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

1st
Sep 2016
japrobo
@japrobo
Sep 01 2016 01:38
@bfaloona back troubleshooting our issue from last night with edison trying to set rate of 0 and failing. Thanks for the idea re: manually setting temp basal. This is what happened:
root@ubilinux:~/myopenaps# echo { \"temp\": \"absolute\", \"rate\": 0.0, \"duration\": 30 } > my_temp_basal.json
root@ubilinux:~/myopenaps# openaps use pump set_temp_basal my_temp_basal.json 
{
  "requested": {
    "duration": 30, 
    "rate": 0.0, 
    "temp": "absolute"
  }, 
  "temp": "absolute", 
  "timestamp": "2016-08-31T21:36:45.124550", 
  "rate": 1.375, 
  "recieved": false, 
  "duration": 0
@bfaloona can't figure out why "recieved": false
japrobo
@japrobo
Sep 01 2016 02:05
@Pazoles I think we are having a problem you mentioned having in June re: having trouble getting the suggested basal enacted on the pump. did you ever figure that out?
Matt Pazoles
@Pazoles
Sep 01 2016 02:29
Yeah, turns out the firmware on the pump we had was too new, and only allowed read access and not write.
I think that was a 523?
japrobo
@japrobo
Sep 01 2016 02:30
@Pazoles ok, think our problem is different. thanks for your reply! Ours was working just fine before with the same pump - and our raspberry pit rig works just fine with this pump
Matthew Williams
@mwilliams
Sep 01 2016 02:30
Well - my first successful run of do-everything from @bewest ns alias's - but still nothing to upload. Only thing I tweaked was to use my clean glucose file to massage the data from my mm enlite cgm. Trying to revert back to using the standard glucose.json output. Any thoughts otherwise from anyone? Really want to go closed-loop but good reporting has me hesitant (I've got push bullet notifications which is nice and corrections have been appropriate, but would be nice to visualize over 24 hours) https://gist.github.com/mwilliams/dab75439e43605f78bb46baef03c9527
Matt Pazoles
@Pazoles
Sep 01 2016 02:31
Cool, @bewest might have some diagnostic suggestions for you @japrobo
japrobo
@japrobo
Sep 01 2016 02:32
@Pazoles @bfaloona we ran this validation command and results look good, but loop is still getting "received: false"
root@ubilinux:~/myopenaps# openaps use pump test_oref0_compat_menu
{
  "completed": true, 
  "results": [
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "ESC", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }, 
    {
      "received": true, 
      "key": "DOWN", 
      "raw": "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
    }
  ]
garykidd
@garykidd
Sep 01 2016 02:37
pi@raspberrypi:~/APS000 $ openaps report add raw-cgm/glucose-raw.json JSON cgm oref0_glucose --no-raw --hours 2.0
added cgm://JSON/oref0_glucose/raw-cgm/glucose-raw.json
pi@raspberrypi:~/APS000 $ openaps use cgm config --G5
{
"model": "G5"
}pi@raspberrypi:~/APS000 $ openaps report invoke raw-cgm/glucose-raw.json
cgm://JSON/oref0_glucose/raw-cgm/glucose-raw.json
raw-cgm/glucose-raw.json raised Python int too large to convert to C long
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 536, in main
for egv, raw in itertools.izip_longest(iter_glucose, iter_sensor):
File "/usr/local/lib/python2.7/dist-packages/dexcom_reader/readdata.py", line 300, in iter_records
for x in reversed(xrange(start, end)):
OverflowError: Python int too large to convert to C long
````````````````````````
Appreciate any help troubleshooting^^
japrobo
@japrobo
Sep 01 2016 02:44
@mwilliams Which file should your data be in? did you try to cat that file so you could confirm the data is there and looks as expected? that is where I would start to troubleshoot...
