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

31st
May 2017
garykidd
@garykidd
May 31 2017 00:12
strange intel edison compute modules have gone up in price quite a bit...had purchased 3 in october of 16 for 42 each..now 53...is that the best deal pretty much now?
Dana Lewis
@danamlewis
May 31 2017 00:13
seems to ebb and flow
garykidd
@garykidd
May 31 2017 00:13
Cool....just got 2 more explorer boardin the mail tonight...yeah...oref1 to be tested soon
rsilvers129
@rsilvers129
May 31 2017 00:23
I paid $37 in April 2017 for my first Edison and $53 for my second.
Dana Lewis
@danamlewis
May 31 2017 00:23
:+1:
rsilvers129
@rsilvers129
May 31 2017 00:29
My myopenaps directory for dev doesn’t have an autotune directory. Is that normal? Should I create one?
Dana Lewis
@danamlewis
May 31 2017 00:29
@garykidd when are you planning to setup your new rigs?
David
@davidkeddydb
May 31 2017 00:36
@danalewis Just an FYI that I just changed the code for 484 and will advise after autotune runs tonight if it worked. (Hopefully I vi'd in and changed the code the right way)
Dana Lewis
@danamlewis
May 31 2017 00:37
@davidkeddydb :+1: great, thanks!
sounds like several people testing tonight; if confirmed, will end up being able to merge tomorrow!
David
@davidkeddydb
May 31 2017 00:46
@danamlewis I have a rig that was originally built in Oct/Nov 16 with the master code set back then. Is there a repository that has all the old code? I am just curious to see the differences with bolus snooze in particular in the code. I can just poke around in my old rig and compare to the new, but I gather if I really new what I was doing in github I could likely just see the changes.
Dana Lewis
@danamlewis
May 31 2017 00:47
@davidkeddydb my way is to use github to refer back to previous versions. the command line lovers will tell you you can do the same on the command line, but if you’re not a command line pro, github poking around works, too
David
@davidkeddydb
May 31 2017 00:49
@danamlewis That I can do.
David
@davidkeddydb
May 31 2017 01:42
My rig with autotune is erroring out getting connected so I may not be able to test that fix as I thought: Starting supermicrobolus pump-loop at Tue May 30 21:39:05 EDT 2017 with 13 second wait_for_silence:
Waiting up to 4 minutes for new BG: glucose.json newer than pump_loop_completed
Radio ok. Listening: .No pump comms detected from other rigs
mmeowlink.exceptions.CommsException: No acknowledgement from pump on wakeup. Is it out of range or is the battery too low?
Error, retrying
It is communicating with one rig but can't seem to get going on the autotune one
David
@davidkeddydb
May 31 2017 01:50
It just took 20 minutes but it is running. Report back tomorrow on the autotune stuff
cameronrenwick
@cameronrenwick
May 31 2017 02:10
@ericfeibelman here is the approach I took with my endo. I said to him look, if I wanted to make a change to my basal every 5 minutes - I could. The decision-making that I do on a daily basis is simply being programmed into this little computer. The computer, following my direction, is doing what I'd normally do, only on a much more timely basis. This, in my opinion, is what a (non-T1D) pancreas does.... so why wouldn't a doctor support this approach? Honestly it's because a) they loose control of how your diabetes is managed (but at the same time they want you to manage your own health) and b) possibly because they don't have a fulsome an understanding of what openAPS is all about. That, they can learn, as you have. When I presented this to my endo he listened carefully. I explained that what I was going to do (and in fact "had" been doing) I wasn't going to stop. It wasn't a threat per se, but more that I believed so strongly in what I was doing. After a 20 minute convo he said, ok, I'm behind you. Ultimately doctors and you and I want the same thing - healthy patients. Good luck with your visit and report back!
mariano42
@mariano42
May 31 2017 02:47
hi all, running the setup script with bluetooth for the first time and it seemed to run into an error. It ran for almost 30 minutes and then popped up with: "couldnt copy bluetoothd.conf" and brought me back to my command line without going to the prompt about the cron or rebooting. This is the first time I am trying it with BT so I am not sure what to expect in terms of how long this script should take... should I just try to run again?
Russell Gardner
@RussGG
May 31 2017 02:51
Just check my Heroku NS site and there is almost nothing there but the time, IOB, Basal, BWP and COB pills. My son his with his mom right now so I can’t check the rig. Papertrail stopped at 3 this afternoon. I have been using this set up for months and even when his BG isn’t uploading or rig connecting I still have much more than this. Any thoughts?
JELCRAWFORD
@JELCRAWFORD
May 31 2017 04:25
@ericfeibelman - our approach with our son's endo was to send an email to her (you could drop off a package if you don't have your endo's email) in advance of our appointment with some basic info about OpenAPS, links to the documents online, and bunch of selected data/graphs from Night Scout. We prefaced it with wanting to update her in advance of our appointment about our going's on and wanting to make the best use of her time when we met. We gave a brief run down of our start up and how everything had gone. We too focused on the fact that the pancreas does, every 5 minutes, using math what we could do, except that we really couldn't do it as well or indefinately as a computer can. We stated we were looking forward to sharing our journey with her when we came the next week. Now, bear in mind, at our previous appointment, 3 months before, we showed up having hooked up a used insulin pump ourselves 2 months previous, with data in hand from NS supporting the risk free start up and history to that point - so I expect that our Endo is not anticipating the most typical of presentations from us. Her overall response has been to be very interested in what we are up to. The last appt we again sent info in about 1 week in advance and again, she had reviewed it all prior to our arrival and she has consistently shared it with our team as well. She did say that she couldn't 'officially' endorse off label use, but that she was excited to see how the system works and she is fully aware and expectant that the closed loop (broadly) is the way of future treatment. We currently send her data about every 6 or 8 weeks with the caveat statement at the start of the email that our data is just to update her between appts and requires no reply from her, acknowledging that she is very busy. We have had a short reply each time and some comment about her interest and curiosity. Overall we feel we are well received and it is pretty hard for there to be push back against the data provided from NS showing the outstanding results. Good luck!
Paul Dickens
@thebookins
May 31 2017 04:42
@tim2000s Any luck with the error message above? I'm afraid I haven't seen that before. I don't think I changed anything in the code that would affect certificates and provisioning - just a bit of code that uploads the BGs to a NS instance (which in our case happens to be the microservice exposed by the edison).
Tim Street
@tim2000s
May 31 2017 05:14
Sadly not Paul. Hopefully getting some help with it.
Martin Haeberli
@mhaeberli
May 31 2017 05:19
@rsilvers129 @scottleibrand indeed IMHO Scott's changes to oref0-online have helped
Can anyone comment on risks/ benefits of having two phones each connected to the same G5 and sharing w dexcom cloud?
rsilvers129
@rsilvers129
May 31 2017 05:21
I just did a fresh install on a friend's Edison, and tried dev, and got the @ date error.
katie disimone
@Kdisimone
May 31 2017 05:22

@scottleibrand having a wee problem tonight. Setup a new rig from scratch (flashed and everything)…setup loop using daughter’s exsiting pump and NS sites (which are currently using SMB on other rigs). The new rig however was NOT smb-enabled. Within about 15 minutes…decided to move the new rig (call it edison1) to a backup pump and test NS site.

Edison1’s loop seemed to have some problems, so I deleted the myopenaps dir, crontab and reinstalled loop from setup script (not runagain). Still have the same problem…COB undefined. (AMA is enabled)

Temp duration 27m or more
glucose.json newer than pumphistory: Refreshed pumphistory
{"carbs":0,"reason":"not enough glucose data to calculate carb absorption"}
{"iob":0.533,"activity":0.0026,"bolussnooze":0,"basaliob":0.533,"netbasalinsulin":0.6,"hightempinsulin":0.85,"microBolusInsulin":0.2,"microBolusIOB":0.18,"time":"2017-05-31T05:16:56.000Z","lastBolusTime":1496205048000}
{"delta":-9,"glucose":63,"short_avgdelta":-8.61,"long_avgdelta":-6.74}
{"duration":117,"rate":0,"temp":"absolute”}
But, this part of the loop log also made me wonder…how does Edison1 know about “microbolusIOB” Where is it getting that number from? This pump has never given a smb. The NS site had it’s mlab emptied out before starting this. The myopenaps dir had been deleted. Is the rig storing data someway I don’t understand?
Feels like I must have a time error somewhere? Pump time is 10:23pm May 30, 2017. NS site profile is set to US/Pacific New. Rig is set to Us/Pacific New.
I guess I’ll go manual and see
Tim Street
@tim2000s
May 31 2017 05:28
@mhaeberli I'm not sure it's possible to do that.
As far as I am aware, G5 is able to do a phone and receiver, but not two phones.
Dana Lewis
@danamlewis
May 31 2017 05:29
@Kdisimone those old 2g might have expired? Want to put in a bigger number to test it?
katie disimone
@Kdisimone
May 31 2017 05:29
:+1:
Scott Leibrand
@scottleibrand
May 31 2017 05:33
@Kdisimone if you're running dev code, any bolus smaller than 30m worth of basal is considered a microbolus, whether SMB is enabled or not.
Martin Haeberli
@mhaeberli
May 31 2017 05:33
@tim2000s thx I'll post any new learnings...
katie disimone
@Kdisimone
May 31 2017 05:33
ahhh
Dana Lewis
@danamlewis
May 31 2017 05:35
Need to add that to the docs. Will confuse for the release, probably. Like the openaps-0.2.0 fun.
PieterGit
@PieterGit
May 31 2017 05:39
@tw_miiko_twitter : the /usr/local/bin/oref0-subg-ww-radio-parameters-timeout: kill no process found: (something like that) can be ignored. on the 0.4.2 release the script that will make sure the ww pump setting scripts ends. on the next release that message will be gone
katie disimone
@Kdisimone
May 31 2017 05:42
cob still undefined. Large (13g and 1.7 bolus) has been read and looped
Dana Lewis
@danamlewis
May 31 2017 05:46
:confused: the only thing AMA related that changed was moving it up out of the sub-menu... ::keeps looking around::
katie disimone
@Kdisimone
May 31 2017 05:51
root@edison1:~/myopenaps# openaps get-ns-bg
ns://JSON/shell/cgm/ns-glucose.json
reporting cgm/ns-glucose.json
root@edison1:~/myopenaps# openaps monitor-pump
pump://JSON/read_clock/monitor/clock.json
reporting monitor/clock.json
pump://JSON/read_temp_basal/monitor/temp_basal.json
reporting monitor/temp_basal.json
pump://JSON/iter_pump_hours/monitor/pumphistory.json
reporting monitor/pumphistory.json
tz://JSON/rezone/monitor/pumphistory-zoned.json
reporting monitor/pumphistory-zoned.json
tz://JSON/clock/monitor/clock-zoned.json
reporting monitor/clock-zoned.json
iob://text/shell/monitor/iob.json
reporting monitor/iob.json
meal://text/shell/monitor/meal.json
Optional feature meal assist disabled: not enough glucose data to calculate carb absorption; found: 21
reporting monitor/meal.json
pump://JSON/reservoir/monitor/reservoir.json
reporting monitor/reservoir.json
pump://JSON/read_battery_status/monitor/battery.json
reporting monitor/battery.json
pump://JSON/status/monitor/status.json
reporting monitor/status.json
feels like user error still….
Dana Lewis
@danamlewis
May 31 2017 05:51
Ohh, maybe have an idea. One min
katie disimone
@Kdisimone
May 31 2017 05:53
root@edison1:~/myopenaps# openaps report invoke monitor/meal.json
meal://text/shell/monitor/meal.json
Optional feature meal assist disabled: not enough glucose data to calculate carb absorption; found: 21
reporting monitor/meal.json
root@edison1:~/myopenaps# cat monitor/meal.json
{ "carbs": 0, "reason": "not enough glucose data to calculate carb absorption" }
Dana Lewis
@danamlewis
May 31 2017 05:56
hmm, no that wasn’t it
Scott Leibrand
@scottleibrand
May 31 2017 05:56
where is this rig getting its glucose data?
Optional feature meal assist disabled: not enough glucose data to calculate carb absorption; found: 21
if you're using g4-upload, you'll need to wait for it to collect enough, or manually run the command to get it all
Aaron Michelson
@Aaybob
May 31 2017 05:59
@mhaeberli With my sets, the newer v1.2 is the only one that doesn't turn on when powered, the only one I have to press the button to turn on.
rsilvers129
@rsilvers129
May 31 2017 06:00
Now that I am on dev, autotune has the issue where it don’t show any changes.
Dana Lewis
@danamlewis
May 31 2017 06:01
@rsilvers129 test openaps/oref0#484 & confirm if it resolves the CSF=0
rsilvers129
@rsilvers129
May 31 2017 06:01

