted.json> <battery.json> <reservoir.json> <status.json>
bewest@bewest-MacBookPro:~/Documents/openaps$
this suggests another field after reservoir named status.json
can you update your openaps
sudo easy_install -ZU openaps
then openaps device show --json ns-status | json
the update is needed for the --json option
yeah

bewest@bewest-MacBookPro:~/Documents/openaps$ ns-status monitor/clock.json oref0-predict/iob.json  oref0-predict/oref0.json  oref0-predict/enacted.json  monitor/battery.json monitor/reservoir.json 
usage:  [ 'node', '/usr/local/bin/ns-status' ] <clock.json> <iob.json> <suggested.json> <enacted.json> <battery.json> <reservoir.json> <status.json>
bewest@bewest-MacBookPro:~/Documents/openaps$ ns-status monitor/clock.json oref0-predict/iob.json  oref0-predict/oref0.json  oref0-predict/enacted.json  monitor/battery.json monitor/reservoir.json monitor/status.json 
{"openaps":{"iob":{"iob":3.16719809199125,"activity":0.02442957070707071,"bolusiob":1.64175806324,"timestamp":"2016-01-25T05:11:41.000Z"},"suggested":{"temp":"absolute","bg":212,"tick":-1,"eventualBG":87,"snoozeBG":198,"reason":"Eventual BG 87<106bolus snooze: eventual BG range 87-198","timestamp":"2016-01-25T05:11:41.000Z"},"enacted":{"bg":235,"temp":"absolute","snoozeBG":254,"recieved":true,"rate":3.275,"reason":"max_iob 2, adj. req. rate:5.1 to maxSafeBasal:3.3, no temp, setting 3.3000000000000003U/hr","eventualBG":254,"timestamp":"2016-01-11T20:45:50.000Z","duration":30,"tick":"+5"}},"pump":{"clock":"2016-01-24T21:07:36","battery":{"status":"normal","voltage":1.56},"reservoir":114.3,"status":{"status":"normal","bolusing":false,"suspended":false,"timestamp":"2016-01-25T05:11:35.000Z"}}}
bewest@bewest-MacBookPro:~/Documents/openaps$
when I try it without all the openaps window dressing....
Aaron Michelson
@Aaybob
Feb 09 2016 07:23

The monitor directory gets cleared every five minutes....but this is the result of what is there now

pi@raspberrypi:~/pancreas/First-commit $ cat monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json
cat: monitor/clock-zoned.json: No such file or directory
{
"bolusiob": 0,
"iob": -0.19718284329573033,
"activity": -0.0018874308333333332
}{"temp":"absolute","bg":162,"tick":"+15","eventualBG":210,"snoozeBG":210,"reason":"Eventual BG 210>120, adj. req. rate:11.3 to maxSafeBasal:2.4, temp 2.375 >~ req 2.4000000000000004U/hr"}
{
"bg": 117,
"temp": "absolute",
"snoozeBG": 161,
"recieved": true,
"rate": 2.375,
"reason": "Eventual BG 161>120, adj. req. rate:6.4 to maxSafeBasal:2.4, temp 0<2.4000000000000004U/hr",
"eventualBG": 161,
"timestamp": "2016-02-08T23:00:55.880277",
"duration": 30,
"tick": "+10"
}cat: monitor/read_battery_status.json: No such file or directory
cat: monitor/reservoir.json: No such file or directory
pi@raspberrypi:~/pancreas/First-commit $

Ben West
@bewest
Feb 09 2016 07:23
oh ok
you have no reservoir/battery
I mean all those errors are telling us what's missing
Aaron Michelson
@Aaybob
Feb 09 2016 07:24
I did have but I think cron deleted it in loop
Ben West
@bewest
Feb 09 2016 07:25
basically all those files plus the pump status have to exist
you need to add pumndp status ot e
to the end
if you see my exampleus above, it errors without the pump stat
pump status
and with the pump status it's ok
this mouse/trackpad is way to "hot"/sensitive
it keeps moving cursor while typing
with my file names here's the same thing using the cat exercise
Ben West
@bewest
Feb 09 2016 07:30
$ cat monitor/clock.json oref0-predict/iob.json  oref0-predict/oref0.json  oref0-predict/enacted.json  monitor/battery.json monitor/reservoir.json monitor/status.json  | json -g
json: error: input is not JSON: Syntax error at line 1, column 22:
        "2016-01-24T21:07:36"{"iob":3.16719809199125,"activity":0.02442957070707071,"bolusiob":1.64175806324}
        .....................^
"2016-01-24T21:07:36"{"iob":3.16719809199125,"activity":0.02442957070707071,"bolusiob":1.64175806324}
{"temp":"absolute","bg":212,"tick":-1,"eventualBG":87,"snoozeBG":198,"reason":"Eventual BG 87<106bolus snooze: eventual BG range 87-198"}
{ 
  "bg": 235,
  "temp": "absolute",
  "snoozeBG": 254,
  "recieved": true,
  "rate": 3.275,
  "reason": "max_iob 2, adj. req. rate:5.1 to maxSafeBasal:3.3, no temp, setting 3.3000000000000003U/hr",
  "eventualBG": 254,
  "timestamp": "2016-01-11T12:45:50.297262",
  "duration": 30,
  "tick": "+5"
}{
  "status": "normal",
  "voltage": 1.56
}114.3{
  "status": "normal",
  "bolusing": false,
  "suspended": false
}
bewest@bewest-MacBookPro:~/Documents/openaps$
hehe, ok slightly different error, but all the files exist
Aaron Michelson
@Aaybob
Feb 09 2016 07:32
I'm at the ns-status and its not an openaps command
Ben West
@bewest
Feb 09 2016 07:32
without the shell word?