Matthew Williams
@mwilliams
Sep 01 2016 02:47
@japrobo I guess that's part of the issue? Following https://github.com/openaps/docs/blob/master/docs/docs/walkthrough/phase-1/openaps-to-nightscout.md I seem to get through the steps, I see data, but nothing seems to make it to my NS instance (and my preflight is good - so my keys and stuff are fine). Just tweaked a few things and do-everything is running again, will dump my nightscout directory contents/file sizes
japrobo
@japrobo
Sep 01 2016 02:48
@Pazoles @bfaloona @bewest We figured out our issue, we believe. thanks for your support. When we are running a combo bolus, we cannot set a temp basal at the same time. We suspended the pump, resumed the pump and now it will accept our temp basal commands. We have been running this loop for weeks now - amazing the details we still haven't learned. :)
Matthew Williams
@mwilliams
Sep 01 2016 02:49
@japrobo 4 -rw-r--r-- 1 pi pi 4 Aug 31 22:47 preflight.json
4 -rw-r--r-- 1 pi pi 2 Aug 31 22:47 recent-treatments.json
4 -rw-r--r-- 1 pi pi 2 Aug 31 22:47 uploaded.json
4 -rw-r--r-- 1 pi pi 2 Aug 31 22:47 recent-missing-entries.json
4 -rw-r--r-- 1 pi pi 2 Aug 31 22:47 uploaded-entries.json
4 -rw-r--r-- 1 pi pi 2004 Aug 31 22:47 openaps-status.json
4 -rw-r--r-- 1 pi pi 2259 Aug 31 22:47 uploaded-recent-devicestatus.json
@japrobo so I've got some content in there, though the pertinent files (uploaded/recent-missing-entries) seem to be empty. I would even settle for even just seeing battery info on my NS dashboard at this point :P
japrobo
@japrobo
Sep 01 2016 02:52
@mwilliams so I followed a different set of instructions - they have been rewritten (this is a good thing!) since mine was first setup, so sorry all I have are questions... where are the pertinent files supposed to get data from (what is the input)? Can you manually run the report to generate the data that you are missing?
garykidd
@garykidd
Sep 01 2016 02:56

pi@raspberrypi:~/APS000 $ openaps report show

cgm://JSON/oref0_glucose/raw-cgm/glucose-raw.json
```````````````
would there be anything in the report if it cant be invoked?
Matthew Williams
@mwilliams
Sep 01 2016 02:57
@japrobo running them manually turns them up empty each time, I'm thinking it's the format of my glucose.json or something because I'm using the enlite cgm
japrobo
@japrobo
Sep 01 2016 02:59
@mwilliams yea, that is obviously what you need to troubleshoot. but you probably knew that already! I would think the next step would be to pick the first one in the loop and study the output of the input files and make sure they look accurate. Good luck!!
Matthew Williams
@mwilliams
Sep 01 2016 03:00
@japrobo Thanks - who knew getting the loop running would be the easiest part out of all of this? :) - fun exercise anyway!
japrobo
@japrobo
Sep 01 2016 03:03
@mwilliams for me the hardest part was getting the upload to nightscout to work, so you are in good company :)
Matthew Williams
@mwilliams
Sep 01 2016 03:05
good to know!
garykidd
@garykidd
Sep 01 2016 03:07
@japrobo lilhelp?
garykidd
@garykidd
Sep 01 2016 03:22
This message was deleted
japrobo
@japrobo
Sep 01 2016 03:23
@garykidd heading to bed, so won't be much help .... but, no if it can't be invoked then there won't be a report. I suggest looking at the data inside the json files that are causing these errors and see if the data looks like you would expect. I don't recognize the error you are getting, sometimes I google errors and get clues that way. Good luck!
garykidd
@garykidd
Sep 01 2016 03:23

