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

15th
Sep 2017
Scott Leibrand
@scottleibrand
Sep 15 2017 00:01
@danimaniac check whether your pump has a temp basal running. If OpenAPS is still setting neutral temps, those will be based on your autotuned basals. If no temp is set, your pump basals are active.
dan lightner
@danimaniac
Sep 15 2017 00:53
OpenAPS is NOT setting neutral temps. Out of curiosity, in what circumstances will OpenAPS continue to set neutral temps? Certain pumps? Certain OpenAPS release? What are the determining factors?
Dana Lewis
@danamlewis
Sep 15 2017 01:03
All versions
Unless you changed that setting in prefeeences
But without CGM data you're not looping and setting neutral temps
garykidd
@garykidd
Sep 15 2017 01:07
small case worked pretty well for 2000mah battery and explorer.
Scott Leibrand
@scottleibrand
Sep 15 2017 01:22
If you have CGM data coming in from a plugged-in receiver but it's invalid (uncalibrated or ???s for example) the rig will set neutral temps. But not if there's no data whatsoever.
garykidd
@garykidd
Sep 15 2017 01:25
@scottleibrand was wondering if it might be beneficial to turn off auto-sensitivity when initially turning on autotune....or would autosensitivity not impact autotune....
Scott Leibrand
@scottleibrand
Sep 15 2017 01:28
They work independently. No need to turn either one off.
Autotune makes tiny changes each day, and then autosens adjusts from there.
garykidd
@garykidd
Sep 15 2017 01:32
right/ understood...just didnt know if ...intially autosense was tweaking figures being analyzed by autotune (assuming autotune was run later say at midnight---sounds like no
Scott Leibrand
@scottleibrand
Sep 15 2017 02:12
No, we specifically designed it to autotune based on what actually happened (relative to the previously autotune directory basals and ratios), not based on what autosens thought at the time.
thisistheusername
@thisistheusername
Sep 15 2017 03:31
Any 712/512 loopers around?
katie disimone
@Kdisimone
Sep 15 2017 03:47
I’ve set one up before, but not currently using one @thisistheusername
thisistheusername
@thisistheusername
Sep 15 2017 04:13
@Kdisimone Could you tell me what you did to get it going? I have followed all the steps on readthedocs for OpenAPS and for the 512/712 settings files that need to be created, along with the aliases, but I still can't things working.
katie disimone
@Kdisimone
Sep 15 2017 04:22
What's happening on yours?
What does cat ~/myopenaps/settings/settings.json show as a result?
Perhaps one of your files has an error in it?
thisistheusername
@thisistheusername
Sep 15 2017 04:45
glenrob1940
@glenrob1940
Sep 15 2017 05:36
Changed/upgraded a phone. In the process, had to reinstall apps including Dexcom Mobile and Nightscout. Dexcom Mobile is tracking glucose, but openaps - on the edison, and in Nightscout is never getting glucose readings - at the moment no data for 7 hours. I remember entering API secret somewhere in this flail and perhaps I entered something wrong there (phone obscures chars after a couple of seconds on entry) I don't think I made a mistake, however, I don't remember what app required that. - I have tried reinstalling apps, but never get asked for the API secret again. I've restarted dyno's at heroku, with no evident change - still never getting glucose data. Any suggestions of where to look/fix? Also during this evolution I rebuilt loop and installed it, so I'll fire off my rileylink to see how it is doing. I'd really like to get the edison going again, I appreciate smb/uam and the like.
katie disimone
@Kdisimone
Sep 15 2017 06:21

@thisistheusername your rf_enable line is set to false in the settings.json file. That's why it is not communicating properly. Change that to true and you should be good. That one definitely will keep you from looping on false.

Also not sure about your timeformat: 120. That's one I never played with.

@thisistheusername your insulin concentration is also set different. I'm not 100% sure whether or not OpenAPS actually uses that value though. Standard novolog or humalog is 100. Your setting has 50.
katie disimone
@Kdisimone
Sep 15 2017 06:32

@thisistheusername also not sure about the raw lines on the bg_targets file. I think you probably were using your 751's file as a template?

