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

26th
Oct 2016
Paul Martin
@Spazholio
Oct 26 2016 00:16
Nice. So is there a way for me to get that into my loop without having to pull the dev branch of anything and just use the oref0-setup script?
garykidd
@garykidd
Oct 26 2016 01:30
so is there a way to offset timing for multiple rigs in cron easily?
Dana Lewis
@danamlewis
Oct 26 2016 01:32
Are you using wait for silence?
Paul Martin
@Spazholio
Oct 26 2016 01:36
The oref0-setup script sets WFS up automagically, right?
Dana Lewis
@danamlewis
Oct 26 2016 01:37
If you put in mmeowlink tty
Chris Oattes
@cjo20
Oct 26 2016 01:49
@riwarren Don't know if anyone has suggested this yet, but get a new inserter. I had that problem of not being released with one of my sensors, and they sent out a new one
garykidd
@garykidd
Oct 26 2016 01:50
yeah see the wait for silence. i think it works out that one stomps the other consistently at times and was just wondering if you could offset one a bit so that one is constantly just waiting
Scott Leibrand
@scottleibrand
Oct 26 2016 01:52
Sorry, I'm confused what the actual problem is. Did you see the random sleep?
garykidd
@garykidd
Oct 26 2016 02:17
I'm not sure there is a problem per se.
While watching the log of 2 different rigs Inoticed on setting basals and reporting BSs etc...and the other basically was waiting for silence mostly..they seemed tobe almost on same rhythm. last night it seemed like one was reporting to NS then the other and on and on and on...
Scott Leibrand
@scottleibrand
Oct 26 2016 02:31
What should happen if both are in range is that one of them will get everything refreshed, and then won't need to talk to the pump until it gets its next BG value, so during that time the other one can refresh everything as well.
There are some random length sleep statements to prevent them from both going active at once, but there's no reason the one with the better signal shouldn't do all the work, and the one that is a bit weaker can try to keep up to speed on what's happening in between while the other is waiting.
Paul Martin
@Spazholio
Oct 26 2016 03:01
The IOB that’s shown in NS - is that IOB that’s over and above what my regular basal would be without the loop?
(plus any boluses, of course)
Scott Leibrand
@scottleibrand
Oct 26 2016 03:02
Assuming it's from OpenAPS, It's net IOB. Zero would be no temp basals or boluses. Can be negative from low temps.
Dana Lewis
@danamlewis
Oct 26 2016 03:03
What version of NS are you using? Not sure if @jasoncalabrese has yet made that loop-based or if it's non net
Scott Leibrand
@scottleibrand
Oct 26 2016 03:03
IOB comes from OpenAPS if it's running. It's COB that's been lagging.
Paul Martin
@Spazholio
Oct 26 2016 03:04
@scottleibrand Ah, ok. Shame Medtronic doesn’t factor basals into its “Active Insulin” entry.
Dana Lewis
@danamlewis
Oct 26 2016 03:04
Nooooo kidding
Or provide ANY place on pump to see temp basal history </mini rant>
Paul Martin
@Spazholio
Oct 26 2016 03:05
Pump just told me to give myself 2.2u, but a quick look at my watch says I still have ~3u IOB from my Happy Happy Fun Time Glucose Rollercoaster earlier today.
Scott Leibrand
@scottleibrand
Oct 26 2016 03:08
You might consider @jasoncalabrese's method to set the bolus wizard target super high and only use the low target for looping...
(Assuming you care about bolus wizard suggestions being accurate)
Paul Martin
@Spazholio
Oct 26 2016 03:09
I don’t think I follow the methodology - can you elaborate? I also don’t use the BW or the BWP all that often so it might not be all that much of an issue.
Mike Stebbins
@mikestebbins
Oct 26 2016 05:16
temptarget question. If I set a temptarget in Nightscout, and then the openaps ns-loop completes successfully, thereby creating the /settings/temptarget.json file, if I then go out of range of the internet, my loop should continue to use that temptarget until it has "expired", correct? This is assuming that my G4 share is plugged into the Raspi (not relying on internet connection for Nightscout).
Scott Leibrand
@scottleibrand
Oct 26 2016 05:26
It depends ok what happens when the next refresh of the temptarget.json report happens. I believe that if you're clearly offline it will keep using the old one, but there are situations where it will get erased. Probably something you should test.
*on
Mike Stebbins
@mikestebbins
Oct 26 2016 05:29
Thank you, @scottleibrand, I'll try to test and report back. I ask, because I frequently set the temp target a bit before the long bike commute home, and I'd like it to continue zero-temping to hit that high target throughout my ride (my ride when I frequently forget to set-up a mobile hotspot).
Scott Leibrand
@scottleibrand
Oct 26 2016 05:31
Best test might be to look back at the pump-loop.log for the time period it was offline and see what target it was using.
Mike Stebbins
@mikestebbins
Oct 26 2016 05:58
@scottleibrand great idea. It looks like the temptarget was "erased" sometime shortly after leaving wifi range. I set the temptarget in nightscout at 16:48 (target of 180, duration 120). Several successful pump loops in pump-loop.log after that time reference the target of 180, with the last occurring at 17:00. I left wifi range a few minutes after 17:00, and the next successful pump loop at 17:08 shows a target back down to 95 (no temptarget).
Saskia
@Missdazzle
Oct 26 2016 11:08
@danamlewis @scottleibrand @bewest and @all the others helping : Thank you so much for making openAPS possible for even non-IT people like me!!! Got my new HbA1c today, it is another 5.x after almost one year with CGM and 5 months looping (coming from 7.x and worse) and even better than the last 5.x. I am SO happy!
Matthias
@ceben80
Oct 26 2016 11:33
That's a great archivment :tada:
emsamom
@emsamom
Oct 26 2016 12:39

Got some more good info overnight (still not physically connected to the pump I'm looping). Hoping to start understanding the logic a bit better.... can anyone give me some feedback?

I set the pump max basal to 3.0 units yesterday, but still haven't seen a basal above my manually programmed basal - even though BG was way high overnight... I think I might be missing something.

Here is one example of log where I think basal should have been set higher:
(pump program at this time is basal = 0.9):

Starting pump-loop at Mon Oct 24 12:32:15 EST 2016:
pump://JSON/read_temp_basal/monitor/temp_basal.json
reporting monitor/temp_basal.json
Profile less than 60m old
temp_basal.json less than 5m old
glucose.json newer than pumphistory: Refreshed pumphistory
{"iob":0,"activity":0,"bolussnooze":0,"basaliob":0,"netbasalinsulin":0,"hightempinsulin":0,"time":"2016-10-24T18:31:32.000Z"}
{"delta":0,"glucose":236,"short_avgdelta":0,"long_avgdelta":-4.9}
{"duration":0,"rate":0,"temp":"percent"}
target_bg unchanged: 110
Carb Impact: -4.9 mg/dL per 5m; CI Duration: NaN hours
Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: NaN hours
Setting neutral temp basal of 0.6U/hr
determine-basal://text/shell/enact/suggested.json
reporting enact/suggested.json
{"temp":"absolute","bg":236,"tick":"+0","eventualBG":207,"snoozeBG":207,"predBGs":{"IOB":[236,232,227,224,220,218,215,213,212,210,209,209,209]},"IOB":0,"
reason":"COB: undefined, Dev: -29, BGI: 0, ISF: 40, Target: 110; Eventual BG 207 > 100 but Min. Delta -4.90 < Exp. Delta -4; setting current basal of 0.6
as temp. Setting neutral temp basal of 0.6U/hr","duration":30,"rate":0.6}
pump://JSON/set_temp_basal/enact/enacted.json
reporting enact/enacted.json
{"requested":{"duration":30,"rate":0.6,"temp":"absolute"},"bg":236,"temp":"percent","snoozeBG":207,"timestamp":"2016-10-24T12:32:51.931435","predBGs":{"I
OB":[236,232,227,224,220,218,215,213,212,210,209,209,209]},"rate":0,"reason":"COB: undefined, Dev: -29, BGI: 0, ISF: 40, Target: 110; Eventual BG 207 > 1
00 but Min. Delta -4.90 < Exp. Delta -4; setting current basal of 0.6 as temp. Setting neutral temp basal of 0.6U/hr","eventualBG":207,"recieved":false,"
duration":0,"tick":"+0","IOB":0}
Settings less than 10m old
Pumphistory-24 less than 20m old
Completed pump-loop at Mon Oct 24 12:32:53 EST 2016

John Sjolund
@sjolundjohn
Oct 26 2016 13:25

Has anybody done work, OR is there any tool which will allow you to take the Nightscout data to help you to optimize your basal rates and sensativies on pump?