pi@raspberrypi:~/APS000 $ openaps report add last_2_hours_cgm.json JSON cgm oref0_glucose --no-raw --hours 2
added cgm://JSON/oref0_glucose/last_2_hours_cgm.json
pi@raspberrypi:~/APS000 $ openaps report invoke last_2_hours_cgm.json

cgm://JSON/oref0_glucose/last_2_hours_cgm.json
last_2_hours_cgm.json raised Python int too large to convert to C long
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 536, in main
for egv, raw in itertools.izip_longest(iter_glucose, iter_sensor):
File "/usr/local/lib/python2.7/dist-packages/dexcom_reader/readdata.py", line 300, in iter_records
for x in reversed(xrange(start, end)):
OverflowError: Python int too large to convert to C long

===========================================
I'm lost why this doesn't work ^^
@japrobo thanks. I dont think looking at data files would help though since report doesn't actually get invoked ever. thanks though. been stuck here more or less for days now....crazy.
garykidd
@garykidd
Sep 01 2016 03:30
==========================================
Can anyone confirm that what I was attempting to do above should work? Just as a sanity check?
diabeticgonewild
@diabeticgonewild
Sep 01 2016 05:49
@garykidd I'm pretty sure this error is occurring because you are trying to get RAW values with a G5, which can't happen (yet).
garykidd
@garykidd
Sep 01 2016 09:18
@diabeticgonewild hmmm I was actually just trying to follow..http://openaps.readthedocs.io/en/dev/docs/walkthrough/phase-1/using-openaps-tools.html#g5-support and was pretty sure that I had taken the appropriate steps w/r/t G5 implementation. You defintely have a lot more experience than I do though....can you think of anything else? Appreciate the feedback and help.
garykidd
@garykidd
Sep 01 2016 09:37
===================================================
So when I openaps use cgm oref0_glucose --no-raw --hours 10 I get back what seems to be the appropriate response. I do have large gaps in the data as a result of being away from my dex receiver, however; not sure of this has anything to do with errors?
also when I openaps use cgm oref0_glucose --no-raw I get back what seems to be an appropriate response......a large pool of formatted cgm records.
garykidd
@garykidd
Sep 01 2016 09:42
further, when I attempt to add a report, in my test example, 10 hours by openaps report add CGM_TEST_10.json JSON cgm oref0_glucose --no-raw --hours 10
that works reporting back:
This message was deleted
added cgm://JSON/oref0_glucose/CGM_TEST_10.json
but still, I cant seem to invoke report.....keep getting that dreadful response....
CGM_TEST_10.json  raised  Python int too large to convert to C long
garykidd
@garykidd
Sep 01 2016 09:49
@bewest @scottleibrand @jasoncalabrese @diabeticgonewild @ecc1 @eluhrs appreciate all the help.....totally stuck.
Scott Leibrand
@scottleibrand
Sep 01 2016 10:46
Are you doing --g5 ?
If so, and it still doesn't work, you can try iter_glucose instead of oref0_glucose.
Maybe try them with use first before you try to do the report stuff.
garykidd
@garykidd
Sep 01 2016 10:47
exactly....weird that the 'use' works but that I still cant seem to invoke a report...right?
so use oref0_glucose works fine though...as does use iter_glucose
but I cant invoke oref0_glucose report successfully
garykidd
@garykidd
Sep 01 2016 10:54
@scottleibrand I was able to invoke a report based on iter_glucose, finally..... will it be fine to use iter_glucose instead of oref0_glucose? Will it impact interface with NS?
Scott Leibrand
@scottleibrand
Sep 01 2016 10:56
That used to be the only way to do it. There might be something with date formats that you need to correct with an additional command: we'll see.
Or maybe we will need to fix oref0 to handle both date formats.
garykidd
@garykidd
Sep 01 2016 10:57
hmmm when you say both date formats....you mean NS format and what's the other?
Tim Street
@tim2000s
Sep 01 2016 10:59
@scottleibrand - I've got the following crontab set up. It worked fine yesterday. It's stopped today. I've restarted, and it seems to make no difference. openaps do-everything runs manually. Any ideas?
SHELL=/bin/bash
PATH=/home/pi/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:home/pi/timsaps
HOME=/home/pi

