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

25th
Jul 2017
aliceinnowonderlands
@aliceinnowonderlands
Jul 25 2017 00:04
ok great, reinstalling it now =)
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:00
I am having a re-occuring issue of bolusing for carbs, and then having an openaps set what I consider unreasonable low basals (like, 0 for a long time after I bolused for ice cream)
This sounds like the bug that scott had posted about, but I have re downloaded and reinstalled master on all of my rigs
bolus snooze is still on 2, I haven't changed that
Screen Shot 2017-07-24 at 9.01.53 PM.png
Also, I keep getting the purple predictor lines of death, and the pill claiming all my predicted BGs' will be in the 30's and lower
Scott Leibrand
@scottleibrand
Jul 25 2017 01:21
@PopcornPam_twitter that usually seems to be a problem with ratios. if you do the BG - (ISF * IOB) + (CSF * COB) calculations manually, what do you get for your eventualBG at the time you're getting the "predictor lines of death"?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:23
lemme go back in time to when I had the death predictors. The other thing is that it will go from all predictions 30 or lower, to predicting about 150 in a pretty short space of time, which seems a bit wierd, since usually the rise rate is consistant
Thacher
@thacherT1D
Jul 25 2017 01:24
so I updated my API secret and now everything is working except for being able to view my openaps data on my nightscout heroku site... any ideas on where to look for the problem?
Dana Lewis
@danamlewis
Jul 25 2017 01:24
/me dislikes jokes about predictor lines
@thacherT1D did you update it in OpenAPS?
Easiest is to change in your runagain and run it again
Thacher
@thacherT1D
Jul 25 2017 01:26
yeah, I did that and then changed it my heroku config vars
when I look at the logs (locally and on papertrail) they are up and working, it's just viewing it on the nightscout site
Scott Leibrand
@scottleibrand
Jul 25 2017 01:27
@PopcornPam_twitter if you have next to no carb absorption yet, then it'll average the carbs in over 4 hours, which usually results in a line that goes up a bit, way down, and then back up
as long as the prediction line eventually gets back above target, it is allowed to bolus snooze and set a neutral temp
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:29
when I do the math myself I get a bg of 260, but the log seems to indicate a bg of 39
{"insulinReq":-1.95,"bg":161,"reservoir":"69.5","temp":"absolute","snoozeBG":156,"rate":0,"predBGs":{"COB":[161,154,146,139,132,125,117,110,104,97,91,85,79,73,67,61,56,50,46,41,39],"aCOB":[161,161,160,157,152,145,137,130,122,115,108,101,95,89,82,76,70,64,58,53,49,44,40,39],"UAM":[161,155,147,139,131,123,116,108,101,93,87,80,73,67,61,54,48,42,39],"IOB":[161,153,145,137,130,122,114,106,99,92,85,78,72,65,59,53,46,41,39]},"minPredBG":-4,"IOB":3.077,"reason":"COB: 13, Dev: 22, BGI: -6.59, ISF: 52, Target: 97, minPredBG -4, IOBpredBG 39, COBpredBG 39, UAMpredBG 39; Eventual BG 39 < 90 insulinReq -1.95; setting 120m zero temp. ","COB":13,"eventualBG":39,"duration":120,"tick":-8,"deliverAt":"2017-07-25T00:41:36.794Z"}
Sorry @danamlewis ... uncouth remark on my part
@scottleibrand , it seems to take a really long time for the predictor line to get back up to what is probably more accurate
Scott Leibrand
@scottleibrand
Jul 25 2017 01:32
ok, let's check your math. what are your ISF and CSF from autotune?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:34
That might be part of it.. Autotune always wants it to be 20+ (like today it thinks 23 g is the correct one), but my pump is set at 15
Autotune has me set at 51
Scott Leibrand
@scottleibrand
Jul 25 2017 01:35
for which?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:35
For CSF autotune says 23 g, but I say 15 g
Scott Leibrand
@scottleibrand
Jul 25 2017 01:35
that's not CSF. you mean CR?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:35
Autotune says ISF is 50
Oh... shoot.. you are right
Dana Lewis
@danamlewis
Jul 25 2017 01:35
@thacherT1D check your mlab size next? If either are around 496, you'll want to compact.
aliceinnowonderlands
@aliceinnowonderlands
Jul 25 2017 01:36
oh
thanks @Kdisimone and @danamlewis
i reinstall everything and now its looping again
don't feel like broken robot anymore
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:36
I am trying to find the CSF in autotune and not seeing it
Scott Leibrand
@scottleibrand
Jul 25 2017 01:36
It's right between the two
CSF is just ISF/CR, so CR would be 2.17 for 50/23, or 3.333 for 50/15.
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:37
and that is something set in preferences, right?
Scott Leibrand
@scottleibrand
Jul 25 2017 01:37
no
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:37
maybe I am in the wrong file, I am not seeing it
Scott Leibrand
@scottleibrand
Jul 25 2017 01:37
autotune just calculates it from your ISF and CR in the pump, and then tunes it accoringly
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:37
oh! There it is.. 4.7
Scott Leibrand
@scottleibrand
Jul 25 2017 01:38
huh what? can you paste those three lines please?
Dana Lewis
@danamlewis
Jul 25 2017 01:38
@aliceinnowonderlands - yay! Glad you're looping again. Btw - that was a git related error (not common) and we have a non-git approach coming in the next release that will cut down on those. But you also know what to do now in future if you see the same error.
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:38
","sensitivity":60,"x":2,"offset":1260}],"first":1},"carb_ratio":23.616,"carb_ratios":{"units":"grams","raw":"0x01 0x00 0x0d 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00","first":1,"schedule":[{"start":"00:00:00","r":13,"ratio":13,"offset":0,"i":0,"x":0}]},"csf":4.698}
Scott Leibrand
@scottleibrand
Jul 25 2017 01:38
oh, no, not that
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:38
I think I'm in the wrong file
Scott Leibrand
@scottleibrand
Jul 25 2017 01:38
cat ~/myopenaps/autotune/autotune_recommendations.log | egrep "\."