ns-status ...
no openaps stuff
if you having missing files, no chance ns-status will work, first resolve the cat errors from file not found/missing
you can cp the needed files to a tmp directory just to play with ns-status directly
if I type just ns-status, I get this message:
$ ns-status
usage:  [ 'node', '/usr/local/bin/ns-status' ] <clock.json> <iob.json> <suggested.json> <enacted.json> <battery.json> <reservoir.json> <status.json>
bewest@bewest-MacBookPro:~/Documents/openaps$
tells me I need the clock, iob, all those things
should be able to cat all of them without any errors
if that works, replacing cat with ns-status should work
Aaron Michelson
@Aaybob
Feb 09 2016 07:36
:~/pancreas/First-commit $ cat monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json
cat: monitor/clock-zoned.json: No such file or directory
{
"bolusiob": 0,
"iob": 0.23208144683149973,
"activity": -0.001930069722222222
}{"temp":"absolute","bg":186,"tick":"+6","eventualBG":204,"snoozeBG":204,"reason":"Eventual BG 204>120, adj. req. rate:10.7 to maxSafeBasal:2.4, temp 2.375 >~ req 2.4000000000000004U/hr"}
{
"bg": 117,
"temp": "absolute",
"snoozeBG": 161,
"recieved": true,
"rate": 2.375,
"reason": "Eventual BG 161>120, adj. req. rate:6.4 to maxSafeBasal:2.4, temp 0<2.4000000000000004U/hr",
"eventualBG": 161,
"timestamp": "2016-02-08T23:00:55.880277",
"duration": 30,
"tick": "+10"
}{
"status": "normal",
"voltage": 1.21
}93.0pi@raspberrypi:~/pancreas/First-commit $
Ben West
@bewest
Feb 09 2016 07:37
missing clock-zoned.json
otherwise all good
Aaron Michelson
@Aaybob
Feb 09 2016 07:37
pi@raspberrypi:~/pancreas/First-commit $ it $ ns-status monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json
cat: mns-status: No such file or directory
cat: monitor/clock-zoned.json: No such file or directory
{
"bolusiob": 0,
"iob": 0.23208144683149973,
"activity": -0.001930069722222222
}{"temp":"absolute","bg":186,"tick":"+6","eventualBG":204,"snoozeBG":204,"reason":"Eventual BG 204>120, adj. req. rate:10.7 to maxSafeBasal:2.4, temp 2.375 >~ req 2.4000000000000004U/hr"}
{
"bg": 117,
"temp": "absolute",
"snoozeBG": 161,
"recieved": true,
"rate": 2.375,
"reason": "Eventual BG 161>120, adj. req. rate:6.4 to maxSafeBasal:2.4, temp 0<2.4000000000000004U/hr",
"eventualBG": 161,
"timestamp": "2016-02-08T23:00:55.880277",
"duration": 30,
"tick": "+10"
}{
"status": "normal",
"voltage": 1.21
}93.0p
Ben West
@bewest
Feb 09 2016 07:37
also, need another file for ns-status
the pump-status
use pump read_status
need a report for that
maybe openaps report add pump monitor/pump-status.json JSON pump read_status
openaps report invoke monitor/pump-status.json
(need to add that to one of your aliases)
also you have no clock-zoned.json, not sure what's up there
do you ahve a clock-raw.json vs a clock.json?
we used to recommend read_clock.json -> clock.json and then figured out we also needed a zoned version so we added -zoned.json after
now we know better so we're starting to recommend saving the first one as -raw.json and the second one as just normal .json
this should be the result of running the tz tool against the clock
Aaron Michelson
@Aaybob
Feb 09 2016 07:43
pi@raspberrypi:~/pancreas/First-commit $ openaps report invoke monitor/read_status.json
pumpbuilding://JSON/read_status/monitor/read_status.json
reporting monitor/read_status.json
pi@raspberrypi:~/pancreas/First-commit $ cat monitor/read_status.json
{
"status": "normal",
"bolusing": false,
"suspended": false
}pi@raspberrypi:~/pancreas/First-commit $
I think this is pump status
]pi@raspberrypi:~/pancreas/First-commit $ cat monitor/clock-zoned.json
"2016-02-08T23:28:44-08:00"pi@raspberrypi:~/pancreas/First-commit $
And this is clock-zoned.json
Ben West
@bewest
Feb 09 2016 07:45
so giving all those things to ns-status should work
Aaron Michelson
@Aaybob
Feb 09 2016 07:49
Same eror
Ben West
@bewest
Feb 09 2016 07:49
that means the file doesn't exist then
should work with cat same thing
if cat complains, something else will too
you added monitor/read_status.json to the end?