*/10 * * * * (cd /home/pi/timsaps && time openaps do-everything) 2>&1 | logger -t openaps-loop
garykidd
@garykidd
Sep 01 2016 11:04
@scottleibrand so do you think that this problem is not specific to me then?
Scott Leibrand
@scottleibrand
Sep 01 2016 11:11
@garykidd I think this is the same thing lots of people using G5 have reported.
@tim2000s what do your logs show?
garykidd
@garykidd
Sep 01 2016 11:13
@scottleibrand thanks...so I guess I should just try to proceed with iter_glucose. Is there a separate mechanism to report this type of problem?
Scott Leibrand
@scottleibrand
Sep 01 2016 11:20
If there's not already a GitHub issue for it, should open one. Under openaps/openaps.
Tim Street
@tim2000s
Sep 01 2016 11:33
@scottleibrand
Sep  1 12:30:04 raspberrypi openaps-loop: model.json  raised  /dev/serial/by-id/usb-0a21_8001-if00-port0 al
ready in use
Does this mean there's an issue with the carelink stick?
garykidd
@garykidd
Sep 01 2016 11:39
@scottleibrand attempted to document the issue. That' my first submitted issue so if you happen to review it, please let me know if I should amend.
Scott Leibrand
@scottleibrand
Sep 01 2016 11:45
@tim2000s you should add something like * * * * * killall -g --older-than 10m openaps to your crontab
and run it manually to see if that frees up the port
looks like you have an old process that is stuck, still using the radio stick.
Tim Street
@tim2000s
Sep 01 2016 11:46
That's what's really odd - it only happens when I run from the crontab. When I run manually, the same error doesn't show!
Is it possible I'm missing an env variable?
Tim Street
@tim2000s
Sep 01 2016 11:55
@scottleibrand Figured it out - due to doing testing, I was overlapping two processes, and the first one hadn't relinquished the stick by the time the second had initiated, generating the error and causing both preflights to fail.
Because the second one with the unresponsive stick was restarting the USB connections.
Nothing like sitting and watching a tail of the logs.
Tim Street
@tim2000s
Sep 01 2016 12:02

What I am seeing is:

Sep  1 13:00:27 raspberrypi Stick transmit[TransmitPacket:ReadPumpModel:data:unknown] reader[ReadRadio:size:0] download_i[2] status[<LinkStatus:0x03:status:size=??LinkStatus:error:True:reason:[]:size(0)>] poll_size[0] poll_i[False] command[<LinkStatus:0x03:status:size=??LinkStatus:error:True:reason:[]:size(0)>]:download(attempts[2],expect[0],results[0]:data[0]):BAD AILING

But the docs so far don't cover what this is and how it might be fixed, and it doesn't seem to affect functioning of the system when it occurs in the pre-flight...!