Thacher
@thacherT1D
Jul 25 2017 01:38
@danamlewis I don't understand what you mean -- looking at my mlab size from heroku?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:39
ISF | 80.000 | 111.412
CSF | n/a | 4.651
Carb Ratio | 15.000 | 23.954
Scott Leibrand
@scottleibrand
Jul 25 2017 01:39
hmm... thought you said your ISF was 50?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:39
Worth noting, right now my logs are giving me an ISF of 50
Scott Leibrand
@scottleibrand
Jul 25 2017 01:40
cat settings/autosens.json ?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:40
"reason":"COB: 13, Dev: 22, BGI: -6.59, ISF: 52,
{"ratio":1.17}
that doesn't make sense
Scott Leibrand
@scottleibrand
Jul 25 2017 01:41
high ratio = more resistant = lower ISF
but those numbers don't quite add up still
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:41
Ya, I get that, I don't understand the ratio of 1.17
Dana Lewis
@danamlewis
Jul 25 2017 01:41
@popcornpam_twitter is Autotune enable in your loop?
Scott Leibrand
@scottleibrand
Jul 25 2017 01:41
what do the Autosens adjusting line and the next line in pump-loop.log show?
Thacher
@thacherT1D
Jul 25 2017 01:43
@danamlewis they're both well under 500 MB
Scott Leibrand
@scottleibrand
Jul 25 2017 01:43
@PopcornPam_twitter and grep autosens_ preferences.json for me please?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:43
sure
"autosens_max": 1.2,
"autosens_min": 0.7,
"autosens_adjust_targets": true,
Here is an example of one of the autosens logs
Jul 24 20:38:23 myopenaps2 pump-loop.log: Autosens adjusting basal from 0.7 to 0.8; target_bg from 102.5 to 97; sens from 60 to 51.7 (autosens ratio 1.16)
Scott Leibrand
@scottleibrand
Jul 25 2017 01:44
that autotune must be old then, and you must have autotune disabled in your loop
ls -la ~/myopenaps/autotune/autotune_recommendations.log to confirm when it was written?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:45
-rw-r--r-- 1 root root 2052 Jul 24 00:07 /root/myopenaps/autotune/autotune_recommendations.log
Scott Leibrand
@scottleibrand
Jul 25 2017 01:45
??? this makes no sense
Dana Lewis
@danamlewis
Jul 25 2017 01:46
Pam, Cat your runagain, and paste here without any URL or API secret?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:46
k
Dana Lewis
@danamlewis
Jul 25 2017 01:47
@thacherT1D k. So what is not showing in NS - everything? Or is temp basals being rendered but just not pill, or ?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:48

To run again with these same options, use:

Screen Shot 2017-07-24 at 9.48.06 PM.png
Dana Lewis
@danamlewis
Jul 25 2017 01:49
K
Scott Leibrand
@scottleibrand
Jul 25 2017 01:49
did you change your pump ISF today from 80 to 60?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:49
no
Scott Leibrand
@scottleibrand
Jul 25 2017 01:49
do you remember when you did?
Thacher
@thacherT1D
Jul 25 2017 01:49
all of NS is up, but it says openAPS was last updated an hour ago
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:49
I never did, it is still 80 in my pump
Thacher
@thacherT1D
Jul 25 2017 01:50
Screen Shot 2017-07-24 at 6.49.11 PM.png
Dana Lewis
@danamlewis
Jul 25 2017 01:50
@thacherT1D blue basal line is rendering temps?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:50
maybe it shoudl be 60, but I have not changed that, since autotune keeps suggesting 100+ (which I don't agree with)
my pump settins are still 15 g, and isf of 80
Scott Leibrand
@scottleibrand
Jul 25 2017 01:51
would you rather try to figure out where it's getting the 60, or just re-install and have it start over?
Dana Lewis
@danamlewis
Jul 25 2017 01:51
Just checking the obvious - you don't have 2 pumps in play, do you?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:52
no, just one minimed
Dana Lewis
@danamlewis
Jul 25 2017 01:52
And you haven't attempted to run manually per phase c directions w creating a new profile manually?
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:53
no
Scott Leibrand
@scottleibrand
Jul 25 2017 01:53
if you want to keep investigating: which file did you see "sensitivity":60 in? can you do an ls -la <that file>?
Thacher
@thacherT1D
Jul 25 2017 01:54
@danamlewis my ns site is just showing my set basals, but the logs are saying they have set a neutral temp also...
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:54
-rw-r--r-- 1 root root 3088 Jul 24 21:47 profile.json
Thacher
@thacherT1D
Jul 25 2017 01:54
which doesn't totally make sense since I'm in the 300s...
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:55
this is a file that is in ~/myopenaps/autotune#
Scott Leibrand
@scottleibrand
Jul 25 2017 01:56
cat autotune/profile.json | jq . | grep sens
and cat autotune/profile.pump.json | jq . | grep sens
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:57
"autosens_max": 1.2,
"autosens_min": 0.7,
"rewind_resets_autosens": false,
"autosens_adjust_targets": true,
"sens": 113.977,
"sensitivities": [
"sensitivity": 113.977,
"sensitivity": 60,
"sensitivity": 60,
and the second...
"autosens_max": 1.2,
"autosens_min": 0.7,
"rewind_resets_autosens": true,
"autosens_adjust_targets": true,
"sens": 80,
"sensitivities": [
"sensitivity": 80,
"sensitivity": 80,
"sensitivity": 80,
Scott Leibrand
@scottleibrand
Jul 25 2017 01:58
hmm, so we have something wonky with multiple ISF schedules getting set the first time you run autotune, and then never adjusted, but used. grr...
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:58
thats wierd, why would it have rewind_resets_autosens true in one and false in the other?
would rerunning setup fix this?