IE if you know that OpenAPS almost almost makes a basal adjustment between 3-6 am, then this gets flagged so that you can manually configure your basal rates on pump

Sandra Keßler
@SandraK82
Oct 26 2016 13:26
@sjolundjohn thinking about it
i have in mind a display like hourly stats or percentile graph for showing variations of basal rate
John Sjolund
@sjolundjohn
Oct 26 2016 13:28
Real time or retrospective
Sandra Keßler
@SandraK82
Oct 26 2016 13:31
for now in reports
garykidd
@garykidd
Oct 26 2016 13:36
trailing hour basal rate would be nice right in NS...as would IOB
both in graph form
not just point in time
Paul Martin
@Spazholio
Oct 26 2016 13:43
@SandraK82 I’d love something like that. The “Render Basal” in NS is nice as an overview, but I’d love some sort of graph where I can corellate where my BG was, what it was doing, and how my basal was compensated to match throughout the day.
garykidd
@garykidd
Oct 26 2016 13:45
was actually thinking this AM...the same and how plotting IOB would be very useful...as well as basal IOB all over time...the data is all there so it obviously can be done....wish I knew more about how.
what does a blinking DS3 led mean on edison breakout board. THe documentation I was referring to didnt say. ??
tazitoo
@tazitoo
Oct 26 2016 14:17
@emsamom - I think the relevant part of the log you posted is:
reason":"COB: undefined, Dev: -29, BGI: 0, ISF: 40, Target: 110; Eventual BG 207 > 100 but Min. Delta -4.90 < Exp. Delta -4; setting current basal of 0.6 as temp. Setting neutral temp basal of 0.6U/hr","duration":30,"rate":0.6}
So even though your BG is higher than your target, it is falling faster than expected, so it sets a "neutral" basal rate.
emsamom
@emsamom
Oct 26 2016 14:18
"falling faster than expected" comes from "Min. Delta -4.90 < Exp. Delta -4"?
Chris Oattes
@cjo20
Oct 26 2016 14:25
@garykidd isn't that the "battery no longer charging" one
garykidd
@garykidd
Oct 26 2016 14:37
@cjo20 from what I could ascertain, the DS3 led is supposed to go out when battery is charged. If I was setting up, I would've had blinking be charging and fully illuminated be be fully charged. I figured that blinking might have meant fully charged, but my rig turned off--probably on account of lack of power---so Im not sure what's what now.
DId I also recall correctly that battery info isnt available when connected via J2?
Chris Oattes
@cjo20
Oct 26 2016 14:42
It depends on how you connect the battery
and I think flashing means it hit the timeout
garykidd
@garykidd
Oct 26 2016 14:42
you mean how I connec the battery on j2?
Chris Oattes
@cjo20
Oct 26 2016 14:43
if you directly connect the battery to J2, you can get voltage information
if you do it via an adafruit board etc, you won't be able to
garykidd
@garykidd
Oct 26 2016 14:44
ok, cool. THis rig will be direct I think...although the timeout is inconvenient for sure
Dana Lewis
@danamlewis
Oct 26 2016 14:46
@sjolundjohn no tool yet. Short term would be to display averages of net basal activity by hour. But that mostly helps nighttime or when carbs are not in play. Long term/better solution will be openaps/oref0#99 which I've just started work on
@emsamom I am seeing some temps at percent in there too? Is that right?
emsamom
@emsamom
Oct 26 2016 14:50
@danamlewis I think I went too far back in my log file - the target BG of 110 is also way off what I set up yesterday.... Any suggestions of how to get the log file off my raspberry pi so I can go through it in something other than nano?
emsamom
@emsamom
Oct 26 2016 15:01

So here is a more correct log record (from early this morning).... I think the issue goes back to expected delta and actual delta...

{"temp":"absolute","bg":338,"tick":-9,"eventualBG":302,"snoozeBG":302,"predBGs":{"IOB":[338,330,322,316,310,304,300,296,293,291,289,289,289]},"IOB":0,"re
ason":"COB: undefined, Dev: -36, BGI: 0, ISF: 40, Target: 160; Eventual BG 302 > 160 but Min. Delta -9.00 < Exp. Delta -5.9, temp 1 ~ req 1U/hr"}
{"bg":349,"temp":"absolute","snoozeBG":354,"recieved":true,"predBGs":{"IOB":[349,350,350,351,352,352,352,353,353,353,353,353,353]},"rate":1,"reason":"COB
: undefined, Dev: 5, BGI: 0, ISF: 40, Target: 160; Eventual BG 354 >= 160, max_iob 0, 20m@0.000 = 0.000 > 2 * req 0. Setting temp basal of 1U/hr. Setting
neutral temp basal of 1U/hr","eventualBG":354,"timestamp":"2016-10-26T04:38:40.060031","duration":30,"tick":"+1","IOB":0}

Darrell Wright
@beached
Oct 26 2016 15:02
Has anyone considered using the pumps suspend feature to enable basal rate change on non-loopable pumps?
Set the basal to X and then interpolate to get the desired value
Dana Lewis
@danamlewis
Oct 26 2016 15:03
@beached there's at least one person who did a micro bolus and suspend version...but lots of risk added when you add in suspends in case you go out of range or battery runs out, not to mention boluses (for if it gets stuck reissuing same command and boluses add up quickly!
@emsamom it looks like iob is always 0 in these examples you're showing?
Have you ever seen a non-zero iob reported?
Also what is your target set at?
Darrell Wright
@beached
Oct 26 2016 15:05
@danamlewis Ah I see, if you go out of range and cannot suspend it you are getting too much
emsamom
@emsamom
Oct 26 2016 15:10

Looking through logs I do have some IOB != 0 - after I've entered carbs/bolus via nightscout...

{"iob":10.808,"activity":0.0692,"bolussnooze":8.028,"basaliob":0,"netbasalinsulin":0,"hightempinsulin":0,"time":"2016-10-24T23:11:00.000Z"}
{"delta":4,"glucose":233,"short_avgdelta":3.11,"long_avgdelta":2.27}
{"duration":0,"rate":0,"temp":"percent"}

Remember I'm not actually using this pump live, so I've not been bolusing from it.

My understanding is that basal insulin won't be included in IOB - so since I'm reviewing overnight logs this is appropriate?

Dana Lewis
@danamlewis
Oct 26 2016 15:12
No. Its net iob so should definitely be + or - with the basal activity
Assuming you are enacting on the test pump
emsamom
@emsamom
Oct 26 2016 15:20
mmmm. I will keep digging

definitely enacting on test pump - I can see the test rate when I check the pump screen.

I think all it is is that the logic is deciding I don't need a higher basal. I found a temp basal rate of 2.275 this morning after I ate breakfast (and didn't tell nightscout about my carbs or bolus)

Dana Lewis
@danamlewis
Oct 26 2016 15:22
I am suspicious that after these temps being shown, iob is still 0
What is your target, btw?
emsamom
@emsamom
Oct 26 2016 15:23

so here is the full loop from after breakfast....

Starting pump-loop at Wed Oct 26 09:27:05 EDT 2016:
pump://JSON/read_temp_basal/monitor/temp_basal.json
reporting monitor/temp_basal.json
Profile less than 60m old
temp_basal.json less than 5m old
glucose.json newer than pumphistory: Refreshed pumphistory
{"iob":-0.871,"activity":-0.003,"bolussnooze":0,"basaliob":-0.871,"netbasalinsulin":-0.95,"hightempinsulin":0,"time":"2016-10-26T13:26:24.000Z"}
{"delta":15,"glucose":119,"short_avgdelta":12.61,"long_avgdelta":5.99}
{"duration":21,"rate":1,"temp":"absolute"}
target_bg unchanged: 160
Carb Impact: 5.4 mg/dL per 5m; CI Duration: NaN hours
Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: NaN hours
determine-basal://text/shell/enact/suggested.json
reporting enact/suggested.json
{"temp":"absolute","bg":119,"tick":"+15","eventualBG":186,"snoozeBG":186,"predBGs":{"IOB":[119,125,130,135,139,143,147,150,153,156,158,160,161,162,164,165,167,168,170,171,172,174,175,176,177,179,180,181,182,183,184,185,186,187,187,188,189,190,190,191,191,191,192]},"IOB":-0.871,"reason":"COB: undefined, Dev: 32, BGI: 0.6, ISF: 40, Target: 160; Eventual BG 186 >= 160, temp 1<2.3U/hr","duration":30,"rate":2.3} pump://JSON/set_temp_basal/enact/enacted.json reporting enact/enacted.json {"bg":119,"temp":"absolute","snoozeBG":186,"recieved":true,"predBGs":{"IOB":[119,125,130,135,139,143,147,150,153,156,158,160,161,162,164,165,167,168,170,171,172,174,175,176,177,179,180,181,182,183,184,185,186,187,187,188,189,190,190,191,191,191,192]},"rate":2.275,"reason":"COB: undefined, Dev: 32, BGI: 0.6, ISF: 40, Target: 160; Eventual BG 186 >= 160, temp 1<2.3U/hr","eventualBG":186,"timestamp":"2016-10-26T09:27:43.390442","duration":30,"tick":"+15","IOB":-0.871}
Settings less than 10m old
Pumphistory-24 less than 20m old
Completed pump-loop at Wed Oct 26 09:27:45 EDT 2016