Scott Leibrand
@scottleibrand
Sep 01 2016 12:03
"Bad ailing" is a warning that is quite common with carelink. If everything seems to recover after that, I wouldn't worry about it.
Tim Street
@tim2000s
Sep 01 2016 12:04
Seems fair to me.
Scott Leibrand
@scottleibrand
Sep 01 2016 12:04
@garykidd that issue looks good to me. @bewest will likely comment on it if he needs more info.
or maybe someone else who actually has a G5 to test with will beat him to it and figure out what needs to be fixed.
Tim Street
@tim2000s
Sep 01 2016 12:06
Do you know if anyone has run an openaps with two Nightscout instances, one for reporting and one for glucose data before? I'm having issue with the 640 uploader and nightscout dev branch at the moment, so having to run master to collect data. Was thinking I might need to run a separate dev for visualistion.
Tim Street
@tim2000s
Sep 01 2016 12:19
I think I can see why people prefer other options over the carelink stick. It's somewhat flakey isn't it?
Eric Jensen
@elnjensen
Sep 01 2016 12:23
@tim2000s Yes, I planned to use Carelink stick for a while, but switched away faster than originally planned because it was so flaky (really short and temperamental range). Using TI stick now - expensive, and a pain to have to flash firmware, but performance is vastly better.
Tim Street
@tim2000s
Sep 01 2016 12:37
I've got a rileylink on the way, so I'm very tempted to use that instead.
@elnjensen - know anyone who has built with one of those? And any reference on the Carelink range?
Dana Lewis
@danamlewis
Sep 01 2016 12:41
@tim2000s I have a RileyLink. It's better than carelink but less than TI
Scott Leibrand
@scottleibrand
Sep 01 2016 12:41
you can solder a ribbon cable to an Edison and plug the RileyLink into it: that is actually our main mobile rig. range is way better than carelink: almost as good as TI (and plenty good for wearing on-body with the pump)
Dana Lewis
@danamlewis
Sep 01 2016 12:42
:+1:
@tim2000s you may be the first running two instances? Or at least talking about it, haven't seen anyone else trying that yet
Tim Street
@tim2000s
Sep 01 2016 12:42
Hmmm. Don't have an edison set up at the moment.... That's a whole other ball game...
Tim Street
@tim2000s
Sep 01 2016 13:00
@danamlewis owIt's a bit of a rename process to get the second device set up, but I'm testing the connection n
Eric Jensen
@elnjensen
Sep 01 2016 13:08
@tim2000s If you wanted to try a rig with better range that can connect to Pi, but cheaper than TI stick, modmypi.com still has some Slice of Radio boards in stock. The manufacturer of them is unfortunately going out of business, so not too many more chances to get them. And it sounds like some good Edison boards are on the way in the next few months, too.
Colin Lennon
@colinlennon
Sep 01 2016 13:08
What model of Pi are you using, @tim2000s ?
Tim Street
@tim2000s
Sep 01 2016 13:09
RPi3
Colin Lennon
@colinlennon
Sep 01 2016 13:10
Eric's suggestion is a good one then, it's easy to just plug in a slice of radio. I've moved onto using my Edison rig 24/7 now and have a couple of Slice of Radio chips which are already flashed. Would be happy to stick one in the post @tim2000s @elnjensen
Eric Jensen
@elnjensen
Sep 01 2016 13:14
Thanks @colinlennon - I just ordered and mine is on the way in the mail now. I'm thinking I'll move the Slice of Radio to my bedside Pi (currently has TI Stick) so I can re-flash the TI stick for serial connection and move it to the Edison I'm working on . Wrestling with Wifi and Bluetooth issues on the Edison right now - seems like I can get wifi working (with wpa_supplicant, and connman disabled), or BLE tethering working (with connman on), but not both. Enabling connman for easy BLE tethering seems to mess up how wifi is working. Are you using both of these, and are you on yocto or ubilinux? I'm working with yocto as it seems to be the future supported OS.
Colin Lennon
@colinlennon
Sep 01 2016 13:16
@elnjensen I'm using Yocto and use connman for BT tethering. I don't really bother much with wifi as my Android phone connects to WiFi networks and shares that connection whenever possible, otherwise it shares 3G/4G (apparently iPhones don't do that).