Parameter | Current | Autotune

ISF | 32.000 | 32.000
CSF | n/a | 0.000
Carb Ratio | 7.500 | 7.500
00:00 | 0.675 | 0.675
katie disimone
@Kdisimone
May 31 2017 06:06
@scottleibrand ahhhh…there is my user error. I think I have g4-local. (does g4-local not pull from NS?)
Oh, and I see now that as I run the script fresh…g4-local isn’t even an option. g4-local-only is.
Scott Leibrand
@scottleibrand
May 31 2017 06:08
it's a substring match: either should work
is the 21 at the end of the Optional feature meal assist disabled warning incrementing every 5 minutes?
katie disimone
@Kdisimone
May 31 2017 06:09
yes it did
Scott Leibrand
@scottleibrand
May 31 2017 06:09
if so, I wouldn't worry too much about the root cause, and just wait it out, as it's just a problem on newly set up rigs
and it just disabled AMA and autosens until it gets enough glucose history for each
(unless of course you want to do some serious spelunking through openaps configs)
katie disimone
@Kdisimone
May 31 2017 06:10
went to 24…wow, I am usually pretty patient. I guess I underestimated on this one? Edit: pretty patient on new pump loops…everything else, not very
Scott Leibrand
@scottleibrand
May 31 2017 06:11
OpenAPS can be more patient than you: what else is new? ;-)
rsilvers129
@rsilvers129
May 31 2017 06:18
I made that change to autotune code, and it didn’t resolve it for me. It is still showing the same values after running as my pump is set for.
Dana Lewis
@danamlewis
May 31 2017 06:19
did you do a manual run of autotune?
rsilvers129
@rsilvers129
May 31 2017 06:20
Yes.
Dana Lewis
@danamlewis
May 31 2017 06:23
k
and CSF is still 0?
Miiko
@tw_miiko_twitter
May 31 2017 06:37
@PieterGit thank you.
katie disimone
@Kdisimone
May 31 2017 06:55
@scottleibrand was there a renaming somewhere for monitor/cgm-glucose.json vs monitor/glucose.json?
root@edison1:~/myopenaps# openaps refresh-old-pumphistory
Old pumphistory: Refreshed pumphistory
rm: cannot remove ‘enact/suggested.json’: No such file or directory
determine-basal://text/shell/enact/suggested.json
Could not parse input data:  { [Error: Cannot find module '/root/myopenaps/monitor/glucose.json'] code: 'MODULE_NOT_FOUND' }
reporting enact/suggested.json
cat: enact/enacted.json: No such file or directory
root@edison1:~/myopenaps# openaps report invoke monitor/glucose.json
usage: openaps-report invoke [-h]
                             {cgm/cgm-glucose.json,cgm/ns-glucose.json,enact/enacted.json,enact/suggested.json,monitor/battery.json,monitor/cgm-glucose.json,monitor/clock-zoned.json,monitor/clock.json,monitor/iob.json,monitor/meal.json,monitor/mmtune.json,monitor/pumphistory-zoned.json,monitor/pumphistory.json,monitor/reservoir.json,monitor/status.json,monitor/temp_basal.json,raw-cgm/raw-entries.json,settings/autosens.json,settings/basal_profile.json,settings/bg_targets.json,settings/bg_targets_raw.json,settings/carb_ratios.json,settings/insulin_sensitivities.json,settings/insulin_sensitivities_raw.json,settings/model.json,settings/profile.json,settings/pumphistory-24h-zoned.json,settings/pumphistory-24h.json,settings/pumpprofile.json,settings/settings.json,upload/pebble.json,xdrip/glucose.json}
                             [{cgm/cgm-glucose.json,cgm/ns-glucose.json,enact/enacted.json,enact/suggested.json,monitor/battery.json,monitor/cgm-glucose.json,monitor/clock-zoned.json,monitor/clock.json,monitor/iob.json,monitor/meal.json,monitor/mmtune.json,monitor/pumphistory-zoned.json,monitor/pumphistory.json,monitor/reservoir.json,monitor/status.json,monitor/temp_basal.json,raw-cgm/raw-entries.json,settings/autosens.json,settings/basal_profile.json,settings/bg_targets.json,settings/bg_targets_raw.json,settings/carb_ratios.json,settings/insulin_sensitivities.json,settings/insulin_sensitivities_raw.json,settings/model.json,settings/profile.json,settings/pumphistory-24h-zoned.json,settings/pumphistory-24h.json,settings/pumpprofile.json,settings/settings.json,upload/pebble.json,xdrip/glucose.json} ...]
openaps-report invoke: error: argument report: invalid choice: 'monitor/glucose.json' (choose from u'cgm/cgm-glucose.json', u'cgm/ns-glucose.json', u'enact/enacted.json', u'enact/suggested.json', u'monitor/battery.json', u'monitor/cgm-glucose.json', u'monitor/clock-zoned.json', u'monitor/clock.json', u'monitor/iob.json', u'monitor/meal.json', u'monitor/mmtune.json', u'monitor/pumphistory-zoned.json', u'monitor/pumphistory.json', u'monitor/reservoir.json', u'monitor/status.json', u'monitor/temp_basal.json', u'raw-cgm/raw-entries.json', u'settings/autosens.json', u'settings/basal_profile.json', u'settings/bg_targets.json', u'settings/bg_targets_raw.json', u'settings/carb_ratios.json', u'settings/insulin_sensitivities.json', u'settings/insulin_sensitivities_raw.json', u'settings/model.json', u'settings/profile.json', u'settings/pumphistory-24h-zoned.json', u'settings/pumphistory-24h.json', u'settings/pumpprofile.json', u'settings/settings.json', u'upload/pebble.json', u'xdrip/glucose.json’)
katie disimone
@Kdisimone
May 31 2017 07:01
root@edison1:~/myopenaps# openaps report invoke monitor/cgm-glucose.json
cgm://JSON/iter_glucose_hours/monitor/cgm-glucose.json
monitor/cgm-glucose.json  raised  'NoneType' object has no attribute 'iter_records'
Traceback (most recent call last):
  File "/usr/local/bin/openaps-report", line 82, in <module>
    app( )
  File "/usr/local/lib/python2.7/dist-packages/openaps/cli/__init__.py", line 51, in __call__
    self.run(self.args)
  File "/usr/local/bin/openaps-report", line 75, in run
    output = app(args, self)
  File "/usr/local/lib/python2.7/dist-packages/openaps/cli/subcommand.py", line 52, in __call__
    return self.method.main(args, app)
  File "/usr/local/lib/python2.7/dist-packages/openaps/reports/invoke.py", line 40, in main
    output = task.method(args, app)
  File "/usr/local/lib/python2.7/dist-packages/openaps/uses/use.py", line 45, in __call__
    output = self.main(args, app)
  File "/usr/local/lib/python2.7/dist-packages/openaps/vendors/dexcom.py", line 628, in main
    for item in self.dexcom.iter_records(self.RECORD_TYPE):
AttributeError: 'NoneType' object has no attribute ‘iter_records’
(yes, still working on this cob undefined thing)
PieterGit
@PieterGit
May 31 2017 07:16
@Kdisimone : are you using the dexcom via USB? Most iter_records problems are caused by the fact that the dexcom_reader can't read it's values. Perhaps check with lsusb if the dexcom reader is properly connected
katie disimone
@Kdisimone
May 31 2017 07:17
It's a spare rig reading from a NS that is getting BGs from share bridge
Rig is paired with a spare pump (not on my daughter) and a spare NS site (pulling from same dexcom account as what we actively use for her rig's). Was setting up this rig to try some dev stuff. I just can't seem to get it running. But, its late now and I'll pick this up tomorrow and start with fresh eyes.
PieterGit
@PieterGit
May 31 2017 07:20
can't test things for you on a rig now. afacit the monitor/cgm-glucose.json will read the stuff from the dexcom by usb. have you tried the openaps ns-loop or ns-glucose.json?
see if openaps report invoke/monitor/ns-glucose.json works
there are various copy commands that will copy the glucose values to monitor/glucose.json
what CGM did you select in oref0-setup? G4-upload?
Miiko
@tw_miiko_twitter
May 31 2017 07:33
@PieterGit How to setting rig for Offline Looping by with 722 Pump plus Enlite Sensor?
PieterGit
@PieterGit
May 31 2017 07:40
@tw_miiko_twitter sorry can't help you with that. Only have experience with dex g4 and 754 pump
Martin Haeberli
@mhaeberli
May 31 2017 07:45
lost in space - just got a G5 working for the first time, but having trouble following or getting the data into Nightscout
advice sought
Tim Street
@tim2000s
May 31 2017 08:08
@mhaeberli iPhone or xDrip?
Martin Haeberli
@mhaeberli
May 31 2017 08:08
iphone
it was flowing fine before with the NS bridge from Dexcom share
but now not seeing it in NS nor on share followers iphones
Tim Street
@tim2000s
May 31 2017 08:09
I assume you have the right config in the phone app for share?
Martin Haeberli
@mhaeberli
May 31 2017 08:09
it says I do
like i look at the G5 app on her phone, and it says that I am being shared with
Tim Street
@tim2000s
May 31 2017 08:12
I'd give Dexcom tech support a call. Have you initiated those other shares from the iPhone app?
Martin Haeberli
@mhaeberli
May 31 2017 08:12
thx - yes - do you know if they are 7x24?
Tim Street
@tim2000s
May 31 2017 08:13
Think so.
Martin Haeberli
@mhaeberli
May 31 2017 08:56
they are 7x24
fidoman
@efidoman
May 31 2017 09:43
@JELCRAWFORD - Thanks so much for the advice which is very helpful in determining my strategy for communicating with my endo about openaps.
@cameronrenwick - Thank you. I will report back on Thursday for sure. He has been open on other things and has mentioned the Minimed hybrid closed loop system to me at several appointments. He doesn't like to debate things or look at any records I have. He prefers to just quickly chat about how things are going, so I'm coming up with an "elevator pitch" for openaps, similar to what you described I think. And I will have what new prescriptions I need ready and printed out. He appreciates brevity.
fidoman
@efidoman
May 31 2017 10:20

