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

24th
Nov 2018
Martin Haeberli
@mhaeberli
Nov 24 2018 00:34
@danamlewis ok thx, later today
glenrob1940
@glenrob1940
Nov 24 2018 03:11
I am missing something obvious. Updated phone to 12.1 and nightscout quit updating. I have both loop and openAPS rigs, but currently are trying to get things back going with just loop.
glenrob1940
@glenrob1940
Nov 24 2018 03:17
Loop works okay, it updates and I can see pump related stuff in nightscout. However, the current basal reading in nightscout is stuck and the verbiage says '19 hours ago' (at the moment). Logs show no delta data present. My dexcom app tracks fine, loop tracks, but nightscout doesn't. I've deployed the dev nightscout branch a couple of times and also master from git, nothing seems to fix it. Anyone have an idea to try. I was using tokens, but this last troubleshooting has been without those fields filled in (in config vars on heroku).
The nightscout site is gtrobfcn.herokuapp.com
Once I get it working, I will kill this site and do a new one with tokens
Martin Haeberli
@mhaeberli
Nov 24 2018 03:50
@glenrob1940 any chance your mongoDB is full? (coincidentally around the time of your iPhone update)
that would explain the symptoms as you describe them
Martin Haeberli
@mhaeberli
Nov 24 2018 04:09
@danamlewis issue posted
Dana Lewis
@danamlewis
Nov 24 2018 04:09
Cool thanks. I meant @cluckj's error but that is good too
Related, feedback welcome on those issue templates if you have any @mhaeberli
I think your issue is the first since I threw those in
glenrob1940
@glenrob1940
Nov 24 2018 04:10
I will check that, I had tried dumping things from the heroku dashboard, but I will see if a more detailed cleaning will do the trick.
glenrob1940
@glenrob1940
Nov 24 2018 04:21
Size on disk is 16 MB. -- so that shouldn't be the problem - and I did run the repair command and the size on disk stayed at 16 MB.
glenrob1940
@glenrob1940
Nov 24 2018 04:32
Think, I'll 'sleep on this' and look at it again from a fresh perspective. This has been an interesting couple of weeks. Three weeks ago, I stayed in borderline ketoacidosis for several days with BG in the 325-440 range. Changed infusion sets, changed pumps, changed vials of Fiasp - twice. Finally went with a fresh vial of novolog, and got back in control. Best guess, is the Fiasp must have frozen (or got to warm) - I was near the end of a 90 day supply. Anyway, I stayed on a 523 (non loopable) pump until just before I started getting this nightscout problem (closed loop had worked just fine with humalog and then a new vial of fiasp in my 722 pump). In my various attempts cloning from git, my deployments to Heroku were to new named url's (I killed each of the previous ones after successful (kind of) deployments).
Martin Haeberli
@mhaeberli
Nov 24 2018 04:51
I recently tried fresh new named deployments to Heroku myself and they all failed. No idea why.
Dana Lewis
@danamlewis
Nov 24 2018 04:51
@glenrob1940 you might post in the Loop forums if that's what you're attempting to troubleshoot with
glenrob1940
@glenrob1940
Nov 24 2018 05:01
May do that tomorrow. I will also re-deploy to openAPS with my latest nightscout url in the morning to see what that shows. When I look at Heroku's logs, it shows time waiting on a successful openaps loop, but also shows the loopkit/loop communication is apparently successful and then one line that says essentially no delta data received. My guess is that there is a problem in communications between the nightscout/heroku/mongo server and Dexcom servers. Loop/iPhone is clearly communicating successfully with Dexcom servers.
Martin Haeberli
@mhaeberli
Nov 24 2018 05:30
@danamlewis props for the new issue templates
Dana Lewis
@danamlewis
Nov 24 2018 05:36
K thanks. Those can also be updated with PRs to the repo if anyone has suggestions to improve them
live4sw
@live4sw
Nov 24 2018 11:56
@glenrob1940 Re: fiasp (and off-topic, but just since no one else mentioned re your message), after about 5 months of use, I started having unexplained highs that were almost impossible to bring down with fiasp alone. Switched back to humalog. I don't like prebolusing but at least the insulin works for me otherwise!
ArvoNiilo
@ArvoNiilo
Nov 24 2018 13:17
Re Fiasp: I had similar experiences last spring and returned to Humalog. Ok but after the meals the BGs were too high. Started in September to mix 1/3 Fiasp and 2/3 Humalog. This mixture seems optimal for me.
Scott Leibrand
@scottleibrand
Nov 24 2018 15:36

Lots of people have had success with mixing like that.