Aaron Michelson
@Aaybob
Feb 09 2016 07:50
$ cat shell monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json
cat: shell: No such file or directory
"2016-02-08T23:28:44-08:00"{
"bolusiob": 0,
"iob": 0.23208144683149973,
"activity": -0.001930069722222222
}{"temp":"absolute","bg":186,"tick":"+6","eventualBG":204,"snoozeBG":204,"reason":"Eventual BG 204>120, adj. req. rate:10.7 to maxSafeBasal:2.4, temp 2.375 >~ req 2.4000000000000004U/hr"}
{
"bg": 117,
"temp": "absolute",
"snoozeBG": 161,
"recieved": true,
"rate": 2.375,
"reason": "Eventual BG 161>120, adj. req. rate:6.4 to maxSafeBasal:2.4, temp 0<2.4000000000000004U/hr",
"eventualBG": 161,
"timestamp": "2016-02-08T23:00:55.880277",
"duration": 30,
"tick": "+10"
}{
"status": "normal",
"voltage": 1.21
}93.0{
"status": "normal",
"bolusing": false,
"suspended": false
}pi@raspberrypi:~/pancreas/First-commit $
Ben West
@bewest
Feb 09 2016 07:50
if it was deleted, it needs to be automated again
ah
get rid of shell
that shell keyword is a wart, sorry
it only applies to openaps commands using the process vendor
that's a technical thing, sorry
should work with ns-status monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json
btw, you can try pressing the tab key while typing these commands
Aaron Michelson
@Aaybob
Feb 09 2016 07:54
$ openaps use ns-status monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json
usage: openaps-use ns-status [-h] USAGE ...
openaps-use ns-status: error: argument USAGE: invalid choice: 'monitor/clock-zoned.json' (choose from 'shell')
Ben West
@bewest
Feb 09 2016 07:54
on rpi it takes a few long moments, but it will print matching suggestions and provide completions for partially typed words
no
so
so first test is cat
second test is tool itself
third test is in openaps with the use commands
fourth test is with reports
fifth test is in loop
so, with cat:
Aaron Michelson
@Aaybob
Feb 09 2016 07:55
$ cat monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json
"2016-02-08T23:28:44-08:00"{
"bolusiob": 0,
"iob": 0.23208144683149973,
"activity": -0.001930069722222222
}{"temp":"absolute","bg":186,"tick":"+6","eventualBG":204,"snoozeBG":204,"reason":"Eventual BG 204>120, adj. req. rate:10.7 to maxSafeBasal:2.4, temp 2.375 >~ req 2.4000000000000004U/hr"}
{
"bg": 117,
"temp": "absolute",
"snoozeBG": 161,
"recieved": true,
"rate": 2.375,
"reason": "Eventual BG 161>120, adj. req. rate:6.4 to maxSafeBasal:2.4, temp 0<2.4000000000000004U/hr",
"eventualBG": 161,
"timestamp": "2016-02-08T23:00:55.880277",
"duration": 30,
"tick": "+10"
}{
"status": "normal",
"voltage": 1.21
}93.0{
"status": "normal",
"bolusing": false,
"suspended": false
}pi@raspberrypi:~/pancreas/First-commit $
Ben West
@bewest
Feb 09 2016 07:56
right
then we can use ns-status directly, no openaps stuff, just repace cat with ns-status
so ns-status monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json
no openaps stuff in there
Aaron Michelson
@Aaybob
Feb 09 2016 07:57
$ ns-status monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json
{"openaps":{"iob":{"bolusiob":0,"iob":0.23208144683149973,"activity":-0.001930069722222222,"timestamp":"2016-02-09T07:30:36.000Z"},"suggested":{"temp":"absolute","bg":186,"tick":"+6","eventualBG":204,"snoozeBG":204,"reason":"Eventual BG 204>120, adj. req. rate:10.7 to maxSafeBasal:2.4, temp 2.375 >~ req 2.4000000000000004U/hr","timestamp":"2016-02-09T07:30:50.000Z"},"enacted":{"bg":117,"temp":"absolute","snoozeBG":161,"recieved":true,"rate":2.375,"reason":"Eventual BG 161>120, adj. req. rate:6.4 to maxSafeBasal:2.4, temp 0<2.4000000000000004U/hr","eventualBG":161,"timestamp":"2016-02-09T07:00:55.000Z","duration":30,"tick":"+10"}},"pump":{"clock":"2016-02-08T23:28:44-08:00","battery":{"status":"normal","voltage":1.21},"reservoir":93,"status":{"status":"normal","bolusing":false,"suspended":false,"timestamp":"2016-02-09T07:41:57.000Z"}}}
pi@raspberrypi:~/pancreas/First-commit $
Ben West
@bewest
Feb 09 2016 07:57
once that works we can try inside openaps
openaps use ns-status shell monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json
here we add the keyword
if that works, that's the report we want to save
Aaron Michelson
@Aaybob
Feb 09 2016 07:58
shell?
Ben West
@bewest
Feb 09 2016 07:58
yeah, shell is for openaps
when we use an external process from outside of openaps inside of openaps, we tell it to do so using the special shell command
use external-virtual-device-aka-process shell <args...> is one way to read it inside openaps
Aaron Michelson
@Aaybob
Feb 09 2016 07:59
fails inside openaps
Ben West
@bewest
Feb 09 2016 08:00
with same exact args?
Aaron Michelson
@Aaybob
Feb 09 2016 08:00
$ openaps use ns-status shell monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json Traceback (most recent call last):
Ben West
@bewest
Feb 09 2016 08:01
openaps device show --ini ns-status?