Scott Leibrand
@scottleibrand
Jul 25 2017 01:59
yeah, that may be another symptom of the same problem
probably, but I wonder if I need you to keep a copy of your myopenaps directory in case we need it for more debugging
it will help, at least
PopcornPam
@PopcornPam_twitter
Jul 25 2017 01:59
ok, that is somewhere in the doc
Scott Leibrand
@scottleibrand
Jul 25 2017 01:59
as would getting rid of your multiple ISF schedules if they're all the same
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:00
I'm not sure where those would be buried.. settings? autotune? detect sensitivies?
Scott Leibrand
@scottleibrand
Jul 25 2017 02:01
multiple ISF schedules would be in your pump
let's back up your myopenaps directory before you change anything: cd && tar cvzf myopenaps.tar.gz myopenaps/
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:01
ok.. got rid of the multiple schedules
Thacher
@thacherT1D
Jul 25 2017 02:02
@danamlewis going to meet friends for dinner and walk away from the computer... maybe it will magically fix itself, haha... will check back in and get back to troubleshooting later, thanks for your help thus far :)
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:02
ok, I ran that, rerun setup? And should I start from scratch versus runagain?
and is there a reason to backup my other rig as well?
Scott Leibrand
@scottleibrand
Jul 25 2017 02:03
if you copy your runagain that should be fine, but we need to delete your myopenaps directory before pasting and running it
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:04
thats just remove myopenaps right?
Scott Leibrand
@scottleibrand
Jul 25 2017 02:04
after you've copied your oref0-runagain, then rm -rf ~/myopenaps/
and then paste it in to re-create it and set everything up w/o no bad memories
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:04
root@myopenaps2:~# ls
crontab.old.txt kernel.config remote_syslog src
crontab.txt myopenaps.tar.gz remote_syslog_linux_i386.tar.gz
root@myopenaps2:~#
Scott Leibrand
@scottleibrand
Jul 25 2017 02:05
looks like it's gone
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:05
Waht about this myopenaps.tar.gz
Scott Leibrand
@scottleibrand
Jul 25 2017 02:05
that's the backup we just made
go ahead and leave that
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:05
ok
Scott Leibrand
@scottleibrand
Jul 25 2017 02:06
I think this multiple ISF schedule overriding autotune thing is the same thing @tim2000s reported at one point, but we never really dug into
so we definitely should open a github issue to track it
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:06
interesting.. would that be a thing to pr, or not yet?
I am not sure how to open a github issue
and should I go ahead and remove and reinstall on all my rigs to be safe?
Scott Leibrand
@scottleibrand
Jul 25 2017 02:07
issue first. we can't PR until we have a fix. https://github.com/openaps/oref0/issues/new
@PopcornPam_twitter let's get this rig up and working the way you want it first
it may be that deleting the multiple ISF schedules is enough to make the other rigs happy: it'd be good to know whether that is true or not
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:08
ok
I am tempted to test it out by eating the last two moche ice creams I have avoided due to being high
Scott Leibrand
@scottleibrand
Jul 25 2017 02:09
heh, let's get it working first, and then once we see it working, maybe? :-)
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:10
probably a bad idea either way with a bg of 250 and heading to bed soon
ok, so at this point, is there anything to check, or just wait and see what happens tomorrow and report back?
Scott Leibrand
@scottleibrand
Jul 25 2017 02:12
after re-running oref0-setup, let's just make sure you're looping happily
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:13
oh, btw.. @scottleibrand and @danamlewis, I am in Seattle Sept 21 and 24. Would be great to meet irl if you guys are around
Scott Leibrand
@scottleibrand
Jul 25 2017 02:14
there's a good chance we will be...
Dana Lewis
@danamlewis
Jul 25 2017 02:15
Yay!
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:15
that would be so cool!
Dana Lewis
@danamlewis
Jul 25 2017 02:15
:+1:
@thacherT1D when you're back or whenever - check your logs and figure out if you're looping, and triple check your maxiob is set. Once you know you're looping properly, triple check capitalization of API secret in all the places. Once you're sure that's right..there's a fix in dev for the looping and temp basals are rendered on blue line but OpenAPS pill says unknown. So you could pull a fresh dev and see if that fixes it. If none of those work..let us know.
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:17
Also.. worth noting that my pump schedule hasn't changed since I started with openaps in April (rates have, just not the schedule). The issue that is hopefully fixed now has only been since using the new master.. I didn't have the issue when I was on the dev branch
Old settings refreshCould not parse autotune_data
ed. find: `monitor/pump_loop_completed': No such file or directory
I assume this is just startup noise?
Dana Lewis
@danamlewis
Jul 25 2017 02:19
Yep
Won't exist until it completes
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:19
right.. because I just deleted it
ok! Loop back up and running, and the numbers match current pump settings
Mollie Petruccelli
@mpets
Jul 25 2017 02:22
Help please, I was trying to rerun the setup script to update to 0.5.2 and change to token based NS authentication, and I think the time changed on my rig somehow? I can't log into it with ssh, but it is on my BT tether and when I look at papertrail, the time stamps on the logs are 3 hrs behind the current time. I tried connecting a cable to the UART and connecting that way but I'm getting lots of funny characters in the terminal window with that.
PopcornPam
@PopcornPam_twitter
Jul 25 2017 02:22
Thank you yet again for helping to figure this out!
Bedtime for me!
Dana Lewis
@danamlewis
Jul 25 2017 02:22
@popcornpam_twitter good job questioning when it didn't "feel" right!
Scott Leibrand
@scottleibrand
Jul 25 2017 02:24
@PopcornPam_twitter you may have been having the issue with not autotuning your ISF schedules all along, but it only became noticeable with the changes to SMB bolus snooze in oref0 0.5.x
if you wanted to you could look back in your /var/log/openaps/pump-loop.log.*.gz files to see if the ISF it was using was ~60 or closer to the autotuned value. not a big deal either way, though: we still have to do the same things to fix it regardless
@mpets you can hit enter a few times, then disconnect and re-connect your UART cable a couple times until the weird characters go away. or if that doesn't work, connect up the UART cable and then use the black power button on the rig to reboot it.