Those probably aren't causing a problem...but if you notice anything wonky after the rf_enable is fixed...I'd maybe change that raw stuff to match as well (but again, I don't think that will be a problem, just different pump models have that raw stuff when it pulls from pump).

glenrob1940
@glenrob1940
Sep 15 2017 06:39
Figured out the problem. Forgot to set up dexcom follow, Forgot that on the old phone, I had set up a redundant share to myself which apparently sets something up in the Dexcom mobile app to allow sharing in general.
garykidd
@garykidd
Sep 15 2017 10:35
where's the latest jubilinux version?
image?
@scottleibrand thanks...
Scott Leibrand
@scottleibrand
Sep 15 2017 10:37
We still recommend jubilinux 0.1.1, not the latest. See the docs.
0.2.0 should work fine with oref0 0.5.3, though, if you you've already flashed it.
garykidd
@garykidd
Sep 15 2017 10:43
thx. I actually had just decided to flash 0.1.1...what I had downloaded already..had just lost the link. I think a PR may be in order for adding a link...will review later.
Darryl Schick
@darrylschick
Sep 15 2017 12:20
Mine came with a sheet with "Linux ubilinux 3.10.17-poky-edison+ Debian GNU/Linmux 8" - It is the pre-flashed version from hamshield. It is working (or at least was, until it started freezing constantly about a week ago). Should I re-flash it?
Tried new cables, charger, still freezing up every few hours. It never (never) froze before we implemented BT tethering so I'd like to undo that. Is it sufficent to re-run the setup string below removing the btmac term, or do I need to do a complete reinstall? The docs explain how to enable BT but not how to disable... '# To run again with these same options, use: oref0-setup --dir=/root/myopenaps --serial=[redacted] --cgm=G4-upload --ns-host=https://[redacted].herokuapp.com --api-secret=[redacted] --tty=/dev/spidev5.1 --max_iob=3 --enable=' autosens autotune meal ' --radio_locale='US' --btmac='[redacted]'
katie disimone
@Kdisimone
Sep 15 2017 13:10
@darrylschick yes, you would just edit out the BT mac part of the runagain line. When your cron rebuilds, it will not include the line for BT tethering then. You can also go into bluetoothctl and use remove AA:BB:CC where the abc part is your BT mac. If you really want to end the pairing even.
Darryl Schick
@darrylschick
Sep 15 2017 13:10
Thanks Katie!
katie disimone
@Kdisimone
Sep 15 2017 13:30
Lol, @darrylschick just realized I didn't say to run the runagain after editing it...hopefully that was implied 🤣
thisistheusername
@thisistheusername
Sep 15 2017 14:02
@Kdisimone I know I changed those values in the settings file and also removed the raw bg values in the other. Anyway....I posted the latest pump_loop log stanza on the GG post I have for tracing this out: https://groups.google.com/d/msg/openaps-dev/ToFrTvhOxT8/NV6ikLN9EAAJ.
Tore Bjørndalen
@ToreBj
Sep 15 2017 14:32
I just discovered when checking autotune in papertrail, following each run I get: autotune.log: /usr/local/bin/oref0-autotune-recommends-report: line 22: warning: setlocale: LC_NUMERIC: cannot change locale (en_US.UTF-8): No such file or directory
Running dev installed aug 15
Craig Brenner
@cbrenner
Sep 15 2017 15:38

Ok, so I'm going to ask this one again since it didn't get a response in the evening, perhaps the morning will be better :smile:

Trying to get some guidance here:
(1) If my son is getting 1 to 1.5 hours of low/"0" temping after meals, if this indicative of his carb ratio being too aggressive for these meals or might there be something else we need to consider like carb absorption times being off?
(2) On the other end, if he underestimates his meal carbs and leaves it to the OpenAPS AMA to high temp him back to a normal range but ends up having too much IOB and goes low, what variables might need tweaking? Would this be the '"current_basal_safety_multiplier', pumps 'max basal rate', something else?