@danamlewis - I manually made the changes to fix the autotune issue and ran autotune manually and it worked for me. Yay! Parameter | Current | Autotune

ISF | 46.000 | 53.106
CSF | n/a | 3.579
Carb Ratio | 12.000 | 14.838
00:00 | 0.650 | 0.650
01:00 | | 0.650
02:00 | | 0.650
03:00 | | 0.650
04:00 | 0.800 | 0.800
05:00 | | 0.800
06:00 | | 0.800
07:00 | | 0.800
08:00 | | 0.800
09:00 | | 0.779
10:00 | 0.650 | 0.662
11:00 | | 0.662
12:00 | | 0.679
13:00 | | 0.650
14:00 | | 0.644
15:00 | 0.800 | 0.759
16:00 | | 0.759
17:00 | | 0.765
18:00 | 0.650 | 0.650
19:00 | | 0.650
20:00 | | 0.650
21:00 | | 0.650
22:00 | | 0.650
23:00 | | 0.650
?dana
rsilvers129
@rsilvers129
May 31 2017 10:53

@scottleibrand It fixed my autotune issue on dev: Parameter | Current | Autotune

ISF | 32.000 | 32.997
CSF | n/a | 4.272
Carb Ratio | 7.500 | 7.724
00:00 | 0.675 | 0.691
01:00 | 0.725 | 0.737
fidoman
@efidoman
May 31 2017 11:54
I read the docs and must have missed what CSF means. I know it is Carb Sensitivity Factor but what is the number? number of grams of carbs that can be absorbed in 5 minutes?
jtsackdex
@jtsackdex
May 31 2017 12:07
So my rig doesn't seem to see the MM pump.
edison@eddie:~$ openaps use pump model
usage: openaps-use [-h] [--format {text,json,base,stdout}] [--output OUTPUT]
[--version]
device ...
openaps-use: error: argument device: invalid choice: 'pump' (choose from )
Any ides on where I should narrow the search? When entering the SN am I safe to assume it is just the SN minus the letters?
My SN # is FRA556677G so I would only use the 556677 correct? Using explorer board . And it seems that MMeowlink launches properly.
rsilvers129
@rsilvers129
May 31 2017 12:11
I got dev working yesteday. I was wondering why I was hanging out at 170 for hours and it was seemingly doing nothing to bring me down. I took a manual bolus. This morning, I noticed that the preferences was reset, and max_iob was reset to 0. I could have sworn I fixed that last night, but I guess not. Well, at least there is an explaination for why it was not working, so really looking forward to today.
@jtsackdex People say to not post your serial number. I guess the concern is that someone will build a database of them, and write custom code and use a powerful transmitter at a diabetes conference and murder the entire crowd using a far more complex method than poisoning the punch bowl.
jtsackdex
@jtsackdex
May 31 2017 12:20
@rsilvers129 That's not my actual SN I fudged the numbers quite a bit. Just trying to make sure I'm doing this right. Is there a setting on pump to make sure radio is on? Should I be looking at Utilities and Turn on meter or sensor?
Miiko
@tw_miiko_twitter
May 31 2017 12:28
After the auto-intalation and running, why in myopenaps directory no "autotune"? And in Subdirectory Settings only three files "autosens.json profile.json temptargets.json"? In Subdirectory Monitor
5 files "carbhistory.json edison-battery.json glucose.json meal.json mmtune.json" and in Subdirectory Encat no things there? What is missing with installation and running, which I still need with setting?
rsilvers129
@rsilvers129
May 31 2017 12:32
@jtsackdex Yes. 556677. Don’t need meter or sensor. Make sure “temp basal type” is not on percent. Put pump within 20 inches of board.
@tw_miiko_twitter The autotune directory will be created when it first runs at midnight.
dan lightner
@danimaniac
May 31 2017 13:36
@ericfeibelman I would be interested in seeing the content of your elevator speech...and to hear how well it is received.
cameronrenwick
@cameronrenwick
May 31 2017 13:53
@ericfeibelman I found that once you state the "I know you can't endorse this but I'm going to proceed regardless" the docs are pretty good with it. Really and truly they "can't" endorse it, nor do any of us expect it. There are lots of non-compliant patients out there; most do things that lead to poorer health; we're the opposite. The A1Cs speak louder than anything else and it's the data that will always set us free.
katie disimone
@Kdisimone
May 31 2017 14:33
@jtsackdex did you flash your edison yourself?
And why are you logged in as edison instead of root? With explorer board setup, you really only need to login as root.
rsilvers129
@rsilvers129
May 31 2017 14:55
@ericfeibelman I wonder if my Dr will evenb say he can't endorse it. I made my appointmenr 90 datys from today, so my A1c test will be 3 months from now. Tomorrow will be my first 7 days on OpenAPS.
By the way, I have seen people post battery life results. My last AAA Duracell lasted 2 days + 7.25 hours. It died at 1.22 volts, and I feel that 1.25 volts is what I want the warning set to for Alkaline. So replacing them every 48 hours seems good. I Ultimate lithium on order from Amazon, but I saw them at Walmart yesterday in four packs for only 25 cents each more than I paid for a 20 bulk pack from Amazon.
jtsackdex
@jtsackdex
May 31 2017 15:01
@Kdisimone no I bought the one preflashed. also the explorer board I've been trying is version from 2016. I have a 2017 now going to start over.
Scott Leibrand
@scottleibrand
May 31 2017 15:11
@rsilvers129 lol at the punch bowl thing. I think the SN posting risk is actually not much more of a risk than the risk of looping in the first place. If someone wants to do something far more complicated than the punch bowl like that, they can just set up a receiver to listen for rigs talking to pumps: the SN is broadcast in the clear. The protection against bad outcomes (other than "poison or a pistol is easier") is mostly down to your max bolus setting in your pump, and having a CGM running so you'll know if BG starts to tank and you need sugar etc. before it's too late.
@ericfeibelman CSF is mg/dL of eventual BG rise per gram of carbs. It isn't a rate and doesn't really have a time component.
JELCRAWFORD
@JELCRAWFORD
May 31 2017 15:51
@ericfeibelman - we finally switched to the ultimate lithium and we get a week + before change. They are expensive in Canada, so I don't know there is much, if any, cost savings, but nice to not have it on the list every other day. They really do go from looking good to empty/dead though - we have not yet determined a good cutoff for the bat. alarm to go off at - If any of you out there using lithium have thoughts about this, would love to hear. The boy carries an extra alkaline battery or two in his low kit 'just in case'.
Ryan Dalisky
@rdalisky
May 31 2017 15:54
Is the Dev branch the most cutting edge, or is auto-maxIOB? I have been running dev for a week and I guess I'm not sure where to view if the oref1 features have been working or not. Any ideas?
Scott Leibrand
@scottleibrand
May 31 2017 16:06
we change the AAA lithiums before bed if they drop below 1.3v during the day
@rdalisky dev should be fine for getting oref1 features, but you need to follow the docs to enable them, and then confirm in your pump-loop.log if you're running the supermicrobolus pump-loop or not
Dana Lewis
@danamlewis
May 31 2017 16:19
@tw_miiko_twitter did you follow the extra steps for manually creating the x12 settings files as described in the docs?
@rdalisky i would not ever use any randomly named branch that you spot - stick with master or dev unless there's an explicit recommendation to try a branch for a particular feature
@jtsackdex what do your logs say?
Ryan Dalisky
@rdalisky
May 31 2017 16:23
@scottleibrand @danamlewis thanks for the info. I set up all of the advanced features in the preferences.json, so I'll check pump-loop.log to see what's been happening.
Matthias
@ceben80
May 31 2017 16:29
@danamlewis #484 is working for me :wink:
Dana Lewis
@danamlewis
May 31 2017 16:29
Super, thanks for testing it!
Matthias
@ceben80
May 31 2017 16:33
you're welcome
Dana Lewis
@danamlewis
May 31 2017 16:33
Props again to @tynbendad for theorizing and PRing the fix to test :D
Matthias
@ceben80
May 31 2017 16:35
@danamlewis very stable my basal settings :star2:
Recommendations Log File: /root/myopenaps/autotune/autotune_recommendations.log