(hold down for up to 15s until red light goes off, then again for a couple seconds until it comes back on)
once you get in, you can run dpkg-reconfigure tzdata to re-set your timezone correctly
your ability to log in with ssh should be unaffected by any timezone stuff. that may be related to your hostname changing or something like that. when you get console working you can do all the steps on the troubleshooting page for accessing your rig and figure that part out
Mollie Petruccelli
@mpets
Jul 25 2017 02:29
Now with the cable it says "cannot exec '/dev/tty.usbserial-*"
Scott Leibrand
@scottleibrand
Jul 25 2017 02:29
that means you're running the command wrong
you're missing the first part of it
Mollie Petruccelli
@mpets
Jul 25 2017 02:30
this is what I'm enteringsudo screen /dev/tty.usbserial-* 115200
Scott Leibrand
@scottleibrand
Jul 25 2017 02:31
hmm, that's the correct command
Mollie Petruccelli
@mpets
Jul 25 2017 02:32
It was working but getting the weird characters a few minutes ago, and now it isn't working at all.
Scott Leibrand
@scottleibrand
Jul 25 2017 02:33
may need some other combination of plug unplug and reboot rig / computer to get it going again. not sure what exactly: I would just start trying things along those lines.
Mollie Petruccelli
@mpets
Jul 25 2017 02:37
Ok. Computer reboot got me back into it and into the console
Katie Aldridge
@kcrcgm
Jul 25 2017 02:54
@scottleibrand Hi Scott, I want to try out the Dex G5 android app with Openaps, (I know we just fixed the xdrip thing! But over the weekend I had a lot of trouble with missed readings with xdrip). When I tried to update/reinstall oref0.5.2 I got an error related to xdrip-fix (will post screenshot). So I tried to start all the way over reinstalling from the wifi and oref0 setup scripts. It seemed to work until at the very end when I chose to remove my existing crontab. I'll post screenshots of that too. It didn't ask me to reboot like it normally does. I checked to see what branch of oref0.5.2 I have and it said "fix xdrip" and "master". Did my setup finish successfully? If not, how can I fix it? Thank you so much.
Trying to reinstall with G5 no reboot prompt0.JPG
Trying to reinstall with G5 no reboot prompt2.JPG
Trying to reinstall with G5 no reboot prompt3.JPG
Trying to reinstall with G5 no reboot prompt4.JPG
Katie Aldridge
@kcrcgm
Jul 25 2017 03:03
I got back in to the rig and it made me change my pw. It is sending temp basals to my pump with up to date bgs. Just want to make sure I didn't mess something up. ;p
Mollie Petruccelli
@mpets
Jul 25 2017 03:05
@scottleibrand So now my issue is the rig won't connect to my wifi I think. I didn't touch any of the wifi setting though, any suggestions for how to trouble shoot? The rig isn't listed in the devices connected to the router and it will only connect to BT.
Screen Shot 2017-07-24 at 11.08.53 PM.png
JazzyCar
@JazzyCar
Jul 25 2017 03:11
Good evening, so just a general question about programming which reveals the sheer magnitude of my ignorance: When you are programming in Terminal if you make a mistake what can you do to fix it? (I.e. Is there a Crtl+Z-like shortcut, or a delete button?)
Katie Aldridge
@kcrcgm
Jul 25 2017 03:15
@JazzyCar I too am pretty ignorant, but I think the answer is... it depends. If you were editing a file (where it says nano or vi in the little snippet you cut and paste from the docs) then there is a way to delete. If you ran a command with a mistake and haven't hit enter yet, you can just backspace. ... If you describe your situation more, you can get more specific help (from me if I can, or from more knowledgeable people).
JazzyCar
@JazzyCar
Jul 25 2017 03:27
Thank you @kcrcgm. I was just trying to start my OpenAPS programming yesterday and copy-and-pasted the first line of code into Terminal. Several lines of code were then generated but they looked pretty different from the lines of code in the sample in the OpenAPS docs and it looked like several lines were missing. We tried to delete and use the ctrl+Z shortcut, but when that didn't work, we closed terminal thinking we could just restart...and then we couldn't get back into the program. So my question is, for future reference, if you think you've screw up...what do you do to go back?
Katie Aldridge
@kcrcgm
Jul 25 2017 03:40
@JazzyCar Your experience sounds almost exactly like mine except it was with the old setup so it would have been with a different copy/paste scenario. I tried to ctrl z and close it too, and had no idea what to do! So don't feel bad! :) Were you trying to run the big block of text for setting up wifi and oref0? The one where you go to the separate website and copy all the text and paste it into terminal? If you had an error running that script, just try pasting it again. It has taken me 2-3 times each time I've tried to do the new set up. Not sure why. A lot of commands are ok to just do over again, but many times commands need to be run in a certain order. And if you're editing a file, that's a whole different ballgame. So I don't have a general answer for you. Each time you need help with a certain error or mistake or whatever, search the Docs to see if you can figure it out, Google it if it's something general, and then ask for help on here or on the fb page. :) Best of luck!
JELCRAWFORD
@JELCRAWFORD
Jul 25 2017 03:52
@scottleibrand @danamlewis Do you suppose the problem described by @popcornpam is the same thing we are seeing? Also multiple ISF's programmed in pump and extended zero temps after bolus and prediction lines in the low's after a bolus.
JazzyCar
@JazzyCar
Jul 25 2017 03:58
@kcrcgm Thanks Katie! Yes, it was that big block of text for setting up wifi and oref0 -- so it's great to know you can just do it over and over again until it works. Thanks for your help. :-)
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 04:00
My name used to be on the "Loops In Progress" page of the docs (I had added it with a PR) (https://openaps.readthedocs.io/en/latest/docs/While%20You%20Wait%20For%20Gear/loops-in-progress.html). Now I don't see it there. Are those names dropped off the list?