One other idea would be to dilute the Fiasp 50% - you’d need to use twice as much volume, so it’d be easiest to do if you have a 72x pump, but that should have the dual effect of reducing the niacinamide concentration while also increasing the volume of tissue and number of blood vessels that can transport the insulin, thereby potentially further improving the speed of the Fiasp.

glenrob1940
@glenrob1940
Nov 24 2018 15:38
@scottleibrand dilute as in distilled water?
Scott Leibrand
@scottleibrand
Nov 24 2018 15:40
No, there is a special diluent that is sold for diluting insulin for small children.
It is basically the same solution your insulin is in, but without all the preservatives and excipients.
glenrob1940
@glenrob1940
Nov 24 2018 15:41
@ArvoNiilo - how did you convince Insurance to pay for 2 analogs?
@scottleibrand Okay I'll investigate obtaining the special diluent, thanks
Scott Leibrand
@scottleibrand
Nov 24 2018 15:42
AFAIK no one has tried diluting Fiasp before for purposes of making it faster, so you’d want to keep very close notes on how well it works, and share your results, if you try that.
glenrob1940
@glenrob1940
Nov 24 2018 15:43
Will do, I also have an endo appt this coming week and I'll run this by her before doing it
Scott Leibrand
@scottleibrand
Nov 24 2018 15:43
Many people do their mixing using their existing supply of Novolog/Novorapid or Humalog, so insurance is only buying one insulin at a time.
Mixing two insulins is a well-tested technique in the DIY community. Dilution is also done more broadly, but usually only for little kids who need less than the minimum amount of insulin a pump can deliver.
glenrob1940
@glenrob1940
Nov 24 2018 15:44
Okay, thanks
Scott Leibrand
@scottleibrand
Nov 24 2018 16:00
I’ll be interested what your endo recommends. :smile:
Martin Haeberli
@mhaeberli
Nov 24 2018 16:50
does anyone know of a tool to check the syntax of wpa_supplicant.conf (or where to see parsing errors in the rig logs)?
ArvoNiilo
@ArvoNiilo
Nov 24 2018 16:55
In Finland there no problem. I have two prescriptions (one for Fiasp, another for Humalon) and the national health insurance pays for the purchases (I have to pay yearly 50 € plus 3.50 € for each of the purchases and they cover the rest). I guess that if the added consumption of the two insulins would be much more than the total amounts per day prescribed for each of them separately they might ask for an explanation.
David
@david-t1
Nov 24 2018 17:35
Anyone have thoughts why my bolus snooze is not activating? .My preferences.json has "bolussnooze_dia_divisor": 1.5," and I am using the rapid-acting curve on a 5hr DIA. SMB is not activated.
Cas Eliëns
@cascer1
Nov 24 2018 17:45
Hey, I just reflashed my pi and am trying to re-install. However, after running oref0-setup I eventually run into this error
added process://ns/nightscout/ns NIGHTSCOUT_HOST API_SECRET
/usr/local/bin/nightscout: line 338: json: command not found
close failed in file object destructor:
sys.excepthook is missing
lost sys.stderr
Traceback (most recent call last):
  File "/usr/local/bin/openaps-import", line 89, 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-import", line 77, in run
    candidates = json.load(args.input)
  File "/usr/lib/python2.7/json/__init__.py", line 291, in load
    **kw)
  File "/usr/lib/python2.7/json/__init__.py", line 339, in loads
    return _default_decoder.decode(s)
  File "/usr/lib/python2.7/json/decoder.py", line 364, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/lib/python2.7/json/decoder.py", line 382, in raw_decode
    raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded
Could not run nightscout autoconfigure-device-crud
I also had some other weird issues with the npm run global-install where it didn't work the first time around
and even after running npm run global-install it told me that oref0-setup didn't exist
(this is on dev, btw)
Dana Lewis
@danamlewis
Nov 24 2018 19:16
David - what version of oref0 are you running? Bolus snooze has been taken out of the last few versions
Martin Haeberli
@mhaeberli
Nov 24 2018 20:20
ok on my wpa_supplicant thread - how to find and fix one whose syntax is wrong - I at least found this hint: - log file is at: /var/log/wpa_supplicant.log (yes, the obvious location!)
David
@david-t1
Nov 24 2018 20:22
@danamlewis the latest version, so now I understand why. tx
Martin Haeberli
@mhaeberli
Nov 24 2018 20:24
this looks fun and relevant: http://www.berrylan.org/ - a tool for configuring WiFi addresses on headless RPis. might work for Edisons, too...
Martin Haeberli
@mhaeberli
Nov 24 2018 20:59
re wpa_supplicant.log - doesn’t seem to exist...
Dana Lewis
@danamlewis
Nov 24 2018 21:06
Ty @audiefile !
Rachel Sandlain
@audiefile
Nov 24 2018 21:07
your welcome @danamlewis ! I guess that means it worked lol
Dana Lewis
@danamlewis
Nov 24 2018 21:13
:+1: saw the PR come in
Brian Rabinovitz
@bdr1177
Nov 24 2018 22:12
How can I (1) reset the autotune settings to match the pump settings, and (2) do so without needing to wait for the next autotune run for SMBs to kick back in.
Dana Lewis
@danamlewis
Nov 24 2018 22:21
A remove b do a manual run
Or just do a new manual run and it'll pull pump settings
Brian Rabinovitz
@bdr1177
Nov 24 2018 22:34
Do I just do the manual run on the past day?
tzachi-dar
@tzachi-dar
Nov 24 2018 22:40
My pi, got totally disconnected from network. When I connect it to a monitor it only shows the screen for the first few seconds.
Any ideas on how to get an ssh back to it again?