*Parameter      | Current  | Autotune*

ISF            | 35.000   | 31.056
CSF            | n/a      | 3.416
Carb Ratio     | 9.000    | 9.091
  00:00        | 1.200    | 1.224
  01:00        | 1.100    | 1.138
  02:00        | 1.100    | 1.111
  03:00        | 1.100    | 1.109
  04:00        | 1.300    | 1.299
  05:00        | 1.300    | 1.300
  06:00        | 1.500    | 1.504
  07:00        | 1.500    | 1.500
  08:00        | 1.500    | 1.548
  09:00        | 1.200    | 1.248
  10:00        | 1.200    | 1.250
  11:00        | 1.200    | 1.202
  12:00        | 0.800    | 0.802
  13:00        | 0.800    | 0.800
  14:00        | 0.800    | 0.800
  15:00        | 0.800    | 0.800
  16:00        | 0.800    | 0.807
  17:00        | 0.800    | 0.803
  18:00        | 0.800    | 0.803
  19:00        | 0.800    | 0.796
  20:00        | 0.800    | 0.800
  21:00        | 0.800    | 0.800
  22:00        | 0.800    | 0.800
  23:00        | 0.800    | 0.826
Dana Lewis
@danamlewis
May 31 2017 16:35
:D
Scott Leibrand
@scottleibrand
May 31 2017 16:35
:+1: but also, one day's data isn't expected to change them much.
Matthias
@ceben80
May 31 2017 16:35
sure
cool my new 2500er lipo is running since 12 hours, EdisonVoltage shows 61% now :+1:
cameronrenwick
@cameronrenwick
May 31 2017 16:37
@ericfeibelman here's the best place in Canada to buy lithiums... https://www.amazon.ca/gp/product/B0002DUQDQ/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1 I order them 2 packs at a time. Still waiting for someone to head to the US and do a battery run for me...
jtsackdex
@jtsackdex
May 31 2017 16:39
@danamlewis started over with all new. guessing i messed it up by being logged in as 2 different users. also had a 2016 explorer board but now using 2017 v1.2. seems to be working right now. how often should my webpage update for openaps and pump on average? 5 min or 15min?
Dana Lewis
@danamlewis
May 31 2017 16:40
Just merged the autotune fix: this is relevant for anyone who’s been testing dev/oref1 (or even dev without oref1). Check your autotune files, and if you had the CSF of 0, pull the latest dev with the fix.
@jtsackdex webpage? do you mean Nightscout? it should be updating as frequently as the rig uploads. But you’ll want to read the logs directly (after setup script, type l) to see what the rig is doing in more detail.
fidoman
@efidoman
May 31 2017 16:42

Elevator Speech for Endo ... I am trying to be as direct and to the point as I can be:
5 weeks ago, I switched from MDI to the Enlite CGM and MiniMed 722 pump in order to leverage the Open Artificial Pancreas System (OpenAPS). This small computer system {show rig } constantly communicates securely with my insulin pump to get CGM data and adjust temp basal rates every 5 minutes as needed. The system has significantly reduced my low blood glucose episodes and makes night-time much safer for me.
{Show NS Typical Days before w/ CGM & Pump only}

{Show NS Now w/ CGM, Pump, OpenAPS, SMB – no manual boluses}

New Prescriptions Needed
• Silhouette Paradigm 23 in. MMT-378
• Enlite Sensor MMT-7008A
• MiniLink REAL-Time Transmitter
• Enlite Serter
• MiniLink Replacement Charger
• Novolog vials
• MiniMed 1.8mL Reservoirs
• MiniMed 722 insulin pump

jtsackdex
@jtsackdex
May 31 2017 16:43
yes NS. time pills seem took a while to update at 1st. seem to be picking up frequency now. doing this standalone with no insulin for now. @danamlewis
Scott Leibrand
@scottleibrand
May 31 2017 16:43

if anyone ever feels annoyed that npm is slow and wants to test out a potentially faster version to see if it works with all our stuff:

"Starting today, typing npm install npm@latest -g will update you to npm version 5.0.1.

npm@5 is all new and packed with performance, reliability, and usability improvements we know you’ll love. These include a new approach to lockfiles, more robust caching, and incredible speed — for many common tasks, npm@5 is up to 5x faster than previous versions."

fidoman
@efidoman
May 31 2017 16:43
days before OpenAPS (CGM and Pump only) - two charts below. Also included is a picture of my rig in its case
image.png
image.png
image.png
Now w/ CGM, Pump, OpenAPS, SMB – no manual boluses.
Dana Lewis
@danamlewis
May 31 2017 16:44
@jtsackdex that’s probably related to the fact that, per the docs, it takes a good 15-20 minutes to gather all the history and run for the first time. you can figure a lot of things out by watching the logs and what’s actually happening on the rig itself, to troubleshoot if info not showing up in NS (or being delayed) is due to a) the rig not actually looping or b ) the rig not having internet or c ) NS being misconfigured, etc.
fidoman
@efidoman
May 31 2017 16:44
image.png
rsilvers129
@rsilvers129
May 31 2017 16:45
@ericfeibelman What are your reasons for getting a prescription for a 722 pump?
fidoman
@efidoman
May 31 2017 16:45
tax purposes
jtsackdex
@jtsackdex
May 31 2017 16:47
@danamlewis Watching the log it starts every 2 min. Very impressive.
rsilvers129
@rsilvers129
May 31 2017 16:47
@ceben80 The Lipo battery life seems mostly related to if the rig is using Bluetooth or Wifi. If I am at home, it doesn't last as long as when I am out and it is tethered.
Dana Lewis
@danamlewis
May 31 2017 16:47
@jtsackdex :+1: glad it’s working now
jtsackdex
@jtsackdex
May 31 2017 16:48
@danamlewis Thank You All SO Much! Gonna go this way for a day or more and just watch
Dana Lewis
@danamlewis
May 31 2017 16:49
:+1:
fidoman
@efidoman
May 31 2017 16:52
In the past couple of days, I disabled SMB on my rig and it seems that using just AMA is possibly slightly better for me. The reason I think is because when I have a meal with significant carbs and BG rises quickly, and say for instance the required insulin is 2.5, then I think that SMB will take 25 minutes to get it delivered where AMA can set my basal to 8 units per hour and get me that 2.5 units in 18.75 minutes (by my math) which is going to have slightly better results in reducing post-prandial BG rise.
Miiko
@tw_miiko_twitter
May 31 2017 16:52
@danamlewis thank you Dana, no I habe not following the manually, I did it automatically, also hier http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-2/oref0-setup.html.
Dana Lewis
@danamlewis
May 31 2017 16:53
@ericfeibelman does AMA go that high with temps for you (i.e. not blocked by safety caps)?
Scott Leibrand
@scottleibrand
May 31 2017 16:54
@ericfeibelman AMA will never deliver a basal rate higher than required to get the insulinReq delivered over the course of 30 minutes, because otherwise it wouldn't be safe if you walked away from the rig before the 30 minutes is up
Matthias
@ceben80
May 31 2017 16:54
@rsilvers129 I am running to 100% via bt tethering, no need for me switching between different networks...
danamlewis @danamlewis goes on her weekly begging spree for someone to better document the x12 setup process since more and more people are now using x12s as the number of loopers goes up and the swath of available pumps is more focused on things like x12s
rsilvers129
@rsilvers129
May 31 2017 16:55
@ceben80 I would like to run on bluetooth but iPhone can't do that unless you turn off WiFi on your phone.
Scott Leibrand
@scottleibrand
May 31 2017 16:56
@ericfeibelman so SMB will always deliver a given insulinReq faster than AMA. the one things oref1 does that some people don't like is it low-temps when insulinReq is negative, vs. doing bolus snooze right after a meal bolus. some people (particularly those who count carbs very precisely and bolus exactly the amount they need) find that the zero temping after a meal bolus makes them end up higher than they'd like after the meal, and therefore requiring SMBs later to catch up.
Dana Lewis
@danamlewis
May 31 2017 16:56
I do not have an x12. I cannot test this myself. The person who wrote the notes about the x12 did so before oref0-setup.sh became the main setup last October. So, if you have an x12 sitting around as one of your backup pumps…please consider helping out by testing a setup with x12 and fixing the documents with any gaps for the x12 or better consolidating those notes for x12 in one place.
if no one does this - we may have to say that x12 is unsupported. I would hate to do this. so hoping someone steps up.
cameronrenwick
@cameronrenwick
May 31 2017 16:58
@scottleibrand I tried to run the npm install and go this when I did.

root@cameront1d:~# npm install npm@latest -g
npm ERR! fetch failed https://registry.npmjs.org/npm/-/npm-5.0.0.tgz
npm ERR! fetch failed https://registry.npmjs.org/npm/-/npm-5.0.0.tgz
npm ERR! fetch failed https://registry.npmjs.org/npm/-/npm-5.0.0.tgz
npm ERR! nospc ENOSPC, write
npm ERR! nospc This is most likely not a problem with npm itself
npm ERR! nospc and is related to insufficient space on your system.