Dana Lewis
@danamlewis
Oct 26 2016 15:23
There's some iob change, that's good
emsamom
@emsamom
Oct 26 2016 15:24
yep. how can iob be negative though?
Dana Lewis
@danamlewis
Oct 26 2016 15:24
Because of lower temp basals compared to preprogrammed rates. It's net iob
(Makes a big different FYI if you go into a meal w lots of negative iob - doing the same meal bolus you'd do otherwise won't go as far as if you were at 0iob; and will have a sharper spike because of the impact on the lack of insulin in your liver when carbs hit)
emsamom
@emsamom
Oct 26 2016 15:25
net compared to preprogrammed?
Dana Lewis
@danamlewis
Oct 26 2016 15:26
Net = any higher and lower temps it gives (preprogrammed meaning what basals are scheduled on your pump)
emsamom
@emsamom
Oct 26 2016 15:26
aha - that is a very important point!
I thought it was against 0 :)
would it be useful to put a glossary of terms in the loop output together? Happy to start it if you think it would be useful...I keep rereading phase 3.
Dana Lewis
@danamlewis
Oct 26 2016 15:30
I would pr to the phase 3 understanding output page..we have/had a glossary but no one bothers to look unless its described on a page
emsamom
@emsamom
Oct 26 2016 15:30
ok will take a stab at it. worth it for someone to tell me where my understanding is bad ;)
Dana Lewis
@danamlewis
Oct 26 2016 15:30
:+1:
Did you see Scott's suggestion for language on your open pr?
Scott Leibrand
@scottleibrand
Oct 26 2016 15:47
@emsamom what is your max IOB in preferences.json? Still zero?
emsamom
@emsamom
Oct 26 2016 15:48
@scottleibrand it IS still zero - I thought that by changing my max basal rate on pump that would be enough?
Dana Lewis
@danamlewis
Oct 26 2016 15:48
Ahhhhh
Nope
emsamom
@emsamom
Oct 26 2016 15:48
so I should edit the preferences file directly to change? or do something else....?
Yep!
Scott Leibrand
@scottleibrand
Oct 26 2016 15:49
Your max basal allows it to high temp when coming out of a low with negative IOB.
But without a positive max IOB it will never high temp IOB above zero when high/rising.
emsamom
@emsamom
Oct 26 2016 15:54
got it. I missed that I this value comes from preferences. I will review all the preference values now, and update docs. @danamlewis I saw (and thought I responded to) Scott's suggestions on my open PR.... but must not have completed it correctly. will check again
Neal
@tnharvey
Oct 26 2016 16:43
I noticed that, after the first run of interactive setup script, the loop seems to be running either every minute or continuosly (and that's how long it takes to run). I read the section on Using Cron in the walkthrough, and it wasn't specific about editing cron, but when I edited crontab it doesn't show specific times for any of the items in crontab, just *s. Does that mean it's running continuously, or am I looking in the wrong place to adjust how often the loop repeats?
garykidd
@garykidd
Oct 26 2016 16:43
regarding powering the intel edison breakout board,,,,was curious if anyone else had experienced any difficulties or inconsistencies with the documentation. My DS3 LED illuminates when there is usb cable power on j16....but doesn't illuminate when there is usb power on j3....(which the documentation states...it should). Despite the LED not lighting up on j3 I am wondering if that's the way to charge becuase when I hook usb up to j16...it lights solid for awhile but then fairly quickly starts to flash---but doesnt seem to hold any charge...well not for very long at least. thoughts?
Luis Betancourt
@betluis
Oct 26 2016 16:43
Hello, Dana, Scott, you were helping me two weeks ago or so with my setup. Just wanted to let you know I had to stop due to the pump failing... I am in the process of getting a new one. not that easy, but making progress. I look forward to engage into the setup again soon..
Neal
@tnharvey
Oct 26 2016 16:46
@garykidd I've only powered it off of J16, and have always had to use both J3 for comms and J16 to power if I don't have a battery hooked up to J2.
garykidd
@garykidd
Oct 26 2016 16:46
J2 is the battery connector. If you want to power the breakout board with a rechargeable lithium-ion battery,
attach it to J2. (Refer to Figure 2 for battery polarity.) When you attach a rechargeable lithium-ion battery,
the breakout board will recharge the battery whenever power is applied via J21 or J22, or via J3 (when the
board is attached to a USB host).
Dana Lewis
@danamlewis
Oct 26 2016 16:47
@tnharvey cron is a universal computer thing, worth googling to help you answer the Q re * 's :)
garykidd
@garykidd
Oct 26 2016 16:48
thanks..... @tnharvey does it charge a connected battery (j2) when you hook up usb power to j16?
Neal
@tnharvey
Oct 26 2016 16:49
@garykidd I was just going to say, I've never had it charge a battery (that I could tell) when connected to J16, so you may have a point there. Maybe it is only charging a battery via J3, but not powering the edison? That might make sense, keeping the charging circuit separate from the powering circuit?
'?' may be taken as an indication of my lack of confidence in that response :D
Chris Oattes
@cjo20
Oct 26 2016 16:51
It only charges via OTG
garykidd
@garykidd
Oct 26 2016 16:51
I'm thinking it could be a mistake becuase I do see DS3 illuminate when j16 is powered....but then again it switches to LED flashing pretty quickly...where the documention indicates it should switch to off when battery is charged...never did that
OTG=J16...correct?
if that's the case the documentation is just wrong then....right?
Neal
@tnharvey
Oct 26 2016 16:59

@danamlewis Thanks, maybe it was because I was trying to google at 1am while sick. I looked it up again and found a clearer answer. So 5 *s in crontab is basically "run continuously", which is what it's doing.
One more question: Is there a reason I see three instances of openaps in there, with one seemingly cut off?

* * * * * sudo wpa_cli scan
* * * * * killall -g --older-than 15m openaps
* * * * * cd /home/edison/edtronic && oref0-reset-git
* * * * * cd /home/edison/edtronic && ps aux | grep -v grep | grep -q 'openaps $
* * * * * cd /home/edison/edtronic && ps aux | grep -v grep | grep -q 'openaps $
* * * * * cd /home/edison/edtronic && ( ps aux | grep -v grep | grep -q 'openap$

I'm thinking I should get rid of the other two, right?

Scott Leibrand
@scottleibrand
Oct 26 2016 17:00
@tnharvey the cron is designed to run every minute, but check whether that process is already running before doing anything.
The $ indicates there is more on that line that your screen is too small to show.
You can make it bigger or scroll over with the arrow keys.
Neal
@tnharvey
Oct 26 2016 17:05
@scottleibrand Yep, looks like they are each unique, thanks for the newb pointers =]
Okay, I'll adjust those. My edison runs a little hot, and I wanted to see if dialing back the frequency of cron would help until I can get a replacement, hopefully it will
Dana Lewis
@danamlewis
Oct 26 2016 17:07
:+1: there's three for a reason :)
Hope you feel better btw
Paul Martin
@Spazholio
Oct 26 2016 17:07

I’m getting a LOT of

Starting pump-loop at Wed Oct 26 13:05:10 EDT 2016:
Radio ok, Listening: .No comms detected
Old pumphistory: 
Error, retrying

in my pump-loop.log file and I don’t know why. Battery’s good on the pump and I’m completely shut down the rig and powered it back on. Any way to see why or increase the verbosity?