this is on 0.0.10
Aaron Michelson
@Aaybob
Feb 09 2016 08:01
but I never started up a git of new openaps like you said once
Ben West
@bewest
Feb 09 2016 08:01
sudo easy_install -ZU openaps to get 0.0.10
need to see the traceback
above you pasted or typed it said agrs
instead of args
or similar
if that's actually in the ini it could cause problems
cat ns-status.ini
Aaron Michelson
@Aaybob
Feb 09 2016 08:02
lol
Ben West
@bewest
Feb 09 2016 08:03
that's impossible afaik using just the tools
so ;-)
I chalked it up to typo
Aaron Michelson
@Aaybob
Feb 09 2016 08:03
$ openaps use ns-status shell monitor/clock-zoned.json monitor/iob.json enact/suggested.json enact/enacted.json monitor/read_battery_status.json monitor/reservoir.json monitor/read_status.json
{
"pump": {
"battery": {
"status": "normal",
"voltage": 1.21
},
"status": {
"status": "normal",
"timestamp": "2016-02-09T07:41:57.000Z",
"bolusing": false,
"suspended": false
},
"reservoir": 93,
"clock": "2016-02-08T23:28:44-08:00"
},
"openaps": {
"suggested": {
"bg": 186,
"temp": "absolute",
"snoozeBG": 204,
"timestamp": "2016-02-09T07:30:50.000Z",
"reason": "Eventual BG 204>120, adj. req. rate:10.7 to maxSafeBasal:2.4, temp 2.375 >~ req 2.4000000000000004U/hr",
"eventualBG": 204,
"tick": "+6"
},
"iob": {
"timestamp": "2016-02-09T07:30:36.000Z",
"activity": -0.001930069722222222,
"iob": 0.23208144683149973,
"bolusiob": 0
},
"enacted": {
"bg": 117,
"temp": "absolute",
"snoozeBG": 161,
"recieved": true,
"reason": "Eventual BG 161>120, adj. req. rate:6.4 to maxSafeBasal:2.4, temp 0<2.4000000000000004U/hr",
"rate": 2.375,
"eventualBG": 161,
"timestamp": "2016-02-09T07:00:55.000Z",
"duration": 30,
"tick": "+10"
}
}
}pi
Ben West
@bewest
Feb 09 2016 08:03
but you were missing several files anyway, so disregarded until now
oh
there you go
:-D
:sparkles: ]
Aaron Michelson
@Aaybob
Feb 09 2016 08:04
wow simple as that
Ben West
@bewest
Feb 09 2016 08:04
yeah, each layer builds on the next
cat can be used to debug the files/args
because each tool generally tries to look up data in the file
so cat is good smoke test to see if there's anything there
then run the tool itself
then try it in side openaps use
then save it as report
in future we'll have a generate-recipe tool that uses the openaps import tool to build these things more reliably
for now this is personalized DIY to extreme ;-)
as we build up more knowledge, it'll make it easier to automate in the future
Aaron Michelson
@Aaybob
Feb 09 2016 08:08
next I execute status-upload with arguments?
Ben West
@bewest
Feb 09 2016 08:09
yeah, each layer builds on last
but it all works same way
so yeah, you can try status-upload outside of openaps to see what happens
although, not sure what it's actually called
openaps device show status-upload maybe
Aaron Michelson
@Aaybob
Feb 09 2016 08:11
$ openaps alias show
status-upload ! bash -c "openaps report invoke monitor/upload-status.json && (openaps use ns-upload shell devicestatus.json monitor/upload-status.json )"
Ben West
@bewest
Feb 09 2016 08:12
ah ok
hmm
Aaron Michelson
@Aaybob
Feb 09 2016 08:13
$ openaps status-upload
ns-status://JSON/shell/monitor/upload-status.json
reporting monitor/upload-status.json
usage: openaps-use [-h] [--format {text,json,base,stdout}] [--output OUTPUT]
[--version]
device ...
openaps-use: error: argument device: invalid choice: 'ns-upload' (choose from u'NS_UpLoad', u'calculate-iob', u'cgmbuilding', u'determine-basal', u'get-profile', u'ns-status', u'oref0', u'pumpbuilding', u'tz')
Ben West
@bewest
Feb 09 2016 08:13
hehe
ok, you named yours NS_UpLoad
case matters
Aaron Michelson
@Aaybob
Feb 09 2016 08:13
lol
Ben West
@bewest
Feb 09 2016 08:13
see in the alias it says ns-upload
try openaps use NS_UpLoad shell devicestatus.json monitor/upload-status.json to test manually once
if that works, change the ini to match
the alias
Aaron Michelson
@Aaybob
Feb 09 2016 08:16
$ openaps status-upload
ns-status://JSON/shell/monitor/upload-status.json
reporting monitor/upload-status.json
[
{
"created_at": "2016-02-09T08:16:25.567Z",
"pump": {
"battery": {
"status": "normal",
"voltage": 1.21
},
"status": {
"status": "normal",
"timestamp": "2016-02-09T07:41:57.000Z",
"bolusing": false,
"suspended": false
},
"reservoir": 93,
"clock": "2016-02-08T23:28:44-08:00"
},
"_id": "56b9a0597f8fc23412c1d347",
"openaps": {
"suggested": {
"bg": 186,
"temp": "absolute",
"snoozeBG": 204,
"timestamp": "2016-02-09T07:30:50.000Z",
"reason": "Eventual BG 204>120, adj. req. rate:10.7 to maxSafeBasal:2.4, temp 2.375 >~ req 2.4000000000000004U/hr",
"eventualBG": 204,
"tick": "+6"
},
"iob": {
"timestamp": "2016-02-09T07:30:36.000Z",
"bolusiob": 0,
"iob": 0.23208144683149973,
"activity": -0.001930069722222222
},
"enacted": {
"bg": 117,
"temp": "absolute",
"snoozeBG": 161,
"recieved": true,
"rate": 2.375,
"reason": "Eventual BG 161>120, adj. req. rate:6.4 to maxSafeBasal:2.4, temp 0<2.4000000000000004U/hr",
"eventualBG": 161,
"timestamp": "2016-02-09T07:00:55.000Z",
"duration": 30,
"tick": "+10"
}
}
}
]p
Good? So it'll upload to NS?
A typo and a forgotten case change....