npm ERR! System Linux 3.10.98-poky-edison
npm ERR! command "/usr/bin/nodejs" "/usr/bin/npm" "install" "npm@latest" "-g"
npm ERR! cwd /root
npm ERR! node -v v0.10.29
npm ERR! npm -v 1.4.21
npm ERR! code ENOSPC
npm ERR! errno 54
npm ERR!
npm ERR! Additional logging details can be found in:
npm ERR! /root/npm-debug.log
npm ERR! not ok code 0

any thoughts ?
dan lightner
@danimaniac
May 31 2017 17:02
@ericfeibelman Personally, with a 722 pump I would ask for 3.0mL reservoirs. Novolog is good in a pump for 6 days. How much Novolog will you use in 6 days?
rsilvers129
@rsilvers129
May 31 2017 17:05
@danimaniac @ericfeibelman Yes. Get 3 ml. You can fill them up to less than full if you want.
Miiko
@tw_miiko_twitter
May 31 2017 17:05
@rsilvers129 The autotune directory will be created when it first runs at midnight? Why?
What time is it then, when man live in Australia?
Dana Lewis
@danamlewis
May 31 2017 17:07
@tw_miiko_twitter it will be midnight for you based on your time zone.
autotune runs as part of the loop overnight so if it fails, it is less likely to cause issues while you are also trying to have OpenAPS fill in gaps around meals.
you can run it manually if you want now, though, to see what it does
Miiko
@tw_miiko_twitter
May 31 2017 17:09
thank you! @danamlewis @rsilvers129
Scott Leibrand
@scottleibrand
May 31 2017 17:11
@cameronrenwick it says your drive is probably full. is it?
cameronrenwick
@cameronrenwick
May 31 2017 17:13
@scottleibrand yes it does.. I tried running the command wired (vs sshing) but same. I haven't put anything else on my rig aside from oref0 so I don't know if there's anything I can dump - ideas?
Dana Lewis
@danamlewis
May 31 2017 17:19
I’d like another tester on openaps/oref0#486 before merging to dev. (So ideally the tester is already running/comfortable on dev). What needs testing is just ensuring changes to oref0-setup.sh work as expected, per this list: https://github.com/openaps/oref0/pull/486#issuecomment-305255238
fidoman
@efidoman
May 31 2017 17:28
@danamlewis - I think I've seen AMA basal rates go as high as 8 when it needs to because my preferences.json has fairly high caps ... "max_iob": 11,
"max_daily_safety_multiplier": 18,
"current_basal_safety_multiplier": 22,
"autosens_max": 1.3,
"autosens_min": 0.6,
"rewind_resets_autosens": false,
"autosens_adjust_targets": true,
"adv_target_adjustments": true,
"maxCOB": 120,
"override_high_target_with_low": false,
"skip_neutral_temps": false,
"bolussnooze_dia_divisor": 2,
"min_5m_carbimpact": 5,
"carbratio_adjustmentratio": 1,
"autotune_isf_adjustmentFraction": 0.5,
"remainingCarbsCap": 72,
"enableUAM": true,
"enableSMB_with_bolus": false,
"enableSMB_with_COB": false,
"enableSMB_with_temptarget": false
mrshapir
@mrshapir
May 31 2017 17:30
Thank you so much to everyone who has helped me so far. I had openaps looping successfully for two days, but when I switched to a new wifi network I was able to connect via "ifup wlan0" but couldnt "ping google.com". Was able to "ping 8.8.8.8". Saw that previously in gitter that someone had this issue and fixed it by editing their interfaces file. I tried to follow this and now when I run "if up wlan0" or "if down wlan0", I get an error that says
/etc/network/interfaces:5: misplaced option
ifup: couldn't read interfaces file "/etc/network/interfaces"
My interfaces file looks like this now:
fidoman
@efidoman
May 31 2017 17:31
@danimaniac - good point - changing to 3.0mL so I can use a single reservoir for 6 days. I only use 35 or so units per day on average.
mrshapir
@mrshapir
May 31 2017 17:32

auto lo iface lo inet loopback

auto usb0 iface usb0 inet static
address 192.168.2.15
netmask 255.255.255.0

auto wlan0 iface wlan0 inet dhcp wpa-conf
/etc/wpa_supplicant/wpa_supplicant.conf

# For WPA wpa-ssid Emutex wpa-psk passphrase For WEP wireless-essid
#Emutex wireless-mode Managed wireless-key s:password

And the following 4 lines are for when using hostapd... auto wlan0

iface wlan0 inet static

address 192.168.42.1 netmask 255.255.255.0

No idea if I screwed up this file???
fidoman
@efidoman
May 31 2017 17:34
@scottleibrand - Yes, the zero temping instead of bolus snooze is likely the difference I'm seeing. Both features AMA and SMB together or separate are quite impressive. I find AMA to be very effective and quite simple. Also, I take a shower after breakfast. Even though I zero temp and suspend, I was delaying some SMB's in some cases. This along with the zero temping were probably contributing to higher peaks than I wanted.
Matthias
@ceben80
May 31 2017 17:36
That's your max basal segment
@ericfeibelman
Just wondering about your pref.json params.....
fidoman
@efidoman
May 31 2017 17:37
@rsilvers129 - will do. FYI - the prescriptions are mostly for tax purposes only. My insurance has such a high deductible that everything is completely out of pocket so I'm somewhat limited by what's available on ebay. ;->
@ceben80 - Is this what you are looking for? "max_daily_safety_multiplier": 18, "current_basal_safety_multiplier": 22,
Matthias
@ceben80
May 31 2017 17:43
Yup :wink:
Dana Lewis
@danamlewis
May 31 2017 17:58
@ericfeibelman ahhhhhhh. that’s kind of…high.
that’s not a good idea
normal is 3 and 4
fidoman
@efidoman
May 31 2017 18:01
@danamlewis - I haven't found it to cause a problem yet. I suppose I was thinking that because temp basals are only set for 1/2 an hour and are set with the thought that it is the last command it can send to the pump. Are there other scenarios I'm not thinking about?
Dana Lewis
@danamlewis
May 31 2017 18:06
The point is to limit so that any amount of insulin delivered can be counterracted by carbs
It also has to help protect against when your settings are off - either if you come in with way off settings, or if your sensitivity changes, etc.
Think about if you have insulin that starts pooling under a site, and it is maxing out at 18x your current, and you get up and start walking, or take a shower, etc. - WHOOSH. and not just a “ok I’ll go down to 70 and be fine”, but a potentially really bad whoosh that gets you into needing glucagon and/or emergency support :(
Martin Haeberli
@mhaeberli
May 31 2017 18:08
as a meta - question re BGs from, say, Nightscout or any source, sometimes the delivery is delayed - does the database (mongoDB, say, or openaps internal) know the difference between “BG was x at 8:05 AM, reported timely” vs “BG was x at 8:05 AM, reported an hour later”?
Dana Lewis
@danamlewis
May 31 2017 18:09
openaps knows “there’s a BG at 8:05 which is an hour ago is too old BG because it’s 9:05am and we need much more recent data to be able to loop"
so from looping operations perspective, yes. from a more technical perspective - not sure of the specifics about all the various pieces that would pay attention to timestamp vs. just log what the data is
Martin Haeberli
@mhaeberli
May 31 2017 18:10
ok, thx; I’ll also ask the Nightscout team … G5 has a nice “backfill” feature but it can also fool you as to when the point was reported vs gathered at the sensor
fidoman
@efidoman
May 31 2017 18:11
@danamlewis - OK you convinced me. I will lower these settings. Looking at my log file, the max temp basal was 6.075 in the past week. 90% of the time it was less than 2.6 when high temping. Given this, then I think that taking my safety multipliers down by at least half will be safer.
Dana Lewis
@danamlewis
May 31 2017 18:12
:+1: :+1: :+1:
I think lowering your caps closer back to normal is a good idea, plus especially now that we have SMB, relying on that to trigger if you have increased insulin need that is capped by temp basal timing.
I also have one other thing that you might could test for us to further determine a knob related to SMB - will PM you
fidoman
@efidoman
May 31 2017 18:14
10-4
francesc0-cgm
@francesc0-cgm
May 31 2017 18:34
@danamlewis @scottleibrand sorry for bothering. Tried to setup simple httpserver but it doesn't update index.html
Dana Lewis
@danamlewis
May 31 2017 18:34
on android or ?
francesc0-cgm
@francesc0-cgm
May 31 2017 18:34
followed instructions on offline monitoring paragraph
rsilvers129
@rsilvers129
May 31 2017 18:34
@danamlewis I feel like it is hard to know which default settings are default because they are best for most people, vs which are default settings as a hugely conservative setting that no one should want to use. For example, since the max_iob is defaulted to 0, and since that is not actually a reasonable setting, it made me think that the 3 and 4 are also not reasonable settings. So, is the 3 and 4 what is truest best for the majority of people, or is it like the max_iob where it just needs to be raised a lot?
Dana Lewis
@danamlewis
May 31 2017 18:35
3 and 4 is what’s true for most people
@rsilvers129 wanna test openaps/oref0#486 for me? related ;)
rsilvers129
@rsilvers129
May 31 2017 18:35
You are a super fast responder.
Dana Lewis
@danamlewis
May 31 2017 18:35
:D
sometimes
rsilvers129
@rsilvers129
May 31 2017 18:35
It is almost like you had that pre typed.
Dana Lewis
@danamlewis
May 31 2017 18:35
haha. no.
@francesc0-cgm I’d maybe tap @tim2000s or @ceben80 or similar for questions about that, I’m not using Android so not a lot of help beyond what’s in the docs.
francesc0-cgm
@francesc0-cgm
May 31 2017 18:36
sorry @danamlewis
Dana Lewis
@danamlewis
May 31 2017 18:36
no worries :)
rsilvers129
@rsilvers129
May 31 2017 18:38
@danamlewis I don't think I can test 486 because I only have one rig until my second Explorer arrives.
Dana Lewis
@danamlewis
May 31 2017 18:38
k
JELCRAWFORD
@JELCRAWFORD
May 31 2017 18:57
@ericfeibelman - dont forget ketone meter and strips on your prescription list. We also always have on our prescription insulin pen cartridges and pen needles - particularly when we travel we always have what we need to divert to MDI if we had to.... in addition to 3 pumps (two that loop and one that does not). I watch expiry on the pen cartridges and extract insulin to fill pump reservoirs so it doesn't go to waste. We also have a long lasting insulin that goes on the script also in event of MDI need....
jmcrawford
@jmcrawford
May 31 2017 18:59
And Robaxocet for the back pain, carrying all this stuff :D
fidoman
@efidoman
May 31 2017 19:01
@JELCRAWFORD - Thanks for the reminder. I'll need a duffle bag I think for the drug store run. ;->
JELCRAWFORD
@JELCRAWFORD
May 31 2017 19:02
yup - you should see our suitcases when we go :)
It's ok though - I make @jmcrawford carry :)
Dana Lewis
@danamlewis
May 31 2017 19:04
(over here laughing because I now load up Scott’s backpack to lighten mine ;D)
Matthias
@ceben80
May 31 2017 19:07
smart woman :wink:
JELCRAWFORD
@JELCRAWFORD
May 31 2017 19:09
really, how much of his own stuff does @jmcrawford need?!?!?
Scott Leibrand
@scottleibrand
May 31 2017 19:10
;-p
jmcrawford
@jmcrawford
May 31 2017 19:16
(searches frantically for speedo emoji)
rsilvers129
@rsilvers129
May 31 2017 20:20
It is kinda interesting that this system has completely changed what I expect of myself. My average BG was about 160 for the last year, and now I start to get concerned if it is over 120, and don't like meals to peak over 150.
Scott Leibrand
@scottleibrand
May 31 2017 20:25
just wait until you notice your body responding to a 160 the way it used to respond to a 240. :-)
live4sw
@live4sw
May 31 2017 20:27
@rsilvers129 For what it's worth my max_iob's are at 1.5 and my daytime basal is 0.6. And I'm pretty average-sized (5'11/175lb adult).
And max_iob usually does not limit my rig, but I usually don't have more than 30-40g/carbs at a sitting (and many of my meals are <15g)
rsilvers129
@rsilvers129
May 31 2017 20:30