Wait…it just started running mmtune. Does it not run that each iteration of the loop anymore?
Neal
@tnharvey
Oct 26 2016 17:11
@danamlewis thanks, a little, this is one of those colds that drag on for a while :/
@Spazholio Mine is doing that too, I haven't checked the script yet to see, but I'm wondering if it doesn't run mmtune immediately, it waits to have multiple "No comms". Once mmtune runs, it seems to work fine
Scott Leibrand
@scottleibrand
Oct 26 2016 17:14
Correct: oref0-setup doesn't run mmtune on every run, or even on every failure. That is primarily to support multiple-rig setups: you don't want a rig just outside usable range to be running mmtune every few minutes and stomping on the comms with another rig that is communicating fine.
garykidd
@garykidd
Oct 26 2016 17:15
can anyone confirm a flashing DS3 led on edison means you've reached timeout?
Scott Leibrand
@scottleibrand
Oct 26 2016 17:15
So instead we run mmtune on a random subset of failures, which means if you have it mistuned, it may take about 5 minutes to try again
garykidd
@garykidd
Oct 26 2016 17:15
breakout board
Scott Leibrand
@scottleibrand
Oct 26 2016 17:15
dunno if it's DS3, but a LED starts flashing after about 6 hours of battery charging
and then it stops charging until you unplug the usb and plug it back in
garykidd
@garykidd
Oct 26 2016 17:15
flasheds sooner than that though
Scott Leibrand
@scottleibrand
Oct 26 2016 17:16
dunno
garykidd
@garykidd
Oct 26 2016 17:16
but apparently I dont have much of a charge
ok thx
Neal
@tnharvey
Oct 26 2016 17:16
@garykidd I haven't had great success charging a J2 battery via J16, it's either slow, or isn't working, but my edison also runs hot and is cranky, so I'm ordering a replacement, which might resolve that. In the meantime, I've just connected an adafruit micro LiPo charger between the battery and J2, and charge via that.
garykidd
@garykidd
Oct 26 2016 17:16
are you using adafruit 100c by chance?
1000c
Neal
@tnharvey
Oct 26 2016 17:17
Oh, and for me, I was getting the flashing led with inadequate charge, it was in a weird boot cycle thing, and once I had it connected to an adequate power source, it ran fine. But it doesn't sound like that's what you're dealing with.
garykidd
@garykidd
Oct 26 2016 17:17
it actually may be
seem to be getting only minutes off of supposed charge
Neal
@tnharvey
Oct 26 2016 17:18
100c? it's the 1623
I'm not completely mobile yet though, no case or anything, just testing it since yesterday. I've been cheating and just using a 2600mAh micro usb battery via J16 whenever my battery runs low. It's an old battery so it doesn't have full life =] Battery and case are on order.
Actually, that reminds me, @Spazholio, what battery are you using, and about how much life do you get?
emsamom
@emsamom
Oct 26 2016 17:22
I'm not ready to loop during the day yet, and I HATE the minimed 522 UI (I wear a T-Slim now). I'm considering swapping pumps for night/day use. Is that ridiculous?
Dana Lewis
@danamlewis
Oct 26 2016 17:22
Nope, some people do it.
Also which parts do you hate?
Paul Martin
@Spazholio
Oct 26 2016 17:23
2200 mAh LiPo from Adafruit, and I tend to get ~16 hours on it. I think it’s decreased slightly with the advent of the new loop since it’s running more often. I’ll take the hit though, as I’d prefer more iterations rather than less. I can always plug it in at work to recharge.
Dana Lewis
@danamlewis
Oct 26 2016 17:23
I primarily bolus with easy bolus button and enter carbs online so I spend very little time touching or looking at pump @emsamom in case that helps
Paul Martin
@Spazholio
Oct 26 2016 17:23
@tnharvey
I have a spare battery but I’m loathe to disassemble the case my rig is in. I have the wires JUST right as it is...
garykidd
@garykidd
Oct 26 2016 17:24
@Spazholio are you using a separate charging board/circuit?
emsamom
@emsamom
Oct 26 2016 17:24
@danamlewis you are why Bigfoot is building the "carb estimate" into it's device!
Neal
@tnharvey
Oct 26 2016 17:24
@Spazholio Okay, that's good to know. I think the heat coming off the edison is wasting a lot of energy, because it's draining pretty quick. Hopefully a new edison will solve that problem
Paul Martin
@Spazholio
Oct 26 2016 17:25
@tnharvey I’m using this to charge: https://www.adafruit.com/products/2465
garykidd
@garykidd
Oct 26 2016 17:25
maybe try that edison for a home rig....or maybe even some silicon grease and a heatsink.
emsamom
@emsamom
Oct 26 2016 17:25
@danamlewis I'm looking forward to having basal correcting for misjudgement of carbs, letting me easy bolus generalized amounts. What a novel idea!
garykidd
@garykidd
Oct 26 2016 17:26
so battery into the battery connector and lipo output from charger into J2?
j2 on edison
Dana Lewis
@danamlewis
Oct 26 2016 17:26
@emsamom you'll probably want to do like me and do carb entry online and easy bolus and then loop takes care of tail :)
emsamom
@emsamom
Oct 26 2016 17:26
Can't wait!
Dana Lewis
@danamlewis
Oct 26 2016 17:27
@emsamom actually no - I'm decent at carb counting, but there's so many things that impact how they actually roll out in your body over time.
Neal
@tnharvey
Oct 26 2016 17:27
@garykidd I've thought about that too, I didn't have any small heatsinks lying around, so I'll have to order one. And besides, I'm home most of the time, so it would still be super useful!
emsamom
@emsamom
Oct 26 2016 17:27
@danamlewis I readily admit to sucking at carb counting. Been doing it for 20 years so you would think I'd have it down by now ;)
garykidd
@garykidd
Oct 26 2016 17:27
@danamlewis I'm sure you have plenty of experience....I suspect you've over bolused before and under-bolused before......does loop handle one better than the other assuming the same amount of deviation from exact bolus
Paul Martin
@Spazholio
Oct 26 2016 17:27
Bingo. I agree with @danamlewis - carb counting is honestly the easy variable to account for. It’s all the other unknowns that screw things up.
@emsamom Without getting all preachy - have you considered a low-carb diet? It helped me literally more than anything else ever has, and I’m 20 years in as well.
Dana Lewis
@danamlewis
Oct 26 2016 17:28
@garykidd I'd say it treats equally well on either side if it knows about the carbs. Although if you're way under, obviously due to safety limits it may not be able to keep up as well.
Neal
@tnharvey
Oct 26 2016 17:29
@Spazholio Does that charger charge much faster than the little one I'm using? I thought with the 500mA soldered I was getting a pretty fast charge, but if not...
garykidd
@garykidd
Oct 26 2016 17:29
still carb counting has its limitations...and I'e been doing it for a long while too---and still can get things wrong----I think being able to concsitently catch it when you're off--is the real difference maker.
Dana Lewis
@danamlewis
Oct 26 2016 17:29
@Spazholio and eating soon mode ;)
emsamom
@emsamom
Oct 26 2016 17:29
Oh shoot. I just realized I can't use the easy bolus AND the carb wizard.... I have varying carb ratios throughout day... not sure I'm willing to start doing that math in my head!
Dana Lewis
@danamlewis
Oct 26 2016 17:29
Ah
emsamom
@emsamom
Oct 26 2016 17:30
@Spazholio I've done low carb before - right now I'm in a serious love affair with wheat in all its forms :(
Paul Martin
@Spazholio
Oct 26 2016 17:30
@tnharvey I honestly don’t know. I’m using @cjo20’s program to monitor the battery charge, but a) it’s non-linear, so I don’t get a TRUE reading of how much charge is left, and b) whenever I plug it in, the charge reading almost immediately goes into the high 90’s.
emsamom
@emsamom
Oct 26 2016 17:30
Also have several logistical challenges to overcome on travelling out of my house with this rig.... I'll get there eventually but for the first few days I'm thinking swapping out pumps will be easiest....
Dana Lewis
@danamlewis
Oct 26 2016 17:31
:+1: make it work for you however you need to!
Paul Martin
@Spazholio
Oct 26 2016 17:31
@danamlewis One could say that my usage of Eating Soon mode is “off label” but yes - it’s worked VERY well for me. Now I just need to be able to temp higher than 6.3u/hr, and I think any miscalculation for protein and such will be handled a lot better than it has been. I don’t have lots of miscalculations though.
Dana Lewis
@danamlewis
Oct 26 2016 17:32
May need to watch iob when you switch though @emsamom i.e. Be careful if you bolus lots on tslim and then shortly after switch to loop pump because it won't have that knowledge
emsamom
@emsamom
Oct 26 2016 17:33
@danamlewis if I'm entering carbs/bolus into nightscout will that ensure openAPS knows my total insulin on board?
Dana Lewis
@danamlewis
Oct 26 2016 17:33
If it's patient created (aka since I created if) I think it's already by nature never been on label? ;) (I hear you) @Spazholio
garykidd
@garykidd
Oct 26 2016 17:33
@danamlewis would you think it useful to have NS plot IOB?
Paul Martin
@Spazholio
Oct 26 2016 17:33
@danamlewis Damn your talons of logic. =)
Dana Lewis
@danamlewis
Oct 26 2016 17:33
@Spazholio I need that on my LI: "talons of logic" ;) lol
Neal
@tnharvey
Oct 26 2016 17:34
lol
@Spazholio Are you using @cjo20 's edison voltage on jubilinux? I just switched, not sure if that's what you're using
Paul Martin
@Spazholio
Oct 26 2016 17:36
@emsamom What logistical challenges are you anticipating when leaving the house?
@tnharvey Yes, that’s what I’m using.
Neal
@tnharvey
Oct 26 2016 17:36
@Spazholio cool, I'll give that a try
garykidd
@garykidd
Oct 26 2016 17:36
@Spazholio regaridng hooking up charger board.... @garykidd
just connect battery output into the battery connector and lipo output from charger into J2 on edison...or did you do some other way?
Paul Martin
@Spazholio
Oct 26 2016 17:37
@garykidd I think that’s how I soldered it up, but I’m not 100%. Sounds right though.
garykidd
@garykidd
Oct 26 2016 17:38
and you're still able to get battery voltage reporting? Cuase I would've thought it would show up 3.7v (plus a little bit) and then just dead...
Paul Martin
@Spazholio
Oct 26 2016 17:38
@garykidd Yessir, I am.
Scott Leibrand
@scottleibrand
Oct 26 2016 17:39
@danamlewis "talons of awesome logic"? ;-)
Dana Lewis
@danamlewis
Oct 26 2016 17:39
@garykidd we've talked about this before. I wonder how many Loop users are really using their iob graph...
@scottleibrand it would make a great tshirt too ;)
garykidd
@garykidd
Oct 26 2016 17:42
wait...I remember the talk...and admittedly just trying to bait you a bit (cause I think it'd be awesome)...but not sure what you mean by loop users using their iob graph--that implies it exists...which I didn't think that it did??
Dana Lewis
@danamlewis
Oct 26 2016 17:43
LoopKit Loop (iOS app) loopers I mean. I believe there is an iob graph. But I'm not sure how many people really use it. But like all things, if you want it in NS, go for it ;) I may not be a keen user, but others might be
The eventual BG predictions + looping take away the need for that for me. If I was not looping and only iob vis I had was what was shown on bolus wizard...then maybe
Paul Martin
@Spazholio
Oct 26 2016 17:44
There’s an IOB graph in NS?
Dana Lewis
@danamlewis
Oct 26 2016 17:44
No
Paul Martin
@Spazholio
Oct 26 2016 17:44
Oh good.
Thought I had missed a checkbox somewhere.
Dana Lewis
@danamlewis
Oct 26 2016 17:45
There is not one. But @garykidd wants one. So if y'all want to do it, go forth and build ;) :)
Paul Martin
@Spazholio
Oct 26 2016 17:47
Truth be told, I’d actually like to see a graph of basal rates right alongside my BG readings.
Dana Lewis
@danamlewis
Oct 26 2016 17:48
Are you not doing basal rendering?
emsamom
@emsamom
Oct 26 2016 17:48
@Spazholio I've not secured a battery for the pi, or figured out how to hotspot it to my phone yet. Trying to develop iteratively and those things seemed to add unnecessary complexity the first time through. The other issue is that I am a consultant so leaving the house usually means leaving the state or country.... which makes me think I need some backup hardware.
@Spazholio And how the heck am I going to carry it all around with me - have to think about that one too :)
Dana Lewis
@danamlewis
Oct 26 2016 17:48
@emsamom which CGM do you have?
emsamom
@emsamom
Oct 26 2016 17:49
Dexcom g5 (so no receiver)
Dana Lewis
@danamlewis
Oct 26 2016 17:49
K. So connectivity needed. Hotspot is fairly easy to add your phones wifi to Pi wifi list when you're ready.
Scott Leibrand
@scottleibrand
Oct 26 2016 17:49
@emsamom why no receiver? are you not in the US?
FDA requires that Dexcom sell the G5 with a receiver (as a backup if your phone isn't working)
emsamom
@emsamom
Oct 26 2016 17:50
@scottleibrand I did the nightscout bridge without putting my receiver hardware in the mix. I have the receiver stored in a box in my bathroom :)
Paul Martin
@Spazholio
Oct 26 2016 17:50
@emsamom Once you get there, it’s easy to set up hotspot connectivity. And using NS to pull BG values completes it.
Scott Leibrand
@scottleibrand
Oct 26 2016 17:50
@emsamom ah, ok. you might want to take it out of the box so you can loop offline by plugging it into the Pi
emsamom
@emsamom
Oct 26 2016 17:51
@scottleibrand if I plug the Dexcom receiver into the pi and don't have internet connection, how will nightscout get the data?
Oh wait! I could do this without nightscout?
Dana Lewis
@danamlewis
Oct 26 2016 17:52
It'll upload later when you get wifi but in meantime act from receiver data
Paul Martin
@Spazholio
Oct 26 2016 17:52
@scottleibrand And there’s currently no way to use the loop with a receiver connected to the G4 receiver? I don’t want to charge it up, just be able to pull values, but apparently that can’t be done due to voltage stuffs?
Dana Lewis
@danamlewis
Oct 26 2016 17:52
Yep!
@Spazholio huh? G4 received plugged in works
Paul Martin
@Spazholio
Oct 26 2016 17:52
Into an Edison? Powered by a 3.3v LiPo?
Dana Lewis
@danamlewis
Oct 26 2016 17:52
Ohhhh not Pi
Paul Martin
@Spazholio
Oct 26 2016 17:53
Sorry - I didn’t mention Edison. My bad.
Dana Lewis
@danamlewis
Oct 26 2016 17:53
;)
garykidd
@garykidd
Oct 26 2016 17:53
Cant you power receiver separately and hook up to edison though?
Dana Lewis
@danamlewis
Oct 26 2016 17:53
@emsamom if there is a receiver connected it'll pull from that, otherwise look to NS. Good redundancy :) and worth it for offline looping!
emsamom
@emsamom
Oct 26 2016 17:54