Ben West
@bewest
Feb 09 2016 08:19
pretty sure you just manually uploaded to NS
if you look at NS pretty sure it's showing something from what you just uploaded
I'm at 88%
Aaron Michelson
@Aaybob
Feb 09 2016 08:20
Docs said it wont take on the first ,,,,has to do it twice before something shows up
Ben West
@bewest
Feb 09 2016 08:20
and missing arg
so do it again :-)
you're in control here
you gotta prime that sucker
Aaron Michelson
@Aaybob
Feb 09 2016 08:21
openaps just changed on nightscout...to 50m ago from unknown
Ben West
@bewest
Feb 09 2016 08:22
might need loop to run again to refresh data
did you include pump-status somewhere in the loop aliases?
should be able to do all that manually a few times, dunno when the next loop will run
I've got a very very simple loop, just runs everything in order every 5 min
I don't even attempt any auto-repair/fix/reset stuff because I want the pain
Aaron Michelson
@Aaybob
Feb 09 2016 08:24
just reran loop now batteries are all going, nightscout shows 15% battery and Openaps went back to unknown status after running loop
Ben West
@bewest
Feb 09 2016 08:24
well, no errors ;-)
Aaron Michelson
@Aaybob
Feb 09 2016 08:25
now status-upload erred.
Ben West
@bewest
Feb 09 2016 08:25
I'm not even using openaps plugin yet
Aaron Michelson
@Aaybob
Feb 09 2016 08:25
I have in loop rm monitor/*
Ben West
@bewest
Feb 09 2016 08:25
that seems like a problem to me
why remove everything?
Aaron Michelson
@Aaybob
Feb 09 2016 08:26
thanks though now i know what to fix
It was suggested to do in the documentation
that way it won't run on old data was the reason
Ben West
@bewest
Feb 09 2016 08:27
hmm
seems like right concern, wrong solution
Aaron Michelson
@Aaybob
Feb 09 2016 08:27
thanks again I'm taking a break now
Ben West
@bewest
Feb 09 2016 08:27
hmm, my loop doesn't do that, but mine's offline often ;-)
pros and cons
yeah, have a good night, congrats on milestone!
Aaron Michelson
@Aaybob
Feb 09 2016 08:28
I like the raw data that shows possible BG in the nex half hour
milestone
Ben West
@bewest
Feb 09 2016 08:33
do you have treatments posting to NS?
hmm, I think that's harder :(
but much more useful
well, I have no basis for saying that
Matthias
@ceben80
Feb 09 2016 11:30
Please can anybody help in how to create "max_iob.json"?
Andrew Warrington
@andrew-warrington
Feb 09 2016 14:27
@ceben80, you can just create it manually. text file. content looks like this:
{
  "max_iob": 8
}
(my own usually has lower than 8, just using that for testing at the moment.)
Matthias
@ceben80
Feb 09 2016 14:30
thx
Scott Leibrand
@scottleibrand
Feb 09 2016 14:31
I wouldn't go any higher than 2 initially.
Maybe lower at first.
Matthias
@ceben80
Feb 09 2016 15:32
@scottleibrand oki, thank you :+1:
Jason Calabrese
@jasoncalabrese
Feb 09 2016 15:54
last night I got autossh setup on my edisons, based on a tip from @oskarpearson, with that I have a reverse ssh tunnel from the rig to my server that works over cell and wifi networks. Now I'm ssh'd to the rig as it's on the way to school. Setup was pretty easy, if anyone is interested let me know.
I used https://raymii.org/s/tutorials/Autossh_persistent_tunnels.html for most of the info, but had to make a couple little changes
JaysonEwer
@JaysonEwer
Feb 09 2016 17:08
Hello All!! I'm very pleased to report my Endo was TOTALLY FLOORED when I showed her my Nightscout and OpenAPS and the 6.8 A1C made her smile; me too! :-D She was writing notes feverishly as I was telling her about it all. She said, "I could talk to you all morning about this!! She also said, how much does this cost, in case I wanted to share this with other people. :-D I'd say my Dr. is somewhat onboard. ;-) @danamlewis @scottleibrand @bewest @jasoncalabrese @dm61 @jmatheson @oskarpearson , et al... I've been looping for just about a month, slightly less, I don't even have the meal assist in place yet or the more aggressive algorithm. I can't wait for my next Endo appt in MAY!!! I'm very pleased with all this! THANK YOU!
My Dr. then proceeded to thank me for coming prepared to the appointment to show her all my gear and everything! :-D I was initially referred to her by my pump trainer 10 years ago due to the fact that they were very "cutting edge" with respect to the technology. ;-)
LadyViktoria
@LadyViktoria
Feb 09 2016 17:14
:+1:
Jim Matheson
@jmatheson
Feb 09 2016 17:18
Awseome, congrats!
JaysonEwer
@JaysonEwer
Feb 09 2016 17:20
Unfortunately, she only did the finger stick A1C, since I just had one a little over a month ago. :-/ Oh well, its still trending in the right direction.
Thank you, @LadyViktoria @jmatheson.