Joshua22395
@Joshua22395
Jul 25 2017 04:00
Anyluck with the 630g pump and Riley link
Scott Leibrand
@scottleibrand
Jul 25 2017 04:10
@kcrcgm we merged the fix-xdrip stuff to dev, so if you want to install that again you'll need to follow the directions for installing oref0 dev.
@JELCRAWFORD perhaps. You could try removing the extra schedules as a workaround for testing...
@dirkgastaldo not intentionally. Check the file's history and see if you can find yourself?
@Joshua22395 nope, and no expectations that it'll ever be fully loopable.
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 04:13
@scottleibrand I just PR'ed it again
Scott Leibrand
@scottleibrand
Jul 25 2017 04:14
@Joshua22395 but you found Medtronic-flash, where most of the efforts to do so are ongoing. Keep an eye on that, but don't hold your breath. (Get yourself another pump to loop with in the mean time.)
Katie Aldridge
@kcrcgm
Jul 25 2017 04:19
@scottleibrand Thanks! Will I have any problems as I have it set up now, with master and xdrip fix but using G5?
Sandra Keßler
@SandraK82
Jul 25 2017 04:31
@PieterGit blueReader can be preordered currently as we are currently fulfilling orders from the crowdfunding. Hopefully we will fulfill any order within a few weeks. Preorder page is at https://blueToolz.de/blueorder/
Scott Leibrand
@scottleibrand
Jul 25 2017 04:49
@kcrcgm you don't have the xdrip fix applied right now, but with the G5 receiver and no xdrip configured, that's fine.
JELCRAWFORD
@JELCRAWFORD
Jul 25 2017 04:53
@scottleibrand - I have removed the extra ISF settings to a single average ISF. I'm not sure how this will work as we have variation of greater than 20% in ISF through the day and so I don't think (if I understand correctly) that autosense/autotune will be able to fully accommodate, but willing to see what happens.
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 05:38
Following docs to add token based authentication and on step 4:
Test the rig, e.g. by running openaps upload or openaps upload-ns-status or just running the pump loop. You’ll see the update from myrigname in the OpenAPS pill in Nightscout.
or easier to read: Test the rig, e.g. by running openaps upload or openaps upload-ns-status or just running the pump loop. You’ll see the update from myrigname in the OpenAPS pill in Nightscout.
I don't see a change, or maybe I don't know what I should be looking for, or what is meant by this: "You’ll see the update from myrigname in the OpenAPS pill in Nightscout."
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 05:57
I'm trying to change to token-based auth. I used the third method of entering the access token to the ns.ini file. When I look at my crontab.txt file I still see the old hash API secret, not the one that starts with token=
Thacher
@thacherT1D
Jul 25 2017 06:39
@danamlewis back up, looping, and ns site reporting correctly -- I think it has to do with the comms on this explore board, ordered another one from Morgan, so hopefully I'll have that fixed soon -- in the meanwhile I'm up on the latest oref0 and running AMA, Autosense and Autotune... we'll see, fingers crossed the basics keep running overnight :smile: :+1:
PieterGit
@PieterGit
Jul 25 2017 06:52
@dirkgastaldo : that's an oversight in the documentation. You must also set it in your crontab, API_SECRET=token=devicename-abcd where devicename-abcd is your hash
can you PR that in the docs?
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 06:53
Thanks, Pieter. To confirm, it is the crontab.txt file in the root directory?
PieterGit
@PieterGit
Jul 25 2017 06:54
you can edit it with crontab -e. i think crontab.txt is a file containing a backup of your crontab
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 06:54
okay, thank you
PieterGit
@PieterGit
Jul 25 2017 06:55
do you also use autotune? Then you also need openaps/oref0#563 (not merged to dev, but available on the autotune-tokenauth branch)
please report succes if you can run that. i tested it, but we always confirm new features/bug fixes by others and it has not been tested by others
PieterGit
@PieterGit
Jul 25 2017 07:07
@dirkgastaldo : forgot to ask: do you run the 0.5.2? Are you able to test the new bugfixes on a testrig? (you should always have a production / working rig)
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 07:11
I run 0.5.2
well, when I installed 0.5.2 I didn't select token-based auth. But reading the docs tonight I decided to install it.
and I guess that might not have been a good idea?
I installed it on both of my rigs, so both have 0.5.2 with token-based auth
and I do have autotune on both
PieterGit
@PieterGit
Jul 25 2017 07:13
the diff between my branch and 0.5.2 is very limited
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 07:13
I don't know what to do with oref0#563
PieterGit
@PieterGit
Jul 25 2017 07:13
you can install it as follows
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 07:13
should I just go back to API secret rather than token-based?
PieterGit
@PieterGit
Jul 25 2017 07:14
cd ~/src/oref0
git pull
git checkout autotune-tokenauth
npm run global-install
it's up to you if you want to go back. the diffs between the versions are very limited.
as long as you don't change your nightscout AUTH_DEFAULT_ROLES from readable, there is no harm
and you can still use plain API_SECRET authentication
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 07:16
I didn't have that variable in my NS, so I added it, but I set it to readable
PieterGit
@PieterGit
Jul 25 2017 07:17
should i help you install it on your rigs?
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 07:18
you mean install oref0#563?
PieterGit
@PieterGit
Jul 25 2017 07:18
yes
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 07:18
I can install it on one rig...
can you confirm...
that a rig running 0.5.2 with token-based auth installed as per docs, is fine?
PieterGit
@PieterGit
Jul 25 2017 07:19
it's fine, but autotune won't work. i noticed the bug after the 0.5.2 release
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 07:20
if I have autotune enabled it's only suggesting basal rate changes, not making them, or is it making them?