OK so let me make sure I understand fully - I plug receiver and carelink into pi.... then opeaps does not need a hotspot or any other internet connection to work?

I could do this ON A PLANE?

Darrell Wright
@beached
Oct 26 2016 17:54
@danamlewis I use my iob and insulin delivery graphs. iob helps me go active as my insulin works 5-10X as much when I am. I use the delivery graphs to see patterns of long periods of high or low temping to adjust basal rates
Dana Lewis
@danamlewis
Oct 26 2016 17:54
Yep!! :)
Paul Martin
@Spazholio
Oct 26 2016 17:54
@garykidd That’s what I don’t get - the receiver IS powered. I mean, it has an internal battery. I don’t need external power, I just want it to give the Edison readings.
emsamom
@emsamom
Oct 26 2016 17:54
Mind blown!
Dana Lewis
@danamlewis
Oct 26 2016 17:55
@emsamom I was really excited when we got that working :) https://mobile.twitter.com/danamlewis/status/621205412310716416?ref_src=twsrc%5Etfw
emsamom
@emsamom
Oct 26 2016 17:55
I cannot wait for my reservoirs to arrive so I can start doing this for real! I'm already changing my Tslim temp basals based on what openaps is recommending.....!
Dana Lewis
@danamlewis
Oct 26 2016 17:55
I spend a lot of time on planes, too. Case in point...on one now :smile:
garykidd
@garykidd
Oct 26 2016 17:55
I think it still attempts to draw power when connected via usb though and its not enough power...I think
emsamom
@emsamom
Oct 26 2016 17:56
@danamlewis you are reminding me that I waited way way to long to do this ;) How do you monitor what openaps is doing when it's offline looping?
Dana Lewis
@danamlewis
Oct 26 2016 17:58
Hotspot and ssh from same network. Or checking temps set on pump. Or you can do https://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-1/offline-monitoring.html
Neal
@tnharvey
Oct 26 2016 17:58
@Spazholio I've wondered about that. Is it possible to modify an OTG cable to remove the power pins and still have data?
Paul Martin
@Spazholio
Oct 26 2016 17:59
@tnharvey Physically, it sounds possible. But I don’t know if either the Edison or the receiver would accept/balk at such a cable.
emsamom
@emsamom
Oct 26 2016 17:59
@danamlewis I'm giddy!
Dana Lewis
@danamlewis
Oct 26 2016 17:59
;)
Paul Martin
@Spazholio
Oct 26 2016 18:00
@beached Are you using Loop or OpenAPS?
Darrell Wright
@beached
Oct 26 2016 18:00
@Spazholio Loop currently
Paul Martin
@Spazholio
Oct 26 2016 18:00
Ah, ok.
Scott Leibrand
@scottleibrand
Oct 26 2016 18:00
@Spazholio you have to have 5V to do USB OTG, regardless of the draw of the receiver. What you can do with the new Explorer board is plug in a USB battery to the UART microUSB port, and then plug in the Dexcom via the OTG port. it will pass the 5V through from the battery to the Dexcom
Darrell Wright
@beached
Oct 26 2016 18:01
@Spazholio and Urchin
Scott Leibrand
@scottleibrand
Oct 26 2016 18:01
not quite as compact as using a LiPo, but more compact than a Pi
garykidd
@garykidd
Oct 26 2016 18:01
@beached trying to lobby folks for usefulness of NS to plot IOB over time--not just point in time. Or are you looking at aggregated IOB graphs?
Paul Martin
@Spazholio
Oct 26 2016 18:01
@scottleibrand So I can have my external charging battery plugged into one port, and my receiver into another?
Scott Leibrand
@scottleibrand
Oct 26 2016 18:01
right
Paul Martin
@Spazholio
Oct 26 2016 18:01
...
emsamom
@emsamom
Oct 26 2016 18:01
so if I have a wireless wifi router (that my kids currently use to play multi-player minecraft while in the car) I could watch the logs, right @danamlewis?
Paul Martin
@Spazholio
Oct 26 2016 18:02
I want that. Now. =)
John Sjolund
@sjolundjohn
Oct 26 2016 18:02
@danamlewis thanks for sharing about the long term solution - that is exactly right - would be so valuable!
Scott Leibrand
@scottleibrand
Oct 26 2016 18:02
there is also an option to turn off charging on the Dexcom, so it doesn't draw down the external battery
Darrell Wright
@beached
Oct 26 2016 18:02
@garykidd so Loop gives me the insulin dosing and decay along with the ability to trace the line and see the values
Scott Leibrand
@scottleibrand
Oct 26 2016 18:02
@Spazholio have you preordered your Explorer board yet?
Paul Martin
@Spazholio
Oct 26 2016 18:02
Might have to wait for @eluhrs to make one of his awesome 3D printed cases. =)
@scottleibrand 2 of ‘em, yessir.
Scott Leibrand
@scottleibrand
Oct 26 2016 18:02
:+1:
Dana Lewis
@danamlewis
Oct 26 2016 18:02
@sjolundjohn help welcome from anyone who wants to help me get it done :) I started stubbing it out in pseudocode
Paul Martin
@Spazholio
Oct 26 2016 18:03
@scottleibrand Wait, so if I disable charging, can I plug it into the OTG port and have it do what I’m looking to do?
John Sjolund
@sjolundjohn
Oct 26 2016 18:03
@danamlewis how/what is best to contribute?
garykidd
@garykidd
Oct 26 2016 18:03
yeah--I think the data is right there---I'm interested in seeing it graphically...wouldnt you like that?
Scott Leibrand
@scottleibrand
Oct 26 2016 18:03
@Spazholio no, you still need 5V to power the USB OTG on the intel breakout board
Dana Lewis
@danamlewis
Oct 26 2016 18:04
@sjolundjohn you'll be the first person to offer help, so we can figure out what works best! I can add you to my google doc later where I am pseudo-coding and we can go from there?
Scott Leibrand
@scottleibrand
Oct 26 2016 18:04
@Spazholio if you have a Sparkfun base block, though, you can do USB OTG there
Paul Martin
@Spazholio
Oct 26 2016 18:04
Rats. And LiPos just don’t do that, do they? 3.3V, IIRC?
John Sjolund
@sjolundjohn
Oct 26 2016 18:04
@danamlewis sure thing, key thing is how to contribute and format since I can't do the coding, but I can do some thinking hopefully. At your disposal
Scott Leibrand
@scottleibrand
Oct 26 2016 18:05
I think they're more like 3.7. Morgan hooked up a volt-meter to the USB OTG on the Sparkfun base block, and was only seeing about ~4V there, and it worked fine.
but the Intel breakout board is wired in a way that doesn't provide LiPo power to the OTG port
garykidd
@garykidd
Oct 26 2016 18:06
@Spazholio you have that charger block with 5v output though---could conceivably use that.
Paul Martin
@Spazholio
Oct 26 2016 18:06
Man, I wish openxshareble were more reliable. It’s so close.
Dana Lewis
@danamlewis
Oct 26 2016 18:06
@sjolundjohn if I can code, you can code ;) but even discussion and review of pseudocode logic to make sure not missing chunks before coding will be great. Can you pm me your preferred email for google docs, and I'll add you in a few days once I clean it up?
Paul Martin
@Spazholio
Oct 26 2016 18:06
@garykidd How do you mean?
Scott Leibrand
@scottleibrand
Oct 26 2016 18:09
ya, there have been people who've used adafruit powerboost to overcome the intel breakout board limitations
I don't recall the details
Neal
@tnharvey
Oct 26 2016 18:10
I've seen the explorer board mentioned elsewhere, but can't seem to find it online, except on github. Is it something that can be purchased?
Dana Lewis
@danamlewis
Oct 26 2016 18:10
Yep
Scott Leibrand
@scottleibrand
Oct 26 2016 18:11
:)
I've given in and just left it open in a tab
Neal
@tnharvey
Oct 26 2016 18:11
lol, thanks @scottleibrand
Dana Lewis
@danamlewis
Oct 26 2016 18:11
Context: https://diyps.org/2016/09/15/openaps-rigs-are-shrinking-in-size/ (also FYI for anyone in future the link is in this blog post)
Neal
@tnharvey
Oct 26 2016 18:11
and @danamlewis
Paul Martin
@Spazholio
Oct 26 2016 18:13
So has anyone started work on a case for the Explorer board yet? I’m not kidding when I say that @eluhrs has ruined me for cases, but I’m game for any others to try. =)
garykidd
@garykidd
Oct 26 2016 18:13
@Spazholio I think that owerboost can step up voltage of your 3.7v to 5v for OTG port....
Paul Martin
@Spazholio
Oct 26 2016 18:13
@garykidd I have no idea how one would accomplish something like that. I’m game to give it a try if there are directions. Preferably with pictures. =)
garykidd
@garykidd
Oct 26 2016 18:13
@Spazholio meaing you like his designs that much...
Paul Martin
@Spazholio
Oct 26 2016 18:13
Big time.
Dana Lewis
@danamlewis
Oct 26 2016 18:14
If anyone has any designs they'd like me to test before boards ship in November; I can give you my address ;)
Paul Martin
@Spazholio
Oct 26 2016 18:15
Dana tends to be REALLY busy, so feel free to send one to me as well.
Dana Lewis
@danamlewis
Oct 26 2016 18:16
Lol.
Paul Martin
@Spazholio
Oct 26 2016 18:16
And since I’ve named myself an official case beta tester, I’m gonna need an Explorer board sooner than expected, right? ;)
Dana Lewis
@danamlewis
Oct 26 2016 18:16
;)
John Sjolund
@sjolundjohn
Oct 26 2016 18:34
I use the Powerboost with the Edison, works great
Neal
@tnharvey
Oct 26 2016 18:35
sorry if this is off topic for here, but my Plugins in NS aren't working. They disappeared from the menu as well. Just in my desktop browser, still there on my phone. Any reason for that? Refresh doesn't change it.
John Sjolund
@sjolundjohn
Oct 26 2016 18:36