JaysonEwer
@JaysonEwer
Feb 09 2016 17:31
@jasoncalabrese Thanks for the note, by the way. I didn't see it before my Dr. appointment and the conversation that took place last night after my initial mention of the Dr. note made me not ask for one. I just wanted to get one incase anyone gave me a hard time. That way they are aware that my Dr. is aware of my personal decision to assume the "risk" involved in using a system like this. Perhaps next time i'll see what the doctor thinks after she does some more reading up on it. She has all the information she needs to build an OpenAPS if she wants too. :-D
Matthias
@ceben80
Feb 09 2016 17:38
@JaysonEwer - Sounds great! I still have a long way to reach looping, but I am highly motivated here
Oskar Pearson
@oskarpearson
Feb 09 2016 17:38
That's really great, @JaysonEwer - so glad to hear of that response :)
Jason Calabrese
@jasoncalabrese
Feb 09 2016 17:48
congrats @JaysonEwer, great to have the support of your Dr
Dana Lewis
@danamlewis
Feb 09 2016 18:17
@JaysonEwer congrats, sounds like an awesome day :)
Jim Matheson
@jmatheson
Feb 09 2016 19:27
Two days ago i was wondering why the heck @danamlewis and @scottleibrand had four rigs. Seemed crazy. Then today I am not blinking twice as I order my 3rd setup...
Jason Calabrese
@jasoncalabrese
Feb 09 2016 19:31
:+1:
Oskar Pearson
@oskarpearson
Feb 09 2016 19:45
Has anyone done any analysis of how much bandwidth the loop takes with nightscout? It seems like the number might be pretty high
(>6gb per month)
(Sighs as gitter mobile app loses some of the messages sent above)
Jason Calabrese
@jasoncalabrese
Feb 09 2016 19:52
it's not a good test
thats multiple devices using the hotspot
my wifes phone, etc
I'll optimize data use after everything else
Dana Lewis
@danamlewis
Feb 09 2016 19:58
lol @jmatheson more like 6 now ;) @scottleibrand
Scott Leibrand
@scottleibrand
Feb 09 2016 20:03
the back and forth with NS is pretty inefficient right now. we could do a lot of optimization with NS I think
Aaron Michelson
@Aaybob
Feb 09 2016 20:03
I like the NS website. I'm using OpenAPS 0.0.10 and status-upload runs error free. But on NS website there's no graph of basal as the pictures from others have shown. I had said I would not implement visualization but was strongly encouraged to do so by this group. There is a lot to like about NS but ought it be a part of the automatic pancreas project? I wish the people that said I must have that graph had said instead that it may cost hundreds of dollars consume as much time as the automatic pancreas project and when everything is complete it won't work.
dougtakeuti
@dougtakeuti
Feb 09 2016 20:06
Hi everyone, is the ´clock´ in the following expression correct?
´openaps report add monitor/pump-history-zoned.json JSON tz clock monitor/pumphistory.json´
Scott Leibrand
@scottleibrand
Feb 09 2016 20:08
no, I think you want rezone. check openaps use tz -h or similar
Dana Lewis
@danamlewis
Feb 09 2016 20:10
@Aaybob yes, I do believe it (or another visualization tool) should be part of OpenAPS. (Btw, the 'a' is for artificial :), as you know by now it's not automatic!). It's critical part of the DIY to be able to see what the system is doing. Unless you're willing to be constantly starting at server auditor on your phone and monitoring the system, the visualization is the best option for system monitoring for most users.
dougtakeuti
@dougtakeuti
Feb 09 2016 20:10
ok thanks
Scott Leibrand
@scottleibrand
Feb 09 2016 20:11
@Aaybob where are you getting the "hundreds of dollars" figure? agree the upload process needs to be simplified, probably to the point of having people run an improved version of ns-uploader-setup.sh first to get NS uploads working before they ever start on closing the loop. if you're interested in helping make that easier for future users, this is only gonna happen if someone is willing to put in the hundreds of hours required to get it done (on top of the thousands of hours we've collectively put in to get us to this point)
Dana Lewis
@danamlewis
Feb 09 2016 20:12
has @dm61 been able to take a look at your setup? He's done the most thorough review of the docs, so hoping he or someone else can give you (@aaybob) some pointers on where else to trouble shoot. (That's also a part of DIY. It's a work in progress, including the docs, and it requires patience and understanding that it's not easy, and it won't be perfect. But, hopefully, the end result is a better system that helps people better manage their diabetes than they could do on their own.)
Ben West
@bewest
Feb 09 2016 20:13
to see baslals, it has to be one of the enabled plugins
also need the profile setup
Jason Calabrese
@jasoncalabrese
Feb 09 2016 20:14
people will be paying thousands for commercial systems that won't be as good, don't see the need to optimize data usage first
but I'm sure there are more easy fixes that could help a lot
Dana Lewis
@danamlewis
Feb 09 2016 20:14
Keep in mind that this, and all the other places that people get stuck, are the same areas we've gotten stuck in before. In fact, when we started…other than @bewest's decocare, there was zero documentation and zero guides online for DIY APS. So, I totally get that it's frustrating, but sometimes context like this is helpful to remember that it's better than it was a year (or even three months!) ago, and as a result, can always get better with contributions from us, you & everyone else who want to pay it forward by helping improve the docs to make it easier for the next person.
Scott Leibrand
@scottleibrand
Feb 09 2016 20:18
One thing we don't want people to do is skip important parts of the system because they're "too hard", if those parts are important for safety. Monitoring the system and understanding what it's doing is safety critical, not a nice to have.
We can and should make those things easier, but need everyone's help to get there, and patience and willingness to work through things in the mean time
Jason Calabrese
@jasoncalabrese
Feb 09 2016 20:34
:+1:
Jim Matheson
@jmatheson
Feb 09 2016 20:40
yup. we need to streamline the parts that are not important. and there are plenty of those parts :)
Dana Lewis
@danamlewis
Feb 09 2016 20:43
hey @jmatheson & everyone else - did you get a chance to review https://github.com/openaps/docs/commit/8ac49f7b0516f145befd8c893b2eb5fb66677f23#diff-ce47a30eb7f9e453a3ca1c86bcf2f2fa as a potential new flow for the docs & process, speaking of stuff to streamline?
got some good feedback from @mccgm, @dm61, @audiefile so far. Will probably start fleshing this out further in the next couple of days, so any major comments/suggestions to change the flow would be great to hear in the next day or so.