PieterGit
@PieterGit
Jul 25 2017 07:33
it's copying the profile. i'm not sure if you're loop will use the basal rates, or only the ISF. good question
you can always disable it in your crontab if you want to use it manual
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 07:35
Okay, I will take a closer look at that and not make anymore changes tonight
PieterGit
@PieterGit
Jul 25 2017 07:37
please tag me if you have questions. i would like to have the PR tested and merged to dev if ok
Dirk Gastaldo
@dirkgastaldo
Jul 25 2017 07:42
will do Pieter
PieterGit
@PieterGit
Jul 25 2017 07:43
@dirkgastaldo : afaict the profile.json is only copied in the autotune directory, see https://github.com/openaps/oref0/blob/autotune-tokenauth/bin/oref0-autotune.sh#L223-L224
@scottleibrand @danamlewis i can't find any code or documentation if the pump basals/isf/csf from autotune is automatically used. I thought the autotune profile was used, but now I'm doubting that. Can you explain?
Tim Street
@tim2000s
Jul 25 2017 08:08
@PieterGit As far as I can tell, the Autotune profile.json is copied into settings and should be used, however I've noticed that on ISF at least, where I have more than one, it isn't being used. It's a little more difficult to tell on the basal rates because I've matched my pump to Autotune.
I had this conversation previously and @scottleibrand confirmed that Autotune data is supposed to replace that pulled from the pump.
PieterGit
@PieterGit
Jul 25 2017 08:28
i thought so too, but i couldn't find the cp command that the autotune profile.json is copied as a settings/profile.json
PieterGit
@PieterGit
Jul 25 2017 08:36
i see https://github.com/openaps/oref0/blob/master/bin/oref0-setup.sh#L975 copies autotune/profile.json to settings/autotune.json in the crontab, but I don't see how settings/autotune.jsonis ever used.
mattmholliday
@mattmholliday
Jul 25 2017 10:08
hello, i'm currently experimenting with autotune do all the values in the profile need to be in mg/dl or just the ISF?
PieterGit
@PieterGit
Jul 25 2017 12:15
testing the PR is appreciated
Tim Street
@tim2000s
Jul 25 2017 12:20
@PieterGit, I don't think the Autotune.json is ever used. All the data from Autotune is put into profile.json and used from there. If you dig through it has all the new basal rates and the ISF changes when you have a single ISF.
PieterGit
@PieterGit
Jul 25 2017 12:22
it's put in the autotune/profile.json. I don't see how it gets into the settings/profile.json.
Brian Rabinovitz
@bdr1177
Jul 25 2017 12:36
I'm getting a lot of communication errors. Could this be an indication that the radio on my rig or pump is going bad? Specifically, the following error:
Old pumphistory: mmeowlink.exceptions.CommsException: No acknowledgement from pump on wakeup. Is it out of range or is the battery too low?
PieterGit
@PieterGit
Jul 25 2017 12:40
@bdr1177 can be various reasons:
Brian Rabinovitz
@bdr1177
Jul 25 2017 12:46
@PieterGit Yes, I've ruled out most of those. Battery is good. Distance is negligible. Error often resolves itself, but can take almost 30 min to do so. It could be other radio devices, but I can't think of any new devices that I've added to the environment. What would be the nature of an error between rig and pump?
PieterGit
@PieterGit
Jul 25 2017 12:48
What signal strength is reported with mmtune? I think some batteries in cases also can block the rf signal, but I don't know what's the best case design.
I also had situations in which rig-pump communication went well after re-inserting the pump battery
I was thinking of adding a rule that: after unsuccesful pump loop AND pump_completed_time > mmtune_time force mmtune else 25% chance on mmtune
that would resolve it more quickly
Brian Rabinovitz
@bdr1177
Jul 25 2017 12:50
I had some bad results (916.636", 0, -99), but then it resolved to 916.828", 5, -52. Unfortunately, a lot of this happens when I'm not with my t1d, so I'm relying on others to tell me that the rig is close to the pump.
So the rule would basically cause mmtune to run more frequently if there are failures?
PieterGit
@PieterGit
Jul 25 2017 12:51
yes, now it mmtunes in 25% of the cases
Brian Rabinovitz
@bdr1177
Jul 25 2017 12:52
Yes, that might be helpful because it does seem to eventually resolve after mmtune runs a couple times. Not sure if it is related to mmtune running or not.
PieterGit
@PieterGit
Jul 25 2017 12:53
what case do you use?
Brian Rabinovitz
@bdr1177
Jul 25 2017 12:53
Usually no case or a soft case.
rodzilla55
@rodzilla55
Jul 25 2017 14:11
@thebookins Thanks... would love to know how the native G5 integration is progressing. The error had to do with incompatible framework something-or-another. I'll have to dig back in. Is it easy for you to give me a summary of what your mods to the master Loop app branch are? Are you just disabling some functionality or ?
@tompetrillo The directions I was using for iPhone Loop App for offline looping: https://gist.github.com/thebookins/fb2f9143ff03172a278c428b35933c04. Ihave the Loop App installed and working, but haven't switched over to loop offline yet... I'm not confident that it won't be a big hassle to keep changing the Nightscout IP address setting for the bluetooth tether.
ejoe132
@ejoe132
Jul 25 2017 18:31
So started using my new G5 today, still have openaps running off of my g4 which is still being used(haha using g4 & g5 together right now) Switching to g5 for openaps though, Im now rerunning my setup script and saying g5 when prompted, what do I have to do with nightscout to get it to switch and pull from my g5 system?
Scott Leibrand
@scottleibrand
Jul 25 2017 18:37
do they both push to the same Dexcom Share/Clarity account? or does the G5 have a new login and password?
ejoe132
@ejoe132
Jul 25 2017 18:40
@scottleibrand I believe they are the same, but am checking now. If their the same it shouldnt need changing correct? Thats what I was thinking but was wondering cause now that I have both running they are slightly different numbers but maybe once I remove the G4, the G5 reading will just take its place?
ejoe132
@ejoe132
Jul 25 2017 18:47
@scottleibrand Yes they are both under the same share and clarity account
Scott Leibrand
@scottleibrand
Jul 25 2017 18:47
AFAIK NS just pulls the same numbers that Dexcom Follow gets from a given Share/Clarity account
ejoe132
@ejoe132
Jul 25 2017 18:56
@scottleibrand Ok Ill try pulling the G4 off in a bit and see if it just starts grabbing the G5 numbers, thanks Scott
BarrasB
@BarrasB
Jul 25 2017 19:18

Hi. I just decided to find a solution for viewing my CGM info on my phone. I understand that my insulin pump (Medtronic 554 with FW 2.8B. European) is not compatible with OpenAPS, but I wonder if this is only in regards to the closed loop-functionality, like giving the insulin pump commands to administer bolus.