Standard Li Ion battery that goes straight into the PowerBoost via a JST connector. Charges via standard Micro-USB.

The tricky part is I needed to solder x2 cables on PowerBoost that would connect to the Edison via J21. Works great. First solder in my life so anybody can accomplish it

Neal
@tnharvey
Oct 26 2016 18:36
Although it could just be my system. It's my work system and it's getting so spotty that I'm ready to do a fresh install
Oh and by plugins, I mainly concerned with the OpenAPS related plugins
@sjolundjohn I had a spare JST that was a little larger and perfectly fit the posts on J2 for the edison, so I soldered those to the LiPo charger, and it makes it easier to pull on and off if I need to. Probably not the best thing long term, but it's nice right now while things are temporary =]
Scott Leibrand
@scottleibrand
Oct 26 2016 18:43
@neal try a couple more reloads of NS
Dana Lewis
@danamlewis
Oct 26 2016 18:45
If refreshes don't ultimately work, wonder if you need to clear device status in mlab. @sulkaharo suggested that for someone the other day, if you're using free tier
Sulka Haro
@sulkaharo
Oct 26 2016 18:48
Plugins disappearing from the view sounds weird though. Clearing cookies could reset the local config for the browser, if that's the issue
For general refresh issues with openaps, it's quite likely that's caused by mongo not scaling. Self hosting on Digital Ocean for $5 a month scales perfectly, the free tier in mLab has maybe a tenth of those resources (or less)
Using Admin Tools / clear device status collection can help in most cases as that collection is semi-expendable data and he hardest for Mongo to process
diabeticgonewild
@diabeticgonewild
Oct 26 2016 19:04
If you get the right sized battery (preferrably a smaller capacity that lasts around 20 hours), an Edison with RL can fit in this case. I would imagine the Explorer board would fit in this too.
emsamom
@emsamom
Oct 26 2016 19:13
New question - if I enter carbs/bolus info to Careportal via nightscout, does this info come to openaps the same as if I entered them into the pump?
Neal
@tnharvey
Oct 26 2016 19:14
@scottleibrand I've refreshed a few times, rebooting to see if that helps, but so far seems not to be sorting itself out.
Dana Lewis
@danamlewis
Oct 26 2016 19:14
@emsamom yes if your OpenAPS rig is connected to Internet it will pull from NS
Neal
@tnharvey
Oct 26 2016 19:14
@danamlewis I'll try that next, would "clearing device status" essentially be deleting all the documents in the device status collection?
emsamom
@emsamom
Oct 26 2016 19:15
does the NS data and the Pump data on carbs/bolus get stored in the same file? (my guess is no but I'm stumbling around the directories)
Neal
@tnharvey
Oct 26 2016 19:15
@sulkaharo Getting to a cheap paid service has been on my radar. I've hear a lot of people referring to heroku, but I haven't heard mention of digital ocean. Any reason to prefer one over the other?
emsamom
@emsamom
Oct 26 2016 19:16
@tnharvey I'll give a vote of support for heroku - was incredibly easy to set up
Matthew Williams
@mwilliams
Oct 26 2016 19:17
@danamlewis - Great success (mostly) with the dev branch of ref0. Both ns-loop and pump-loop churning along. However, only data making it to NS appears to be my temp basal rates. No glucose data. Logging from each looks pretty reasonable, however of the three files data dumped into the cgm directory (cgm-glucose.json, glucose.json, and ns-glucose.json), ns-glucose is just an empty array, while the other two are full of data. Trying to follow the breadcrumbs to see how ns-glucose comes to be
I'm also getting bolus markers in NS as well which is good
Mike
@libxmike
Oct 26 2016 19:19
@/all who struggle with Rpi (in my case Zero) and Carelink and the death stick after few to maybe 9h, the only solution is to power cycle (reboot,hub_ctrl etc. Does not work) I will do it now with transistor and GPIO HIGH/LOW , transistor on USB power line (cut power line plus, transistor in between and GPIO to control) python to control transistor over GPIO, for now the only way to power cycle
Scott Leibrand
@scottleibrand
Oct 26 2016 19:21
@mwilliams the dev branch doesn't do BG upload
Matthew Williams
@mwilliams
Oct 26 2016 19:21
@scottleibrand: Ha, well that'd explain it...
@scottleibrand: I'm running a fresh loop with the new setup script in the dev branch
Scott Leibrand
@scottleibrand
Oct 26 2016 19:23
There is a WIP in the oref0-setup branch, but it also loops with raw data, which needs to be disabled by default before we merge it.
Matthew Williams
@mwilliams
Oct 26 2016 19:26
Interesting, OK - so if I want to get glucose and pump status to NS - I should have all the glucose data (I may need reports for battery, reservoir, etc etc), yes? And I'll just look at crafting an additional ns-upload report to push it up?
Scott Leibrand
@scottleibrand
Oct 26 2016 19:27
Please do it right using the ns device not ns-upload
Someone needs to re-do what's in oref0-setup to use that.
Matthew Williams
@mwilliams
Oct 26 2016 19:29
alright, I'll poke around for some prior art with the upload then - do you know if either cgm-glucose.json or glucose.json is formatted properly for NS (especially since mine is enlite data, don't know if it needs to be massaged)?
Neal
@tnharvey
Oct 26 2016 19:32
@danamlewis @scottleibrand looks like it was Chrome related, still not sorted out, but FireFox and IE are normal. Probably something with the cookies like you said. Mmmm. Cookies. Is it lunchtime?
Neal
@tnharvey
Oct 26 2016 19:38
yep! Cookies did it!
cassknox
@cassknox
Oct 26 2016 21:20
pi@raspberrypi:~/myopenaps $ openaps do-everything
PREFLIGHT
Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
File "/home/pi/src/openaps/bin/openaps-report", line 82, 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-report", line 69, in epilog
super(ReportToolApp, 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 83, in create_git_commit
if self.repo.is_dirty( ) or self.repo.index.diff(None):
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.8-py2.7.egg/git/repo/base.py", line 608, in is_dirty
len(self.git.diff('--cached', default_args)):
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.8-py2.7.egg/git/cmd.py", line 466, in <lambda>
return lambda
args, kwargs: self._call_process(name, *args, kwargs)
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.8-py2.7.egg/git/cmd.py", line 934, in _call_process
return self.execute(make_call(), **_kwargs)
File "/usr/local/lib/python2.7/dist-packages/GitPython-2.0.8-py2.7.egg/git/cmd.py", line 718, in execute
raise GitCommandError(command, status, stderr_value)
git.exc.GitCommandError: 'git diff --cached --abbrev=40 --full-index --raw' returned with exit code 128
stderr: 'error: object file .git/objects/c8/bb35c6487c9a37afefe96dbd42bc8d383955ba is empty
error: object file .git/objects/c8/bb35c6487c9a37afefe96dbd42bc8d383955ba is empty
fatal: loose object c8bb35c6487c9a37afefe96dbd42bc8d383955ba (stored in .git/objects/c8/bb35c6487c9a37afefe96dbd42bc8d383955ba) is corrupt'
PREFLIGHT FAIL
Can anyone help me out with this?
@scottleibrand if you're around, can you check this out and let me know what you think? Im not sure if you're up in the bay area with Chris and Ben, but now I'm flying solo over here and Ben has re-configured my Pi and now I'm not so sure how to fix it... Help!
Diabetic_Dude
@diabetic_dude_twitter
Oct 26 2016 21:29
I am reading this section of setting up my APS http://www.nightscout.info/wiki/labs/how-to-heroku I noticed that there is no reference to setting up a DB on MLAB... is the db @ MLAB set up when you deploy to heroku from git?
diabeticgonewild
@diabeticgonewild
Oct 26 2016 21:29
I hate the ambiguous errors python gives
Matt Walker
@walker0
Oct 26 2016 21:30
@diabetic_dude_twitter yes heroku automatically sets up its own Mlab if you follow the instructions
diabeticgonewild
@diabeticgonewild
Oct 26 2016 21:30
@diabetic_dude I believe I have heard that you can set up mlab either alone at mlab.com or within Heroku itself.
I would set it up externally cause if you use multiple sites, it's to your benefit.
Diabetic_Dude
@diabetic_dude_twitter
Oct 26 2016 21:31
I have it set up externally @mlab but the data is not getting seen by heroku
it is driving me batty, my mongodb is sitting there collecting the data but heroku will not see it and there are no darn logs
Paul Martin
@Spazholio
Oct 26 2016 21:40
@cassknox Sounds like you’ve got git corruption going on. Do you have the oref0-reset-git command in your crontab? Or if not, have you tried running that command in your openaps directory?
emsamom
@emsamom
Oct 26 2016 21:44