Jim Matheson
@jmatheson
Feb 09 2016 20:44
nope, will check it out
Dana Lewis
@danamlewis
Feb 09 2016 20:46
thx
Dana Lewis
@danamlewis
Feb 09 2016 21:08
@JaysonEwer I like that idea (to teach how to PR). I was thinking we need a "I'm in the pipeline" list in addition to the "I'm looping" list, and cool to do it through a PR in the docs. :+1:
JaysonEwer
@JaysonEwer
Feb 09 2016 21:13
You/we could even have it auto-populate your form or have a step in your form have them go back, update the PR to include the "finish" date, which would be a good metric for how quickly folks are progressing through?
Jim Matheson
@jmatheson
Feb 09 2016 21:20
It would be helpful to have some sort of place to go for help and to interaction on each step of the process. could be a forum thread, channel in a slack group, etc. gitter digging sometimes feels like half the problem. the answer is in here somewhere but hard to find.
JaysonEwer
@JaysonEwer
Feb 09 2016 21:25
One other note that came out of my Dr visit today, My Dr. Noticed I was using much more basal insulin, more than the age old rule of thumb that your basal/bolus ratio is roughly 50/50. I think this may be somewhat of a paradigm shift in thinking about that old "rule-of-thumb". I'm not quite sure where the 50/50 "rule" even come from? I think we would all agree that the temp micro basals that are employed to maintain our targets would not exactly be classified as basal insulin, right? This may already be explained somewhere else, but it would be cool to be able to report on this phenomenon using NS. Perhaps it already can be reported on, but I haven't done that yet.
Looking back at my NS data, it's pretty apparent where I need to adjust my actual underlying basal rates, perhaps that's all I need to do to make the old rule of thumb work out better, if that is even a goal, not sure it really needs to be.
Jim Matheson
@jmatheson
Feb 09 2016 21:28
Yes. It’s sort of a 3rd category. I think it’s worth adjusting underlying when it’s consistently off.
Dana Lewis
@danamlewis
Feb 09 2016 21:30
@jmatheson to be honest, that's the idea of the docs. if a good discussion happens here…it should be transported to the docs and PRed in to the 'main source of truth' there.
Jim Matheson
@jmatheson
Feb 09 2016 21:31
sure. i’ve just noticed things are moving and changing fast.
Dana Lewis
@danamlewis
Feb 09 2016 21:32
@JaysonEwer not sure it would fit in with the form for my end tracking, but I think a list to PR yourself in to is great
@jmatheson exactly. someone's gotta document stuff :)
Jim Matheson
@jmatheson
Feb 09 2016 21:32
and some things are “I tried this and it worked” and not confirmed by the group. but we have 2-3 more people in the pipeline the next few days asking the same questions.
Dana Lewis
@danamlewis
Feb 09 2016 21:33
yep. even copying and pasting that type of thing from gitter to an 'issue' for someone to further validate before adding to the docs would be great
the challenge is getting everyone into the mindset of capturing info wherever the convos happen, and turning it piece by piece into somethign for the docs.
Here's a good example - was talking to @eszcloud about adding some language to the docs. She opened an issue (openaps/docs#84) and dropped a note there so it'd get on the 'list' to add/review for the docs, since she wasn't sure what the content would be and where to PR it to
Jim Matheson
@jmatheson
Feb 09 2016 21:34
right, and teaching folks to use github issues from day 1.
Dana Lewis
@danamlewis
Feb 09 2016 21:35
yep. good ideas.
but also, for people who speed read over the docs/skip things, needs reinforcing here as we go. "By the way, this has been a great discussion. Want to copy/paste it and open an issue in the docs for someone to review later/put into the docs?" "That's a great expalantion, want to PR it into the docs?"
@JaysonEwer i'm not sure the 50/50 rule applies anymore…or that I ever fully understood the idea other than it being an arbitrary goal suggestion..I think it says more about your ISF and how many carbs you're eating than anything else?
Jim Matheson
@jmatheson
Feb 09 2016 21:37
maybe. i am not completely sold.
Dana Lewis
@danamlewis
Feb 09 2016 21:38
@jmatheson not completely sold on which part?
JaysonEwer
@JaysonEwer
Feb 09 2016 21:41
@danamlewis true, when I go ultra low carb for a time, the 50/50 is out the window for sure. Like I said, I'm not sure it even needs to be a goal, I know I can tell you I have never once thought about 50/50 when I'm eyeing up my next GF 🍩!! I think the paradigm shift I was referring too was my Dr's view of that general rule. :-D
Jim Matheson
@jmatheson
Feb 09 2016 21:41
if that whole discussion was a thread tied to that page of instructions then the several days it takes to "Want to copy/paste it and open an issue in the docs for someone to review later/put into the docs?” would be a mute point. Because the person going there at 11pm tonight would see the thread and be able to move ahead.
the rest of the process is the same that as a group we need to improve the main doc as you suggested.
Dana Lewis
@danamlewis
Feb 09 2016 21:43
that's the point of PRing into the docs, though.
someone can see it, and learn from it
JaysonEwer
@JaysonEwer
Feb 09 2016 21:43
@danamlewis RE the auto updating the tracking... Please strike that from the record your honor. :-).
Dana Lewis
@danamlewis
Feb 09 2016 21:43
if no one updates the docs, we end up with where we are today. so people have to get comfortable putting what they learn into the docs.