Do any of you know if it will be possible with OpenAPS to listen in on the signals sent to the pump from my Medtronic CGM, without interacting with it?

The reason I'm asking is because this seems possible with Nightscout in combination with the MMCommander, and with Loop. The problem with the MMCommander is its need to be pysically plugged into my phone with a cable, and also the lack of it to work when offline (at least as far as I've found out this far). The problem with loop is I would rather not switch to iPhone.

Scott Leibrand
@scottleibrand
Jul 25 2017 19:32
Yes, you can set up an OpenAPS rig just like you would for a loopable pump, and it will upload CGM and insulin dosing data to NS. It'll just fail to set any temp basals.
BarrasB
@BarrasB
Jul 25 2017 19:34
Great :smile: That settles it then. Off to buy an intel edison and the explorer board immediately! Thanks a lot :smile:
Dana Lewis
@danamlewis
Jul 25 2017 20:00
@BarrasB you could probably still use OpenAPS as an open loop if you want to get pushover alerts in case of any carbs needed/bolus needed (and you can configure the threshold for how much insulin you want to get alerted at)
Scott Leibrand
@scottleibrand
Jul 25 2017 20:12
some of that is built into oref1 if you turn SMBs on. not sure if that'll be everything you want, but it'd be a start
Dana Lewis
@danamlewis
Jul 25 2017 20:13
:+1:
ejoe132
@ejoe132
Jul 25 2017 20:26
So pulled off the G4 sensor, new G5 is running and they were under the same account but ns is not getting the bg any longer. Is there something Id have to change in heroku possibly? Thinking it should have just kept grabbing it since it was the same account
Dana Lewis
@danamlewis
Jul 25 2017 20:27
I thought G5 needed a new account for share, or some people had similar issues. @ELUTE might have some ideas, but if you want a more rapid response and no one knows here, I'd ask on CGMITC on Fbk
Kate Farnsworth
@ELUTE
Jul 25 2017 20:29
It should be fine. @ejoe132 did you turn sharing on in the g5?
danamlewis @danamlewis loves the bat signal
Dana Lewis
@danamlewis
Jul 25 2017 20:29
thanks @ELUTE
Kate Farnsworth
@ELUTE
Jul 25 2017 20:30
np
DaisyHK
@laurathiessen
Jul 25 2017 20:31
image.png
ejoe132
@ejoe132
Jul 25 2017 20:32
@ELUTE that may be the ticket, feel like an idiot didnt think of that, was thinking it would just go as it had with g4. But looking on my g4 share(blue app) I just have myslef listed as a follower and my email is in there, would I just need to do the same thing in the g5 app?
Kate Farnsworth
@ELUTE
Jul 25 2017 20:32
Usually you just need to turn the sharing on with g5. But adding yourself as a follower and confirming that the follow app is receiving data won't hurt
ejoe132
@ejoe132
Jul 25 2017 20:34
@ELUTE Makes sense that that would be my issue, top right corner of my g5 app is now "colorful" so Ill see if it pops up in ns in a few. Thank you so much
Kate Farnsworth
@ELUTE
Jul 25 2017 20:35
:)
Martin Haeberli
@mhaeberli
Jul 25 2017 20:40
@ejoe132 Went through this two months ago with our own G4 / G5 transition. after much difficulty (and lots of calls with Dexcom support, etc) turned out it was VERY IMPORTANT to set up a completely new share account for the G5. Among other challenges ...
Dana Lewis
@danamlewis
Jul 25 2017 20:41
ok I'm NOT crazy, I did see/hear that somewhere :) thanks for sharing @mhaeberli
Martin Haeberli
@mhaeberli
Jul 25 2017 20:41
the G5 would not flow data to the old share account.
Dana Lewis
@danamlewis
Jul 25 2017 20:41
wonder if they've fixed that since then
Martin Haeberli
@mhaeberli
Jul 25 2017 20:41
I would hope, but ...
also, other free advice - with G5, it uses bluetooth of course to talk to iPhone (haven’t tried android version (yet))
our initial iPhone had some kind of marginal flaky bluetooth behavior, one symptom is it would take all 30 minutes to even PAIR with the phone.
so we had to get a new iPhone.
THEN we learned that having the phone paired with the G4 receiver caused some kind of ‘interference’ with G5 data flowing - our T1 was bravely wearing G4 on one arm and G5 on the other, which would have been fine, but for...
ALSO we learned - ‘Tile’ app - so useful for finding lost Dex receiver and lost openaps - interferes with G5 data from G5 transmitter to iPhone. So don’t run the Tile app unless you need it, then force quit it if need be.
ejoe132
@ejoe132
Jul 25 2017 20:53
@mhaeberli What was the issue you were experiencing? Only ask because turning on the share has my loop running properly, am now recieving new bg and setting temp basals appropriately
Dana Lewis
@danamlewis
Jul 25 2017 20:54
:+1:
Martin Haeberli
@mhaeberli
Jul 25 2017 20:54
@ejoe132 yay!
issue we were experience was NO data flowed to share cloud nor to Nightscout bridge ...
so maybe this is cleaned up
ejoe132
@ejoe132
Jul 25 2017 20:56
Ok just checking! Have you liked the g5? kinda on the fence so far, but I hate change so itll take a few days. I do like that I can change the Y axis from 400 to 300 in the app( wish it would go to 200 or less as I hardly ever see anything that high now thanks to openaps
Martin Haeberli
@mhaeberli
Jul 25 2017 20:57
overall, our t1 likes it a lot - easily displays on apple watch, back fills data (tho obviously openaps can’t really use backfilled data for live decisions). sometimes surprisingly accurate. sometimes some calibration drift ...
ejoe132
@ejoe132
Jul 25 2017 21:00
@mhaeberli ok awesome, sounds like g4 accuracy wise.