@scottleibrand I am trying the new npm on a friend's future install. I get this with master:
npm WARN npm npm does not support Node.js v0.10.29
npm WARN npm You should probably upgrade to a newer version of node as we
npm WARN npm can't make any promises that npm will work with this version.
npm WARN npm You can find the latest version at https://nodejs.org/

/home/.rootfs/usr/local/lib/node_modules/npm/lib/install.js:298
[this, (next) => { computeMetadata(this.idealTree); next() }],
^
npm ERR! Unexpected token >

npm ERR! A complete log of this run can be found in:
npm ERR! /root/.npm/_logs/2017-05-31T20_27_36_600Z-debug.log

live4sw
@live4sw
May 31 2017 20:31
I tend to be a fairly nervous person so I try to err on the side of conservatism - my A1c's are mid/high 6's, not nearly as good as many/most of the people on here
Scott Leibrand
@scottleibrand
May 31 2017 20:32
@rsilvers129 ok, sounds like someone needs to untangle some dependencies before we try using npm version 5. thx for checking.
RabidMoleMan
@RabidMoleMan
May 31 2017 21:25

Hi everyone, this is my first time posting. I've been working on my rig for a few days now. It consists of an edison, explorer board, and a WW minimed 512. I've run through steps 0, 1, and 2 of the phase 2/oref0 setup a number of times now. I've put in the proper data into step 2 (ie. correct serial number and using the WW setting) but am really struggling to get the pump and edison communicating, I believe. When I run the line "tail -F /var/log/openaps/pump-loop.log" to watch the logs, I am unable to pull the history from the pump and get "Error, retrying" with all of the following displayed. It seems to repeat this over and over for as long as I leave it. This is my first real experience with unix and I've been doing lots of reading to learn it. I'm getting much better at navigating, but I am still struggling with understanding what is going on at times. Any help is greatly appreciated.

root@ubilinux:~# tail -F /var/log/openaps/pump-loop.log

  • sleep 0.5
  • ./change_setting.py /dev/spidev5.1 0x0B 0x38
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • exit 0
    /usr/local/bin/oref0-subg-ww-radio-parameters-timeout: 8: kill: No such process

mmtune: pump://JSON/mmtune/monitor/mmtune.json
reporting monitor/mmtune.json
Listening: .No pump comms detected from other rigs
Old pumphistory:
Error, retrying
Listening: .No pump comms detected from other rigs
subg-ww-radio-parameters:running /usr/local/bin/oref0-subg-ww-radio-parameters for max 30 seconds
Your TI-stick is located at /dev/spidev5.1

  • cd /root/src/subg_rfspy/tools
  • ./reset.py /dev/spidev5.1
    RileyLink OK
    Version: subg_rfspy 0.8
    OK
  • sleep 2
  • ./change_setting.py /dev/spidev5.1 0x06 0x00
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • ./change_setting.py /dev/spidev5.1 0x0C 0x59
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • ./change_setting.py /dev/spidev5.1 0x0D 0x66
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • ./change_setting.py /dev/spidev5.1 0x0E 0x33
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • ./change_setting.py /dev/spidev5.1 0x0F 0x62
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • ./change_setting.py /dev/spidev5.1 0x10 0x1A
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • ./change_setting.py /dev/spidev5.1 0x11 0x13
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • ./change_setting.py /dev/spidev5.1 0x09 0x24
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • ./change_setting.py /dev/spidev5.1 0x0A 0x2E
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • ./change_setting.py /dev/spidev5.1 0x0B 0x38
    RileyLink OK
    Version: subg_rfspy 0.8
    SUCCESS
  • sleep 0.5
  • exit 0
    /usr/local/bin/oref0-subg-ww-radio-parameters-timeout: 8: kill: No such process

mmtune: pump://JSON/mmtune/monitor/mmtune.json
reporting monitor/mmtune.json
Starting pump-loop at Thu Jun 1 02:41:32 +06 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Old pumphistory: ^C

Dana Lewis
@danamlewis
May 31 2017 21:25
hi @RabidMoleMan - welcome!
first off - did you do the manual settings file for the x12 as described in the link here? http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-0/hardware/pump.html#why-do-i-need-a-certain-pump-firmware
RabidMoleMan
@RabidMoleMan
May 31 2017 21:30
@danamlewis I will attempt to do both of those now
Tim Street
@tim2000s
May 31 2017 21:32
@francesc0-cgm - have created a script that runs from a cron entry to pull the data every five mins?
francesc0-cgm
@francesc0-cgm
May 31 2017 21:33
@tim2000s i followed that instructions. I can browse the html but it doesn't update itself
Tim Street
@tim2000s
May 31 2017 21:34
What does your cron entry say?
It should be something like */5 * * * * (bash /root/http.sh) 2>&1 | tee -a /var/log/openaps/http.log
Jane83
@Jane83
May 31 2017 21:37
Hello. I am just thinking about getting started with my own APS. Would love to hear thoughts on what you love and what your frustrations are. My A1c already sits 6.3-6.8% and I'm wondering if this will provide much better control or even reduce hypoglycemia.
Tim Street
@tim2000s
May 31 2017 21:40
@Jane83 It can do a lot of things, depending on how you set it, but the biggest difference is that you won't have to think about it as much to get there once it's set up.
francesc0-cgm
@francesc0-cgm
May 31 2017 21:40
@tim2000s i have /5 * (bash /root/http.sh) 2>&1 | tee -a /var/log/openaps/http.log
Dana Lewis
@danamlewis
May 31 2017 21:40
@Jane83 hi, Jane. most people find it reduces their hypoglycemia, further improves mealtime outcomes, but also reduces the amount of work you need to do overall, too.
francesc0-cgm
@francesc0-cgm
May 31 2017 21:41
Of course with 5 asterisks
Tim Street
@tim2000s
May 31 2017 21:41
YOu're missing the * before the /5
francesc0-cgm
@francesc0-cgm
May 31 2017 21:41
I have it
It is the same as written in the guide
Dana Lewis
@danamlewis
May 31 2017 21:42
@RabidMoleMan and @tw_miiko_twitter and everyone else I’ve talked to about x12 pumps (512 712) lately: I put those notes all in one section here. http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-2/troubleshoot-oref0-setup.html#users-712-users-x12-users
Tim Street
@tim2000s
May 31 2017 21:42
What happens when you run bash http.sh from the command line?
francesc0-cgm
@francesc0-cgm
May 31 2017 21:42
Something wrong with script creation in my opinion
Dana Lewis
@danamlewis
May 31 2017 21:42
But, please, please, please do as the note says and if there’s an extra step you need to do - PR it in or let me know so I can. Or if that’s all you have to do, let me know that, too.
francesc0-cgm
@francesc0-cgm
May 31 2017 21:42
Now i'm on loop again i will try tomorrow
Had some problem with uam enabled
Tim Street
@tim2000s
May 31 2017 21:43
(@frances-cgm - the script is a direct copy of what works on mine, so I'm comfortable it works)
francesc0-cgm
@francesc0-cgm
May 31 2017 21:43
I have to paste it on the terminal or i have to add lines in enact file?
Tim Street
@tim2000s
May 31 2017 21:43
Do nothing in enact file.
francesc0-cgm
@francesc0-cgm
May 31 2017 21:43
I make rm command
Then touch
Tim Street
@tim2000s
May 31 2017 21:44
You make the http.sh file in your home directory.
francesc0-cgm
@francesc0-cgm
May 31 2017 21:44
And i expexted to see a new file to write in
Tim Street
@tim2000s
May 31 2017 21:44
(/root)
francesc0-cgm
@francesc0-cgm
May 31 2017 21:44
I made something wrong...
Tim Street
@tim2000s
May 31 2017 21:44
then the script removes any index.html file and then creates a new one.
francesc0-cgm
@francesc0-cgm
May 31 2017 21:44
Let me see
So i have to create a script with those data in?
Tim Street
@tim2000s
May 31 2017 21:46
Which data?
francesc0-cgm
@francesc0-cgm
May 31 2017 21:46
Those lines sorry
Tim Street
@tim2000s
May 31 2017 21:47
Can you post what your script looks like please?
francesc0-cgm
@francesc0-cgm
May 31 2017 21:47
I have any script i pasted those lines in terminal
I have to create a script and paste those lines in...this is my problem
Tim Street
@tim2000s
May 31 2017 21:48
Okay - if you don't have the http.sh script then it won't work, no.
Jane83
@Jane83
May 31 2017 21:49
Do people use this all day or just at night?
francesc0-cgm
@francesc0-cgm
May 31 2017 21:49
May i ask you how to create it...so sorry i'm quite stupid
So after touch command i have to use nano or vi and write others lines in?
Is it correct?
Dana Lewis
@danamlewis
May 31 2017 21:51
@Jane83 some people start by only using it at night but most people use 24/7
Tim Street
@tim2000s
May 31 2017 21:51
Yes - the simplest way to make a script is to do vim or nano <scriptname> replacing <scriptname> with what it is called, in this case, http.sh.
francesc0-cgm
@francesc0-cgm
May 31 2017 21:52
Ok...then i have to copy from cat command to the end
?
I'm following your guide here
Tim Street
@tim2000s
May 31 2017 21:53
Then you copy everything off the page
So:
rm ~/myopenaps/enact/index.html
touch ~/myopenaps/enact/index.html

(cat ~/myopenaps/enact/smb-enacted.json | jq -r .timestamp | awk '{print substr($0,12,5)}') >> ~/myopenaps/enact/index.html