I have noticed that connman doesn't automatically connect to a WiFi network when it is in range. I have a cron job which executes every minute, to forcefully re-establish the BT tethering connection if it is ever dropped
Eric Jensen
@elnjensen
Sep 01 2016 13:17
Yeah, the iPhone tethering is 4G only, disables wifi when tethering is on. Will concentrate on wifi for now - as long as I can connect at both work and home, then that covers most of my needs. Would be nice to get a seamless handoff from wifi to BLE working, though.
Colin Lennon
@colinlennon
Sep 01 2016 13:18
I think someone else mentioned that on the newest Yocto image (released June I think), connman is more reliable and behaves more like you'd expect it to.
I'm on the March 15th image I think
Tim Street
@tim2000s
Sep 01 2016 13:19
I've just ordered one thanks @colinlennon
Eric Jensen
@elnjensen
Sep 01 2016 13:20
OK - will keep working with it, since I have the latest image. The instructions on the mmeowlink wiki page for getting BLE tethering working worked seamlessly, and supposedly wifi works with connman, too, but I haven't yet figured out how - wifi won't work when connman is on. I'm sure it's possible, just need to figure it out. Documentation is pretty sparse for connman, much better for wpa_supplicant. (Though it seems that connman is somehow using wpa_supplicant, too?)
Colin Lennon
@colinlennon
Sep 01 2016 13:20
No worries @tim2000s . The improvement in range makes a real difference.
@elnjensen Oh, okay. Apologies for asking an obvious question - do WiFi networks show up at all in connman?
Tim Street
@tim2000s
Sep 01 2016 13:24
Okay - another daft question. When my loop is running, the enacted result sometimes includes the temp basal it's pulled from the pump (which exists in the file) and sometimes doesn't. Any idea why? is it a timing thing?
This is the output:
cat oref0-enacted/enacted-temp-basal.json
{
  "bg": 131, 
  "temp": "absolute", 
  "snoozeBG": 92, 
  "timestamp": "2016-09-01T14:20:57.403749", 
  "missing": [
    "duration", 
    "rate"
  ], 
  "reason": "Eventual BG 46<90, bolus snooze: eventual BG range 46-92, temp 0.8 ~ req 0.8U/hr", 
  "eventualBG": 46, 
  "error": "missing required input fields", 
  "input": {
    "reason": "Eventual BG 46<90, bolus snooze: eventual BG range 46-92, temp 0.8 ~ req 0.8U/hr", 
    "bg": 131, 
    "eventualBG": 46, 
    "temp": "absolute", 
    "snoozeBG": 92, 
    "mealAssist": "Off: Carbs: undefined Boluses: undefined ISF: 50.4, Target: 90 Deviation: 2 BGI: -1.66", 
    "tick": -1
  }, 
  "mealAssist": "Off: Carbs: undefined Boluses: undefined ISF: 50.4, Target: 90 Deviation: 2 BGI: -1.66", 
  "tick": -1
Eric Jensen
@elnjensen
Sep 01 2016 13:33
@colinlennon Yes, it sees the wifi network - but I get some errors connecting. So hardware is powered up, but something else is funky. Can't remember errors right now - need to head to work now, but will try more tonight and see if I can document it better. Thanks for the help!
Colin Lennon
@colinlennon
Sep 01 2016 13:34
@elnjensen :+1:
Scott Leibrand
@scottleibrand
Sep 01 2016 15:18
@tim2000s it won't set another identical temp if one is already running with enough time left.
garykidd
@garykidd
Sep 01 2016 15:37
@scottleibrand Thanks for feedback. I'm going to try to use iter_glucose for now....thinking it wont work out of the gate but hopeful someone here may be able to help me.
Scott Leibrand
@scottleibrand
Sep 01 2016 15:38
I think there's only one new feature that breaks with iter_glucose output.
All of the old stuff was built when it was the only thing available.
Tim Street
@tim2000s
Sep 01 2016 15:56
@scottleibrand What's the tolerance on that? I had one this afternoon where the active was 0.28 and the suggested was 0.24, and it left the 0.28 in place.
Scott Leibrand
@scottleibrand
Sep 01 2016 15:57
.1u/hr in master, 20% in dev
Tim Street
@tim2000s
Sep 01 2016 15:59
Okay - makes sense then.
Brandon Faloona
@bfaloona
Sep 01 2016 18:19
@scottleibrand @tim2000s I'm trying to parse the tolerance statements above.
New temp basal will be set if it differs from the current temp basal by .1u/hr in master branch, and by 20% in dev branch.
Did I get that right?
Brandon Faloona
@bfaloona
Sep 01 2016 18:56