Diadon81
@Diadon81
Jul 25 2017 21:06
Dears somebody know how to remove that annoying debugging messages from jubilinux 0.1.1 serial console?
182867] dhd_bus_init: enable 0x06, ready 0x06 (waited 0us)
[   71.193344] wifi_platform_get_mac_addr
[   71.200096] wifi_get_mac_addr_intel: unable to open /config/wifi/mac.txt
[   71.211504] Firmware up: op_mode=0x0015, MAC=90:b6:86:01:13:41
[   71.243523] Firmware version = wl0: Aug 15 2014 15:09:17 version 6.20.190.2 (r497061) FWID 01-bb9bbd07
[   71.254823] dhd_preinit_ioctls wl ampdu_hostreorder failed -23
[   76.018686] wl_bss_connect_done succeeded with 64:3e:8c:ba:6a:40
[   76.045023] wl_bss_connect_done succeeded with 64:3e:8c:ba:6a:40
Diadon81
@Diadon81
Jul 25 2017 21:08
thanks a lot :)
Dana Lewis
@danamlewis
Jul 25 2017 21:09
:+1:
Diadon81
@Diadon81
Jul 25 2017 21:12
Just want to discuss one idea. Everybody said that true loop device should deliver both of insulin able to deliver both insulin and glucagon. Is it possible to use two medtonic pumps one with insulin another one with glucagone, to implement this scheme? Of course oref0 algorithm should be carefully adopted for such scheme.
Dana Lewis
@danamlewis
Jul 25 2017 21:15
"Some" people think that.
Diadon81
@Diadon81
Jul 25 2017 21:15
and what the conclusion?
Tim Street
@tim2000s
Jul 25 2017 21:15
@Diadon81 I wouldn't want to carry two pumps around.
Dana Lewis
@danamlewis
Jul 25 2017 21:15
Yes, you could do something like that, but I don't believe glucagon is shelf-stable enough to make that worthwhile reight now, and someone would have to figure out the glucagon curve like the insulin curves so it could be added to an algorithm.
Tim Street
@tim2000s
Jul 25 2017 21:16
Then there are the glucagon problems.
Dana Lewis
@danamlewis
Jul 25 2017 21:16
Doesn't seem worth it for every day. Might be interesting for all-day hike scenario where you have to eat even with no insulin in order to keep BGs high enough...but not a regular situation for most people to make all that work worth it.
Tim Street
@tim2000s
Jul 25 2017 21:17
(And to add to that, iLet have been having issues tuning theirs!)
Dana Lewis
@danamlewis
Jul 25 2017 21:17
yep
might be because they over-rely on it, though
Diadon81
@Diadon81
Jul 25 2017 21:17
but at least it will be interesting from conceptual point of view, is this scheme makes compesation better.
of course I agree that such scenario with two pumps isn't very comfortable. But maybe it will push someone to start working on such kind of pump.
Diadon81
@Diadon81
Jul 25 2017 21:21
as always that product will be ready in next 5-7 years :)
Scott Leibrand
@scottleibrand
Jul 25 2017 21:21
@ejoe132 the G4 and G5 use the exact same sensors and calibration / filtering algorithms. the only differences are in how the data flows, and where the calibration occurs (receiver vs. transmitter)
rodzilla55
@rodzilla55
Jul 25 2017 21:22
yep, too slow... I think a significant portion of the barrier is the stability of the glucagon. If there was a stable version available, I'd wear 2 pumps now... someone in this brilliant community would build a awesome 3d case for 2 pumps and a rig.
Scott Leibrand
@scottleibrand
Jul 25 2017 21:22
@Diadon81 probably more like 2-3 years now, but I doubt it will outperform OpenAPS in the "highly motivated" population when it's released
Diadon81
@Diadon81
Jul 25 2017 21:24
how long it last? glucagon
Scott Leibrand
@scottleibrand
Jul 25 2017 21:26
right now it's supposedly only good for a couple days
in the iLet trials they've been replacing it every day or two
Diadon81
@Diadon81
Jul 25 2017 21:28
not so bad, I am using 3ml insulin vial per one week.
or less
one box (5 vials) per one month.
Tim Street
@tim2000s
Jul 25 2017 21:49
@scottleibrand - depends where you are WW never got the 505 algo so G5 is different to G4.
(And better in my experience)
Scott Leibrand
@scottleibrand
Jul 25 2017 21:58
ah, didn't know that. (and that's dumb)
Cameron Chunn
@applehat
Jul 25 2017 22:06
Just had my first Endo appointment since using OpenAPS. Went from a 7.8 a1c to 6.2 :). When I told my Endo that "I plan to get into the 5s" she said "are you not already fighting lows to achieve a 6.2?".. nooe. I made a believer today. Haha.
She also said "I read an article on Reddit about some lady from Alabama who made something like what your using" @danamlewis :p
Shortest Endo visit ever, too. "Uh, I don't really have any suggestions to make. Guess we're all good!"
Scott Leibrand
@scottleibrand
Jul 25 2017 22:12
:-D :+1: :clap: :confetti_ball: :microphone: :droplet:
Ryan Dalisky
@rdalisky
Jul 25 2017 22:37
@applehat Solid work sir!
PieterGit
@PieterGit
Jul 25 2017 23:19
did quite some work on out-of-the-box pebble pancreabble support. also added insulinReq and carbsReq notifcation (similar to pushover)
will test again and release the PR tommorow
Cameron Chunn
@applehat
Jul 25 2017 23:29
Nice @PieterGit!