Dana Lewis
@danamlewis
Sep 15 2017 15:39
For 1, can we see a screenshot and logs? For 2) can we see screenshot and logs ? :smile:
There can be a couple of things going on in each of those, so the visual helps identify
garykidd
@garykidd
Sep 15 2017 15:50
Didn't get to check logs yet but any reason that's readily identifiable that UAM didnt kick in here......
[working on pasting in image] hmmm..
anyone else ever get error when trying to upload image?
Could not find a bored instance Error Uploading File Maybe firewall??
Craig Brenner
@cbrenner
Sep 15 2017 15:52
@danamlewis OK, I will try to pull some screen shots
David
@davidkeddydb
Sep 15 2017 16:08
Can anyone recommend a micro usb cable you have purchased in the past that has a small end with the micro usb? Trying to find one that I can use that will work easily with the rig case that Robert Silvers had put out. Some of mine I have to take the whole thing apart to get to the Explorer board vs being able to just plug into the micro usb port.
Craig Brenner
@cbrenner
Sep 15 2017 16:40
@danamlewis Ok, the first graphic and PDF of the pump log represents a recent low temping situation. The second graphic and log represents a recent high temping situation
Too Much Low Temping.jpg
Low Temping Pump Log.pdf
Too Much High Temping.jpg
High Temping Pump log.pdf
katie disimone
@Kdisimone
Sep 15 2017 16:45
@cbrenner how wide is your target BG range? What's your DIA?
Craig Brenner
@cbrenner
Sep 15 2017 16:46
BG range is 120 -110 and DIA is 4
fidoman
@efidoman
Sep 15 2017 18:00
Anyone know if there is a way to remove a few erroneous blood glucose entries in NightScout. My prototype g5 collector sent out a few readings that were totally bogus (> 1000) and it has really messed up my NightScout visualization.
David
@davidkeddydb
Sep 15 2017 18:06
see if you can delete it in reports, treatments in NS
Scott Leibrand
@scottleibrand
Sep 15 2017 18:11
@ToreBj I've noticed that warning as well. It seems to be cosmetic (doesn't really affect anything), so I haven't looked into it.
@efidoman if NS doesn't let you delete those entries, you'll need to go into mLab to do so directly
fidoman
@efidoman
Sep 15 2017 18:16
Thanks - I will do it from mLab
Dana Lewis
@danamlewis
Sep 15 2017 18:19
(ICYMI - wanted to make sure you had a chance to see what's coming!) https://twitter.com/danamlewis/status/908753105226829824
Jon Cluck
@cluckj
Sep 15 2017 18:33
nice!
Jon Cluck
@cluckj
Sep 15 2017 18:39
Also @danamlewis ping when you get a chance re: "other researchers interested in this" :)
Craig Brenner
@cbrenner
Sep 15 2017 19:09
@Kdisimone Any concerns with the BG range and DIA as it relates toe the low or high temping issue?
Dana Lewis
@danamlewis
Sep 15 2017 20:45
@cluckj please email me for a longer lasting ping :D :)
Martin Haeberli
@mhaeberli
Sep 15 2017 20:46
```
Getting log entries via papertrail:
Sep 15 13:44:11 myopenaps autosens-loop.log: Fri Sep 15 13:44:03 PDT 2017
Sep 15 13:44:11 myopenaps autosens-loop.log: settings/autosens.json
Sep 15 13:44:11 myopenaps cgm-loop.log: AttributeError: 'NoneType' object has no attribute 'iter_records'
Sep 15 13:44:11 myopenaps network.log: At Fri Sep 15 13:44:02 PDT 2017, my wifi network name is , and my public IP is: 166.137.242.17
Sep 15 13:44:11 myopenaps network.log: Finished oref0-online.
trying to understand the NoneType thing … - missing iter_records
running 0.5.3 master
Peter Miller
@ochenmiller
Sep 15 2017 21:04
@danamlewis , very excited to read about your work sponsored by RWJF!
Dana Lewis
@danamlewis
Sep 15 2017 21:16
@ochenmiller thanks! excited about the opportunity to bring some data science resources into the community :)