@JaysonEwer lol.
JaysonEwer
@JaysonEwer
Feb 09 2016 21:43
:+1:
Call that section of the intro" GETTING COMFORTABLE" and let @jmatheson come up with a nice picture of a man holding his 👜 in a very comfortable way!😏
Jim Matheson
@jmatheson
Feb 09 2016 21:46
i don’t think bucketing the conversations by topic instead of in a firehose here would have any negative impact on updating the docs. Seems like it would have a positive impact because as PRs are reviewed for release it is easier to find all the components of discussion.
Dana Lewis
@danamlewis
Feb 09 2016 21:46
my point - gitter doesn't do threading. I'm all for that. so we need to use the docs, or create "issues" to track conversations around things that can be tracked via threading if you want that kind of thing.
(I think we're agreeing re: tracking conversations is good)
Dragan Maksimovic
@dm61
Feb 09 2016 22:15
belated congrats for your endo appointment @JaysonEwer !
JaysonEwer
@JaysonEwer
Feb 09 2016 22:20
Thanks @dm61 it was pretty cool. I'm seriously just getting started with this stuff and have already turned things around significantly! :-D. I'm looking forward to an A1C of 6ish at my May appointment.
Dragan Maksimovic
@dm61
Feb 09 2016 22:20
@JaysonEwer, I would entirely ignore the 50/50 basal/bolus "rule" - I do not think it has any relevance to good bg control. A well adjusted basal rate does make a difference.
Awesome!
I'd say things are moving too quickly for me - I can't even catch up - try the newest meal assist, replace the stupid CareLink with a TI stick, try edison, more docs, backups, git, ... lot's of fun, wish I had more time available :worried:
JaysonEwer
@JaysonEwer
Feb 09 2016 22:26
Yes, I totally agree with you @dm61. I'm not going to over think the 50/50 thing, it was just a very clear glimps into how Dr's will need to update their standards of care moving forward with the new paradigm.
I also agree with the velocity of things here, it's actually pretty awesome! 🚀 Hold on!!!
ah, here's the kind of docs fda (and other stakeholders) are used to working with
has nice table with identifiers to enumerate/link to each "harm" and discuss how it's covered
Jason Calabrese
@jasoncalabrese
Feb 09 2016 23:05
was wondering what a TOE was, started thinking we could have an OpenTOE with flipflop logo, but was just a boring "target of evaluation"
Screen Shot 2016-02-09 at 3.06.13 PM.png
looks a lot like our systems
JaysonEwer
@JaysonEwer
Feb 09 2016 23:07
🙂
:+1:
Sulka Haro
@sulkaharo
Feb 09 2016 23:14
Ha, talked with a friend who's using the 640G and apparently they have exact same issues as OpenAPS with the APS noticing BG going up too slowly after zero temping to cover the low, resulting in a high after the low.
Scott Leibrand
@scottleibrand
Feb 09 2016 23:18
not sure I follow
if OpenAPS notices BG going up too slowly after zero-temping, it will continue low-temping until BG rises enough, and then high-temp to blunt the rise
what's the failure mode there?
JaysonEwer
@JaysonEwer
Feb 09 2016 23:25
That happened to me this morning on my way to dr. The reason I didn't take any carbs was I wanted to be fasting for bloodwork. But then due to multiple temp 0.0's I ended up high later. I'm not sure I saw the blunting @scottleibrand mentioned. :-). Bottom line, apparently by basal is set a bit too high. This is one of the drawbacks of the single hormone systems, no?
Jason Calabrese
@jasoncalabrese
Feb 09 2016 23:29
out of space on edison
git corrupted
Scott Leibrand
@scottleibrand
Feb 09 2016 23:30
yeah, OpenAPS really assumes you tune your basals correctly. it can help if you don't, but only so much
JaysonEwer
@JaysonEwer
Feb 09 2016 23:30
I did this a week and a half or so ago. I restarted my Dexcom to get a couple more days out of it, but I end up having drops which negatively effects the ability for OpenAPS to react and respond appropriately.
I need to NOT try to stretch my Dexcom sensors! <<NOTE to self>>
Scott Leibrand
@scottleibrand
Feb 09 2016 23:31
how long do they run before starting to see issues?
JaysonEwer
@JaysonEwer
Feb 09 2016 23:32
I get the recommended 7 days, then on the second pass I don't get very far, I'm only a day or so into the second session.
Scott Leibrand
@scottleibrand
Feb 09 2016 23:33
bummer. @danamlewis usually manages to get 2 full weeks out of hers, and then they start to get wonky.
Dragan Maksimovic
@dm61
Feb 09 2016 23:34
The biggest problem is the huge delay, 1-2 hours at least, between any temp and its effect on the bg. In spite of this, in my experience, OpenAPS has been working pretty well overnight when bg tends to move really slowly (assuming the basal profile is well set, which is really important in my opinion). Nevertheless, I would still sometimes see "limit-cycle oscillations" from decreasing bg leading to zero temping, leading to increasing bg, leading to high temp, etc.
JaysonEwer
@JaysonEwer
Feb 09 2016 23:35
I've had a few, no CGM data events, perhaps I've walked away from my murse one too many times today as well. 👜😳
Scott Leibrand
@scottleibrand
Feb 09 2016 23:57
"limit-cycle oscillations": that's a good term for it.
two things that affect OpenAPS' propensity to do that: sensitivity and DIA
if your sensitivity is too low, it will overreact in both directions. you can increase ISF to dampen that.
if your DIA is too short, that will have a similar effect.