@jasoncalabrese Your comment in a Mar 30 PR seems to relate to the feature I'm suggesting.

Also I'm wondering if we should have an option to pick the IOB source instead of doing the auto detect, maybe that's for later.

If I understand what you are saying, the IOB source could be optionally configurable, as opposed to the current behavior of flexible.

nightscout/cgm-remote-monitor#1560
Tim Street
@tim2000s
Sep 01 2016 19:03
@bfaloona that's my reading, yes.
David
@davidkeddydb
Sep 01 2016 20:12
ok, I am working through the code to set up. I have come to the pump-history and requisite report. (I really wish I studied Linux). I understand that I am trying to get the pump-history-raw over from the pump into the monitor. Should I be writing something like the following or am I way off:
openaps device add tz pumphistory
git add pumphistory.ini
openaps report add monitor/pumphistory-zoned.json JSON tz pumphistory monitor/pumphistory.json
garykidd
@garykidd
Sep 01 2016 21:36
so simple question....in trying to configure openaps and following guide.... the rtd at one point indicates openaps report add settings/model.json JSON pump model followed by openaps report invoke model.json
just was plugging along and cutting and pasting but....when I got an error with second command...was wondering why it shouldn't be openaps report invoke settings/model.json ...which btw I did run and it did work?
mariano42
@mariano42
Sep 01 2016 22:32
has anyone seen a dexcom transmitter mini usb port crap out? My rig stopped working yesterday and it seems like I am unable to get it to connect to my G4 now. I get AttributeError: 'none type' for any command.. for example openaps use dex GetFirmwareHeader. Or if I do openaps use dex scan it returns "". I assume that means it can't connect but the transmitter is showing that it is getting power from the usb.
my preflights pass too... so that seems odd to me
Eric Luhrs
@eluhrs
Sep 01 2016 22:36
I had the transmitter USB crap out once. Dexcom replaced it for no charge. I got lucky though because I was less than a week from my warranty running out. I hope you have the same (or better) luck.
Yes, that does seem odd. Even impossible I would think.
mariano42
@mariano42
Sep 01 2016 22:37
@eluhrs thanks.. this is only a month old, yea I don't understand the preflight thing at all
Eric Luhrs
@eluhrs
Sep 01 2016 22:39
Mine still worked with OpenAPS but it wouldn't take a charge. I don't think there were any errors.
mariano42
@mariano42
Sep 01 2016 22:48
@eluhrs yea I am watching the screen as its plugged in now and it looks like it loses usb connection then reestablishes it way too often. I do have a usb power cycle in my cron when it fails, but its happening much more often than every 5 mins
Brandon Faloona
@bfaloona
Sep 01 2016 22:49
@garykidd ^ Yes, looks like you found a typo. Any chance you could submit a PR to the docs repo? I think the master branch is the correct target for typos.
Let me know if you have questions about that. I'm glad to help.
garykidd
@garykidd
Sep 01 2016 23:19
@eluhrs what's shaking man?
@bfaloona are you sure you supposed to PR on master or dev? and you're sure it's a typo?
Eric Luhrs
@eluhrs
Sep 01 2016 23:24
@garykidd Not much. Thanks for asking.
Brandon Faloona
@bfaloona
Sep 01 2016 23:25
If you are unsure, ask for review in the PR. Sounds like you tested it though! :smile:
For docs, small changes should go in master. At least I get that impression from @danamlewis
garykidd
@garykidd
Sep 01 2016 23:32
@eluhrs I actually got last part today for edison rig, the connectors. I think I have everything I need now. Unfortunately, I still have a lot of work to do to get through the docs. Progress has been pretty painful. Goal is to get test rig up on RP3 and then hopefully it's a much easier process to move instance over to edison.