(cat ~/myopenaps/enact/smb-enacted.json | jq -r .reason) >> ~/myopenaps/enact/index.html
(echo -n 'TBR: ' && cat ~/myopenaps/enact/smb-enacted.json | jq .rate) >> ~/myopenaps/enact/index.html                                  
(echo -n 'IOB: ' && cat ~/myopenaps/enact/smb-enacted.json | jq .IOB) >> ~/myopenaps/enact/index.html
(echo -n 'Edison Battery: ' && cat ~/myopenaps/monitor/edison-battery.json | jq -r .battery | tr '\n' ' ' && echo '%') >> ~/myopenaps/enact/index.html
(echo -n 'Insulin Remaining: ' && cat ~/myopenaps/monitor/reservoir.json) >> ~/myopenaps/enact/index.html
That all goes into the script
francesc0-cgm
@francesc0-cgm
May 31 2017 21:54
So i have to go in my root then type nano http.sh and paste it in
Scott Leibrand
@scottleibrand
May 31 2017 21:54
@tim2000s have you considered adding that script to oref0 to skip all this?
or putting it in your own repo so someone can check it out (or even wget it)?
francesc0-cgm
@francesc0-cgm
May 31 2017 21:55
Ok i got it
Thank you @tim2000s
Other bothering sorry
Is it possible to add an echo command on glucose.json to get also the last bg?
Tim Street
@tim2000s
May 31 2017 21:57
@scottleibrand Just haven't had time to sit down and fit it into the oref0 setup model.
@francesc0-cgm yes, you can.
Scott Leibrand
@scottleibrand
May 31 2017 21:57
@tim2000s k
Tim Street
@tim2000s
May 31 2017 21:57
You can pul it from the same file - it's in smb-enacted or enacted.
francesc0-cgm
@francesc0-cgm
May 31 2017 21:59
So jq .glucose?
I have to check that json to see which is the name for new bg
Maybe jq .bg
Tim Street
@tim2000s
May 31 2017 22:00
I think that's right
francesc0-cgm
@francesc0-cgm
May 31 2017 22:00
Will try thank you very very much
Tim Street
@tim2000s
May 31 2017 22:00
jq -r .bg
francesc0-cgm
@francesc0-cgm
May 31 2017 22:00
Thank you
Maybe @scottleibrand in the guide could you add the vi http.sh step?
Could be clearer for n00b like me
And another question for other topic...my son often eats pasta that takes about 6 hours of absorption time
My dia is 4
Openaps has many difficult to manage it
Cob goes to 0 too early
How can i manage this situation @scottleibrand ?
Dana Lewis
@danamlewis
May 31 2017 22:05
@francesc0-cgm how confident are you in your carb counts?
francesc0-cgm
@francesc0-cgm
May 31 2017 22:05
About 99%
:smile:
I always use a scale before cooking
Scott Leibrand
@scottleibrand
May 31 2017 22:07
are you using master or dev of oref0?
francesc0-cgm
@francesc0-cgm
May 31 2017 22:07
Dev
Scott Leibrand
@scottleibrand
May 31 2017 22:07
have you enabled UAM yet?
francesc0-cgm
@francesc0-cgm
May 31 2017 22:07
Tried today...
Worse than other times
Scott Leibrand
@scottleibrand
May 31 2017 22:08
worse how? how did it dose differently?
francesc0-cgm
@francesc0-cgm
May 31 2017 22:08
Cob went to 0 before using it w/o uam
Scott Leibrand
@scottleibrand
May 31 2017 22:08
I would expect that UAM would take over once COB goes to zero due to the DIA-hour carb cutoff
francesc0-cgm
@francesc0-cgm
May 31 2017 22:09
In reality it went to 0 after 2 hours
Should it go to 0 as long as dia is?
Can't use smb...do you remember :smile: ?
Scott Leibrand
@scottleibrand
May 31 2017 22:09
depends on what BG is doing. if it sees enough positive deviations to "use up" the carbs, then COB will decay faster
if it doesn't decay by then, the rig will "forget" about carbs after DIA hours
francesc0-cgm
@francesc0-cgm
May 31 2017 22:10
For me it is a big limitation...
Scott Leibrand
@scottleibrand
May 31 2017 22:10
even without SMB, you can try using UAM with temp basals. that hasn't been very well tested, though
francesc0-cgm
@francesc0-cgm
May 31 2017 22:10
I will
So uam could avoid after-DIA peaks?
Scott Leibrand
@scottleibrand
May 31 2017 22:11
possibly
or at least get it to high temp a bit more aggressively to reduce their size
francesc0-cgm
@francesc0-cgm
May 31 2017 22:11
Ok i will try and i will give you a feedback about that
Scott Leibrand
@scottleibrand
May 31 2017 22:12
UAM changes the behavior earlier in the meal as well, though. I'm not sure exactly how that will work without SMB: haven't tested that myself.
francesc0-cgm
@francesc0-cgm
May 31 2017 22:12
I noticed it is less flat
Using openaps i have often flat bgs
Tonight i had a peak than a drop then a peak
Scott Leibrand
@scottleibrand
May 31 2017 22:13
UAM was really designed with SMB in mind, so it may not behave quite right, and you may be better off with just AMA
francesc0-cgm
@francesc0-cgm
May 31 2017 22:13
I'm using ama as well
But i have that problem after dia
I have almost the 20% of carbs still there and 0 cob
Scott Leibrand
@scottleibrand
May 31 2017 22:13
with dev we can now set min_5m_carbimpact a bit higher, so it might be safer to give AMA a longer carb memory...
we added that cutoff because of the "reappearing COB" problem
francesc0-cgm
@francesc0-cgm
May 31 2017 22:14
So i have to change that in preferences?
Scott Leibrand
@scottleibrand
May 31 2017 22:14
min_5m_carbimpact is in preference, but the DIA-hour cutoff is in the code
Dana Lewis
@danamlewis
May 31 2017 22:15
@elodaille01 - since you have the CSF =0 issue, like I mentioned in channel earlier, pull a fresh copy of dev and re-run and it should resolve.
francesc0-cgm
@francesc0-cgm
May 31 2017 22:15
Ok will try to tune that parameter
elodaille01
@elodaille01
May 31 2017 22:16
ok @danamlewis
Tim Street
@tim2000s
May 31 2017 22:16
I run DIA-hour cutoff longer as it handles protein better that way. And my insulin duration at 3 hours is too short.
Scott Leibrand
@scottleibrand
May 31 2017 22:16
a higher min_5m_carbimpact will cause it to decay carbs faster if you didn't have enough observed carb absorption after the meal for whatever reason
francesc0-cgm
@francesc0-cgm
May 31 2017 22:16
Last question if my son eats very quickly a meal and prebolusing is not enough could i set to 0 the bolussnooze
Scott Leibrand
@scottleibrand
May 31 2017 22:17
bolus snooze just prevents it from low-temping after you bolus, so I don't think that'd be relevant there
unless you're trying to do a super bolus where you do a bigger than normal bolus and want it to zero temp immediately?
francesc0-cgm
@francesc0-cgm
May 31 2017 22:17
But it could be useful when my son starts to play
Running after meal
And dropping a lot
?
Scott Leibrand
@scottleibrand
May 31 2017 22:18
I would do your bolus and immediately set an activity mode high temp target. If you don't want bolus snooze to prevent zero-temping, setting the bolussnooze_dia_divisor higher will get it to turn off sooner
francesc0-cgm
@francesc0-cgm
May 31 2017 22:18
Ok i will
This is my problem
So to cover pasta after hours i have to change the oref0 code
And do as @tim2000s did
Tim Street
@tim2000s
May 31 2017 22:19
Or try enabling UAM as @scottleibrand suggested.
francesc0-cgm
@francesc0-cgm
May 31 2017 22:20
With uam without smb i have some problems early after meal
It becomes a bit fuzzy
However i will make some tests
And i will let you know
Thank you for your kindness
Scott Leibrand
@scottleibrand
May 31 2017 22:21
remind me what your basal ranges are?
francesc0-cgm
@francesc0-cgm
May 31 2017 22:22
0.125 to 0.175
From midnight to 8 pm
Then he has a peak later during the evening and i have to use 0.50
Scott Leibrand
@scottleibrand
May 31 2017 22:22
ok. the SMB cutoff is 0.3U/hr.
francesc0-cgm
@francesc0-cgm
May 31 2017 22:22
I saw it
During evening it could work
Scott Leibrand
@scottleibrand
May 31 2017 22:23
do you have an x23 pump that can do 0.05U boluses, or just 0.1U?
francesc0-cgm
@francesc0-cgm
May 31 2017 22:23
723 one
Diadon81
@Diadon81
May 31 2017 22:24
''' Test '''
Scott Leibrand
@scottleibrand
May 31 2017 22:24
if you wanted to make some more serious code changes, you could try setting the minimum SMB size to 0.05U and the SMB cutoff to 0.15U/hr or so...
would require lots of dev and testing though
francesc0-cgm
@francesc0-cgm
May 31 2017 22:24
I will...
I have only to learn how to
:smile:
I have to edit something on the rig?
Or on github?
Scott Leibrand
@scottleibrand
May 31 2017 22:25
do you have a disconnected 723 pump you can test on?
francesc0-cgm
@francesc0-cgm
May 31 2017 22:25
Yes i have
I have a 723 and a 722 as backup
Scott Leibrand
@scottleibrand
May 31 2017 22:27
I wouldn't go anywhere near the insulin-dosing pump until your code is really thoroughly tested (and probably by other people as well), but once you get familiar with making tweaks like the DIA-hour carb cutoff like @tim2000s suggested, you might be able to do some "bench testing" and development if you're up for it
Diadon81
@Diadon81
May 31 2017 22:27
Dears why I see only part of the data in autotune recommendation log.
Parameter      | Current  | Autotune
-------------------------------------
ISF            | 27.000   | 27.000
CSF            | n/a      | 0.000
Carb Ratio     | 8.000    | 8.000
Basal Profile  |          |
  00:00        | 1.500    | 1.500
  00:30        |          |
  01:00        |          |
  01:30        |          |
  02:00        |          |
  02:30        |          |
  03:00        |          |
  03:30        |          |
  04:00        |          |
  04:30        |          |
  05:00        |          |
  05:30        |          |
  06:00        |          |
  06:30        |          |
  07:00        |          |
  07:30        |          |
  08:00        |          |
  08:30        |          |
  09:00        |          |
  09:30        | 1.500    | 1.500
  10:00        |          |
  10:30        |          |
  11:00        |          |
  11:30        |          |
  12:00        |          |
  12:30        |          |
  13:00        |          |
  13:30        |          |
  14:00        |          |
  14:30        |          |
  15:00        |          |
  15:30        |          |
  16:00        |          |
  16:30        |          |
  17:00        |          |
  17:30        |          |
  18:00        |          |
Scott Leibrand
@scottleibrand
May 31 2017 22:29
a fix for that CSF=0 bug was just pushed to dev. please update your dev and re-run oref0-setup
Diadon81
@Diadon81
May 31 2017 22:29
I am using this function approx. 3 days
Scott Leibrand
@scottleibrand
May 31 2017 22:29
(I'm assuming you're already on dev, as that's the only place we've seen this bug)
Diadon81
@Diadon81
May 31 2017 22:29
ok
thank you
garykidd
@garykidd
May 31 2017 22:33
@danamlewis planning on setting up new rigs this weekend.
francesc0-cgm
@francesc0-cgm
May 31 2017 22:33
Ok thanks @scottleibrand
francesc0-cgm
@francesc0-cgm
May 31 2017 22:40
Is this the code?
openaps/oref0@137304b
garykidd
@garykidd
May 31 2017 22:41
looks like there's a a bit of a run on the smaller footprint 2000mah lipo's....anyone know if they are in stock anywhere?
Scott Leibrand
@scottleibrand
May 31 2017 22:44
@francesc0-cgm in dev it's back down to just DIA, not 1.5x: https://github.com/openaps/oref0/blob/dev/lib/meal/total.js#L40
francesc0-cgm
@francesc0-cgm
May 31 2017 22:45
Ok...i just saw it...if i edit this in that way it could be better for me
Scott Leibrand
@scottleibrand
May 31 2017 22:46
just be on the lookout for reappearing carbs, and if you see them, see if raising min_5m_carbimpact helps eliminate them
francesc0-cgm
@francesc0-cgm
May 31 2017 22:50
Ok...found also the smb settings at line 760 in basal-set-temp.js
Diadon81
@Diadon81
May 31 2017 22:51
Dears, I am trying to update my dev branch on a rig by issue command cd ~/src/oref0 && git checkout dev && git pull but there are no any changes Already on 'dev' Your branch is up-to-date with 'origin/dev'. Already up-to-date.
But after manually checked my branch content, i didn't last commit (#484) on file. Am I doing something wrong?
francesc0-cgm
@francesc0-cgm
May 31 2017 22:51
Sorry in determine-basal
tynbendad
@tynbendad
May 31 2017 22:51
@thebookins does lookout app work for you/are you able to BT tether when there is no cell phone signal? @tim2000s wasn't able to...
Scott Leibrand
@scottleibrand
May 31 2017 22:52
@Diadon81 what does git show give you?