@diabetic_dude_twitter if you do the install by button click from github it will create an MLAB database for you - I ended up with two, but just used the one automatically created with the Heroku install.

Anyone know how we make an update to the Nightscout documentation? This should be called out.

@diabetic_dude_twitter and unfortunately I don't know how to associate heroku to a different DB, but I'm sure that's feasible too
Diabetic_Dude
@diabetic_dude_twitter
Oct 26 2016 21:46
@diabeticgonewild thanks, for now I am going to change some info on my phone and see if it uploads to heroku as opposed to the one I created.
Matt Walker
@walker0
Oct 26 2016 21:46
@diabetic_dude_twitter assuming you only want to use one Mlab that could be used for azure or heroku then in heroku you probably want to overview, then configure add ons then edit next to m lab and remove that and then add your mongo data base in settings as you would in Azure
Diabetic_Dude
@diabetic_dude_twitter
Oct 26 2016 21:47
@emsamom heck ask long as I know this, all I ahve to do is change some info on my uploader and see if it popluates heroku
awedixon31
@awedixon31
Oct 26 2016 21:51

so when I try to run the alias get-ns-bg (get-ns-bg ! bash -c "openaps get-ns-glucose && cat cgm/ns-glucose.json | json -c \"minAgo=(new Date()-new Date(this.dateString))/60/1000; return minAgo < 10 && minAgo > -5 && this.glucose > 30\" | grep -q glucose && cp -pu cgm/ns-glucose.json cgm/glucose.json; cp -pu cgm/glucose.json monitor/glucose.json")

cp: cannot stat ‘cgm/glucose.json’: No such file or directory

I thought the point of the end of that script is to copy/replace cgm/glucose.json if older OR non-existent. What am I missing?

Diabetic_Dude
@diabetic_dude_twitter
Oct 26 2016 21:54
@walker0 thanks for the tip, I copied that sentence and try that if changing the upload config on my phone does not work
garykidd
@garykidd
Oct 26 2016 21:58
what's the command for monitoring edison battery voltage
Paul Martin
@Spazholio
Oct 26 2016 22:30
That’s the breakout board, but not the Edison itself.
Matt Walker
@walker0
Oct 26 2016 22:44
@Spazholio it has a picture of the actual edison and the specs so I'd hope it came with one. Especially for that price
Paul Martin
@Spazholio
Oct 26 2016 22:45
Oh hey, you’re right. My bad.
garykidd
@garykidd
Oct 26 2016 22:48
@Spazholio how'd you get battery voltage of edison?
emsamom
@emsamom
Oct 26 2016 22:55
@Spazholio @garykidd - I think this is for the board and Edison..... I was looking into the preorder Edison explorer board that @danamlewis posted earlier, and I think I don't need the board if I'm going to go that route.
Paul Martin
@Spazholio
Oct 26 2016 22:56
@garykidd Gotta install/run this: https://github.com/cjo20/EdisonVoltage
Yeah, but it’s not a bad idea to get both. If the explorer board goes bad or something, it’s good to have a backup.
I think I have dupes of each piece of hardware involved in setting up a rig.
emsamom
@emsamom
Oct 26 2016 22:57
Can you take the edison off the board sold by amazon easily? (I'm more a build it once, see if it works, then order backups type developer)
Paul Martin
@Spazholio
Oct 26 2016 22:57
Oh yeah.
$50.
So for an extra $9, you can get a BO board. Worth it.
R. Dennis Steed
@rdsteed
Oct 26 2016 23:00
For a bargain, try microcenter - they are selling it with the breakout board for less than the Edison alone: http://www.microcenter.com/product/458394/Edison_Breakout_Board_Kit
Paul Martin
@Spazholio
Oct 26 2016 23:01
@rdsteed Good call. I think I actually got it from them when I bought mine. Even with the $6 shipping, it’s still a deal.
emsamom
@emsamom
Oct 26 2016 23:01
Thanks team!
emsamom
@emsamom
Oct 26 2016 23:07
Sooooo close. How do I charge the LiPO battery?

Will this work? https://www.adafruit.com/product/259

And can I charge the battery while It's plugged into the Edison?

Matt Walker
@walker0
Oct 26 2016 23:13
@emsamom yes and yes. That particular one is a bit bigger than you need though. Mine is running a Micro lipo which works fine and I know many others are using the powerboost 1000c
emsamom
@emsamom
Oct 26 2016 23:14
What's the right mAH to get through a 14-16 hour day?
Matt Walker
@walker0
Oct 26 2016 23:15
You will want something 2000mah+ to reliably get a full day
emsamom
@emsamom
Oct 26 2016 23:15
I'm guessing it's the one adafruit is out of stock on!
Screen Shot 2016-10-26 at 7.15.32 PM.png
Matt Walker
@walker0
Oct 26 2016 23:16
you can get some really nice 3500mah panasonic 18650 batteries if you are happy with the round form factor.
Just make sure whatever you buy has a protection circuit that prevent overcharge/discharge and short circuit
diabeticgonewild
@diabeticgonewild
Oct 26 2016 23:19
I just ordered the new Explorer board. Do we need LiPo in addition to the Explorer board? Also, when are they expected to ship?
Scott Leibrand
@scottleibrand
Oct 26 2016 23:36
Yes, you'll need a LiPo with a JST connector, and the Edison itself. Sometime in November.
cassknox
@cassknox
Oct 26 2016 23:36
@Spazholio when i run oref0-reset-git in my opens directory it returns this:
fatal: bad object HEAD
error: object file .git/objects/9e/d497b137f3c6432dedfefbf54f1307ac2c7222 is empty
error: object file .git/objects/9e/d497b137f3c6432dedfefbf54f1307ac2c7222 is empty
fatal: loose object 9ed497b137f3c6432dedfefbf54f1307ac2c7222 (stored in .git/objects/9e/d497b137f3c6432dedfefbf54f1307ac2c7222) is corrupt
fatal: bad object HEAD
fatal: bad object HEAD
Again
Paul Martin
@Spazholio
Oct 26 2016 23:37
@scottleibrand might be able to shed more light, but maybe try it again, and then run the loop to see if it helped?
cassknox
@cassknox
Oct 26 2016 23:39
okay, from here how do i try and run the loop again?
okay @Spazholio heres what i got:

pi@raspberrypi:~/myopenaps $ openaps do-everything
PREFLIGHT
PREFLIGHT OK
cgm://JSON/oref0_glucose/monitor/glucose-raw.json
reporting monitor/glucose-raw.json
tz://JSON/rezone/monitor/glucose.json
reporting monitor/glucose.json
pump://JSON/read_temp_basal/monitor/temp-basal-status.json
monitor/temp-basal-status.json raised /dev/serial/by-id/usb-0a21_8001-if00-port0 already in use
Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
File "/home/pi/src/openaps/bin/openaps-report", line 82, in <module>
app( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 51, in call
self.run(self.args)
File "/home/pi/src/openaps/bin/openaps-report", line 75, in run
output = app(args, self)
File "/home/pi/src/openaps/openaps/cli/subcommand.py", line 52, in call
return self.method.main(args, app)
File "/home/pi/src/openaps/openaps/reports/invoke.py", line 40, in main
output = task.method(args, app)
File "/home/pi/src/openaps/openaps/uses/use.py", line 44, in call
self.before_main(args, app)
File "/home/pi/src/openaps/openaps/vendors/medtronic.py", line 61, in before_main
self.setup_medtronic( )
File "/home/pi/src/openaps/openaps/vendors/medtronic.py", line 159, in setup_medtronic
self.uart = stick.Stick(link.Link(self.scanner( )))
File "/home/pi/src/decocare/decocare/link.py", line 23, in init
raise AlreadyInUseException("{port} already in use".format(port=port))
decocare.link.AlreadyInUseException: /dev/serial/by-id/usb-0a21_8001-if00-port0 already in use
pump://JSON/reservoir/monitor/reservoir.json
monitor/reservoir.json raised /dev/serial/by-id/usb-0a21_8001-if00-port0 already in use
Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
File "/home/pi/src/openaps/bin/openaps-report", line 82, in <module>
app( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 51, in call
self.run(self.args)
File "/home/pi/src/openaps/bin/openaps-report", line 75, in run
output = app(args, self)
File "/home/pi/src/openaps/openaps/cli/subcommand.py", line 52, in call
return self.method.main(args, app)
File "/home/pi/src/openaps/openaps/reports/invoke.py", line 40, in main
output = task.method(args, app)
File "/home/pi/src/openaps/openaps/uses/use.py", line 44, in call
self.before_main(args, app)
File "/home/pi/src/openaps/openaps/vendors/medtronic.py", line 61, in before_main
self.setup_medtronic( )
File "/home/pi/src/openaps/openaps/vendors/medtronic.py", line 159, in setup_medtronic
self.uart = stick.Stick(link.Link(self.scanner( )))
File "/home/pi/src/decocare/decocare/link.py", line 23, in init
raise AlreadyInUseException("{port} already in use".format(port=port))
decocare.link.AlreadyInUseException: /dev/serial/by-id/usb-0a21_8001-if00-port0 already in use
pump://JSON/iter_pump_hours/raw-pump/pump-history-raw.json
raw-pump/pump-history-raw.json raised ACK is 0 bytes:

Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
File "/home/pi/src/openaps/bin/openaps-report", line 82, in <module>
app( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 51, in call
self.run(self.args)
File "/home/pi/src/openaps/bin/openaps-report", line 75, in run
output = app(args, self)
File "/home/pi/src/openaps/openaps/cli/subcommand.py", line 52, in call
return self.method.main(args, app)
File "/home/pi/src/openaps/openaps/reports/invoke.py", line 40, in main
output = task.method(args, app)
File "/home/pi/src/openaps/openaps/uses/use.py", line 44, in call
self.before_main(args, app)
File "/home/pi/src/openaps/openaps/vendors/medtronic.py", line 61, in before_main
self.setup_medtronic( )
File "/home/pi/src/openaps/openaps/vendors/medtronic.py", line 160

Paul Martin
@Spazholio
Oct 26 2016 23:42
monitor/temp-basal-status.json raised /dev/serial/by-id/usb-0a21_8001-if00-port0 already in use <— that’s your problem. Unfortunately, I’m unsure how to fix it. Have you already put your loop into cron? Was it already running maybe?
cassknox
@cassknox
Oct 26 2016 23:44
hmm, im not sure? i mean, it was running at one time... but it has not been working for over a week now.
it's not working now because there's no temp basals running
R. Dennis Steed
@rdsteed
Oct 26 2016 23:45
You could try unplugging the stick and plugging it back in.
Paul Martin
@Spazholio
Oct 26 2016 23:46
Oh yeah - what are you using? TI? Carelink?
cassknox
@cassknox
Oct 26 2016 23:46
Carelink
Paul Martin
@Spazholio
Oct 26 2016 23:46
On a Pi?
cassknox
@cassknox
Oct 26 2016 23:46
Yes, RPi2
Paul Martin
@Spazholio
Oct 26 2016 23:46
Yeah, try unplugging it and re-plugging it.
Then try the loop again.
cassknox
@cassknox
Oct 26 2016 23:47
Okay I just did that, trying again...
pi@raspberrypi:~/myopenaps $ openaps do-everything
PREFLIGHT
raw-pump/clock-raw.json raised 'Pump' object has no attribute 'model'
Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
File "/home/pi/src/openaps/bin/openaps-report", line 82, in <module>
app( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 51, in call
self.run(self.args)
File "/home/pi/src/openaps/bin/openaps-report", line 75, in run
output = app(args, self)
File "/home/pi/src/openaps/openaps/cli/subcommand.py", line 52, in call
return self.method.main(args, app)
File "/home/pi/src/openaps/openaps/reports/invoke.py", line 40, in main
output = task.method(args, app)
File "/home/pi/src/openaps/openaps/uses/use.py", line 45, in call
output = self.main(args, app)
File "/home/pi/src/openaps/openaps/vendors/medtronic.py", line 327, in main
return self.pump.model.read_clock( )
AttributeError: 'Pump' object has no attribute 'model'
PREFLIGHT FAIL
taylorfowler
@taylorfowler
Oct 26 2016 23:48
@emsamom I think that adafruit battery is in stock on Amazon
(not 100% but it's worth a try)
cassknox
@cassknox
Oct 26 2016 23:53
why can it not read my pump model now?
i just moved the stick closer to my pump and...
pi@raspberrypi:~/myopenaps $ openaps do-everything
PREFLIGHT
PREFLIGHT OK
cgm://JSON/oref0_glucose/monitor/glucose-raw.json
reporting monitor/glucose-raw.json
tz://JSON/rezone/monitor/glucose.json
reporting monitor/glucose.json
pump://JSON/read_temp_basal/monitor/temp-basal-status.json
reporting monitor/temp-basal-status.json
pump://JSON/reservoir/monitor/reservoir.json
reporting monitor/reservoir.json
pump://JSON/read_status/monitor/status.json
reporting monitor/status.json
pump://JSON/read_battery_status/monitor/battery.json
reporting monitor/battery.json
pump://JSON/iter_pump_hours/raw-pump/pump-history-raw.json
reporting raw-pump/pump-history-raw.json
tz://JSON/rezone/monitor/pump-history.json
reporting monitor/pump-history.json
pump://JSON/read_bg_targets/raw-pump/bg-targets-raw.json
reporting raw-pump/bg-targets-raw.json
units://JSON/bg_targets/settings/bg-targets.json
reporting settings/bg-targets.json
pump://JSON/read_insulin_sensitivities/raw-pump/insulin-sensitivities-raw.json
reporting raw-pump/insulin-sensitivities-raw.json
units://JSON/insulin_sensitivities/settings/insulin-sensitivities.json
reporting settings/insulin-sensitivities.json
pump://JSON/read_selected_basal_profile/raw-pump/selected-basal-profile.json
reporting raw-pump/selected-basal-profile.json
oref0://JSON/shell/settings/selected-basal-profile.json
reporting settings/selected-basal-profile.json
pump://JSON/read_settings/raw-pump/settings.json
reporting raw-pump/settings.json
oref0://JSON/shell/settings/settings.json
reporting settings/settings.json
get-profile://text/shell/oref0-monitor/profile.json
Could not sort temptargets_data. Optional feature temporary targets disabled.
reporting oref0-monitor/profile.json
calculate-iob://text/shell/oref0-monitor/iob.json
reporting oref0-monitor/iob.json
determine-basal://text/shell/oref0-predict/oref0.json
{"iob":5.061,"activity":0.0342,"bolussnooze":3.03,"basaliob":0,"netbasalinsulin":0,"hightempinsulin":0,"time":"2016-10-26T23:54:04.000Z"}
{"delta":-9,"glucose":130,"short_avgdelta":-9.94,"long_avgdelta":-5.29}
{"duration":13,"rate":1.3,"temp":"absolute"}
Adjusting target_bg from 87.5 to 88
Carb Impact: -4.8 mg/dL per 5m; CI Duration: NaN hours
Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: NaN hours
reporting oref0-predict/oref0.json
pump://JSON/set_temp_basal/oref0-enacted/enacted-temp-basal.json
reporting oref0-enacted/enacted-temp-basal.json
ns://JSON/shell/nightscout/recent-treatments.json
reporting nightscout/recent-treatments.json
ns://JSON/shell/nightscout/uploaded.json
Nothing to upload.
reporting nightscout/uploaded.json
ns://JSON/shell/nightscout/recent-missing-entries.json
reporting nightscout/recent-missing-entries.json
ns://JSON/shell/nightscout/uploaded-entries.json
reporting nightscout/uploaded-entries.json
Paul Martin
@Spazholio
Oct 26 2016 23:58
There you go.
And the Carelink has notoriously short range, and the orientation can affect its efficacy.