it should be:

commit dcb08e70020dcc0a779b45b81a866176aa929fc6
Author: tynbendad <tynbendad@gmail.com>
Date:   Wed May 31 10:36:53 2017 -0600

...

+    else
+        # Copy tuned profile produced by autotune to profile.json for use with next day of data
+        cp newprofile.$run_number.$i.json profile.json
Tim Street
@tim2000s
May 31 2017 22:52
@tynbendad That's not a function of lookout. It's how the iPhone works. And all that lookout does is take advantage of tethering to connect.
Diadon81
@Diadon81
May 31 2017 22:52
commit 9a1e8f347d16a828a46e88377422bf393789f6d3 Author: John Benjamin <johnbenjamin1061@gmail.com> Date: Tue May 23 22:41:40 2017 -0400
Scott Leibrand
@scottleibrand
May 31 2017 22:53
how about git remote -v ?
Diadon81
@Diadon81
May 31 2017 22:53
origin git://github.com/Bender1061/oref0.git (fetch)
origin git://github.com/Bender1061/oref0.git (push)
Scott Leibrand
@scottleibrand
May 31 2017 22:54
there's your problem. You're using @Bender1061's repo, not the oref0 one
simplest fix is to rm -rf ~/src/oref0 and clone it again from openaps/oref0
Diadon81
@Diadon81
May 31 2017 22:55
but I need #478 fix, is it merge to original branch?
John Benjamin
@Bender1061
May 31 2017 22:57
What I normally do is just delete oref0 and do the whole command, it makes it much easier than to run the command from the docs you are running.
$ cd ~/src/ && rm -r oref0
cause git is "Fun"
Scott Leibrand
@scottleibrand
May 31 2017 22:58
@Diadon81 in that case you could do git pull git@github.com:openaps/oref0.git dev
that would merge the oref0 dev branch into your branch with the 478 fix
Diadon81
@Diadon81
May 31 2017 22:59
ok
Scott Leibrand
@scottleibrand
May 31 2017 23:00
you might need to replace the git@ URL with https://github.com/openaps/oref0.git
John Benjamin
@Bender1061
May 31 2017 23:00
hmm. thats a good way, I just had made dev1 myself to test that fix
Scott Leibrand
@scottleibrand
May 31 2017 23:00
there's always 5 more ways to do everything in git. :)
John Benjamin
@Bender1061
May 31 2017 23:00
yes there is
Diadon81
@Diadon81
May 31 2017 23:02
Ok, it works
``` From https://github.com/openaps/oref0
  • branch dev -> FETCH_HEAD
    Merge made by the 'recursive' strategy.
    bin/oref0-autotune.sh | 3 +++
    lib/basal-set-temp.js | 2 +-
    lib/determine-basal/determine-basal.js | 14 ++++++++++----
    3 files changed, 14 insertions(+), 5 deletions(-)
    ```
tynbendad
@tynbendad
May 31 2017 23:05
@tim2000s right, but not much use in lookout unless tether works... Although u guess you could use it to avoid sending BG to share, or to reduce data cost a little, but the real point of it I thought was for offline monitoring, ie when theres no cell service.
Scott Leibrand
@scottleibrand
May 31 2017 23:07
FWIW, if you establish a tethering connection before losing cell signal, it will usually stay up until something changes. I've managed to stay connected through entire long flights that way.
garykidd
@garykidd
May 31 2017 23:08
anyone use these by chance?
Tim Street
@tim2000s
May 31 2017 23:08
Hmmm. Testing it on the underground this evening and as soon as I went in the tunnel, the tether dropped.
Scott Leibrand
@scottleibrand
May 31 2017 23:08
another option is to use an external wifi hotspot like an Edimax or Hootoo travel router to establish the network connection between the phone and rig, and then do the Lookout thing via that link
@tim2000s did the phone drop it, or oref0-online?
Diadon81
@Diadon81
May 31 2017 23:09
@scottleibrand I prefer to use Enlite sensor. It helps me to loop offline all the time.
Tim Street
@tim2000s
May 31 2017 23:09
Good point. It may we'll have been oref0-online.
Scott Leibrand
@scottleibrand
May 31 2017 23:09
yeah, that's an option, as is plugging in a Dexcom receiver with USB OTG and an external battery
Tim Street
@tim2000s
May 31 2017 23:11
Doesn't explain why I haven't seen it drop on an android device though?
jtsackdex
@jtsackdex
May 31 2017 23:12
So I'm obvisouly been staring at my computer to long. Got dyslexic with my SSID. Back working again.
John Benjamin
@Bender1061
May 31 2017 23:14
@Diadon81 :+1: on the Enlite is awesome for that!
Diadon81
@Diadon81
May 31 2017 23:15
Why most of the people here are using dexcom?
Dana Lewis
@danamlewis
May 31 2017 23:16
Because the improved Medtronic sensors aren't available everywhere
Diadon81
@Diadon81
May 31 2017 23:16
I have tried all kind of CGM.. . Enlite, Libre and Dexcom and finally I decide to use Enlite as most confortable
Dana Lewis
@danamlewis
May 31 2017 23:17
@tomasboudr scroll back up; did you accidentally hit enter here instead of providing a value?
Diadon81
@Diadon81
May 31 2017 23:17
So, and you can't just come and buy it?
The price in Russia is 32USD per sensor. One sensor I can use approx. 2-3 weeks
Dana Lewis
@danamlewis
May 31 2017 23:19
Those sensors aren't approved in the US so they're hard to get
So for many people Dexcom is available and superior to what's available in the US
Diadon81
@Diadon81
May 31 2017 23:20
That sad.
I like many things in USA, but not Medical services.
Tim Street
@tim2000s
May 31 2017 23:24
As far as I am aware the Enlite II sensors are the same everywhere, the big difference is in the transmitters algorithms. The only non-worldwide sensor is the new Enlite for the 670.
Diadon81
@Diadon81
May 31 2017 23:29
I have NA version of the pump and also I have NA transmitter, so I don't think that enlite isn't approved in USA. I think it closely depends on insurance plan.
Scott Leibrand
@scottleibrand
May 31 2017 23:30
lots of people prefer Dexcom over Enlite because the Dexcom is more forgiving about "calibration hygiene" and the sensors last longer.
Dana Lewis
@danamlewis
May 31 2017 23:31
There are different kind of enlites that are approved and available in different places. The older worse mard enlite is what is in the US
Diadon81
@Diadon81
May 31 2017 23:31
@scottleibrand in that case, I choice is Libre :)
It doesn't need any calibration
Dana Lewis
@danamlewis
May 31 2017 23:32
Libre not approved in US yet either
garykidd
@garykidd
May 31 2017 23:32
hmmm where can I read about oref1?
Diadon81
@Diadon81
May 31 2017 23:32
@danamlewis same with russia
Dana Lewis
@danamlewis
May 31 2017 23:33
This message was deleted
Diadon81
@Diadon81
May 31 2017 23:33
but i order it from UK, 10 sensors
Dana Lewis
@danamlewis
May 31 2017 23:33
wrong paste :D
Diadon81
@Diadon81
May 31 2017 23:33
10 sensors for 5 month.
garykidd
@garykidd
May 31 2017 23:35
thx
@Diadon81 not sure if folks care---I actually would not---but I don't think that Enlite is approved for making dosing decisions currently (is it?)
Scott Leibrand
@scottleibrand
May 31 2017 23:39
neither is G4, but everyone does it anyway. :)
garykidd
@garykidd
May 31 2017 23:49
and understandably so....have been doing so for years even with far less precise technology....if you're at 260 on any CGM....pretty sure you're gonna need some sort of correction....
@scottleibrand regarding oref1 and SMBs....understanding that SMB's are limited to 1/3 of then estimated dose.....at the next 5 min interval...assuming the prior estimated dose was exactly correct....would/should oref1 issue another 1/3....and so on and so on...
Scott Leibrand
@scottleibrand
May 31 2017 23:57
yes, although in practice any changes in BG tend to mean that the nice decay curve you would expect never really shows up
either it keeps dosing about the same amount every 5m (or even increases), or it drops way down (and often stops entirely)