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

25th
Dec 2017
Samuchco
@Samuchco
Dec 25 2017 03:09
I need help with autotune and autosense.
The BGs are going very high at around 12 am. The ISF in the pump is 35 at 12 am. autosense/autotune is changing it to 45, even though he is high. Any ideas?
Samuchco
@Samuchco
Dec 25 2017 03:17
here are the log results for papertrail filtered by pump-log-adjust starting yesterday at 11:30 PM. It shows that prior to 12 am the isf is 22 or so (the pump is set to ISF 25). After 12am (the pumps is set to 35), the ISF goes to 40 something.
Dec 23 11:30:34 SammysAPS pump-loop.log: Autosens ratio: 1.07; Adjusting basal from 1.6 to 1.75; ISF from 25 to 23.4; CR: 6 
Dec 23 11:42:17 SammysAPS pump-loop.log: Autosens ratio: 1.07; Adjusting basal from 1.6 to 1.75; ISF from 25 to 23.4; CR: 6 
Dec 23 11:45:21 SammysAPS pump-loop.log: Autosens ratio: 1.07; Adjusting basal from 1.6 to 1.75; ISF from 25 to 23.4; CR: 6 
Dec 23 at 11:30 PM
Dec 24 02:51:15 SammysAPS pump-loop.log: Autosens ratio: 1.02; Adjusting basal from 1.4 to 1.45; ISF from 43.3 to 42.4; CR: 6.329 

Dec 24 02:55:30 SammysAPS pump-loop.log: Autosens ratio: 1.03; Adjusting basal from 1.4 to 1.45; ISF from 43.3 to 42; CR: 6.329 
Dec 24 03:55:39 SammysAPS pump-loop.log: Autosens ratio: 1.05; Adjusting basal from 1.4 to 1.45; ISF from 43.3 to 41.2; CR: 6.329
I changed the settings on the pump to one ISF- 25 throughout the day. By changing in the pump, will autotune get the info from the pump? or do I have to change it on the rig too?
Samuchco
@Samuchco
Dec 25 2017 03:30
Here is the Autotune recommendation log
Dec 24 00:11:42 SammysAPS autotune.log: p50deviation: 0.39 p50BGI -2.05 p50ratios: 0.854 Old ISF: 44.552 fullNewISF: 38.047 adjustedISF: 38.047 newISF: 43.251 
Dec 24 00:11:42 SammysAPS autotune.log: Autotune pump profile recommendations: 
Dec 24 00:11:42 SammysAPS autotune.log: --------------------------------------------------------- 
Dec 24 00:11:42 SammysAPS autotune.log: Recommendations Log File: /root/myopenaps/autotune/autotune_recommendations.log 
Dec 24 00:11:42 SammysAPS 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 
Dec 24 00:11:45 SammysAPS autotune.log: Parameter      | Pump     | Autotune  
Dec 24 00:11:45 SammysAPS autotune.log: ------------------------------------- 
Dec 24 00:11:45 SammysAPS autotune.log: ISF [mg/dL/U]  | 35.000   | 43.251    
Dec 24 00:11:45 SammysAPS autotune.log: Carb Ratio[g/U]| 8.000    | 6.329     
Dec 24 00:11:45 SammysAPS autotune.log: Basals [U/hr]  | -        | 
Dec 24 00:11:45 SammysAPS autotune.log:   00:00        | 1.700    | 1.230     
Dec 24 00:11:45 SammysAPS autotune.log:   01:00        |          | 1.367     
Dec 24 00:11:45 SammysAPS autotune.log:   02:00        |          | 1.398     
Dec 24 00:11:45 SammysAPS autotune.log:   03:00        | 1.600    | 1.381     
Dec 24 00:11:45 SammysAPS autotune.log:   04:00        |          | 1.323     
Dec 24 00:11:45 SammysAPS autotune.log:   05:00        |          | 1.323     
Dec 24 00:11:45 SammysAPS autotune.log:   06:00        |          | 1.292     
Dec 24 00:11:45 SammysAPS autotune.log:   07:00        |          | 1.237     
Dec 24 00:11:45 SammysAPS autotune.log:   08:00        | 1.450    | 1.326     
Dec 24 00:11:45 SammysAPS autotune.log:   09:00        |          | 1.460     
Dec 24 00:11:45 SammysAPS autotune.log:   10:00        |          | 1.603     
Dec 24 00:11:45 SammysAPS autotune.log:   11:00        |          | 1.659     
Dec 24 00:11:45 SammysAPS autotune.log:   12:00        |          | 1.624     
Dec 24 00:11:45 SammysAPS autotune.log:   13:00        | 1.550    | 1.500     
Dec 24 00:11:45 SammysAPS autotune.log:   14:00        |          | 1.445     
Dec 24 00:11:45 SammysAPS autotune.log:   15:00        |          | 1.433     
Dec 24 00:11:45 SammysAPS autotune.log:   16:00        |          | 1.431     
Dec 24 00:11:45 SammysAPS autotune.log:   17:00        |          | 1.409     
Dec 24 00:11:45 SammysAPS autotune.log:   18:00        |          | 1.362     
Dec 24 00:11:45 SammysAPS autotune.log:   19:00        |          | 1.360     
Dec 24 00:11:45 SammysAPS autotune.log:   20:00        |          | 1.393     
Dec 24 00:11:45 SammysAPS autotune.log:   21:00        | 1.750    | 1.414     
Dec 24 00:11:45 SammysAPS autotune.log:   22:00        |          | 1.339     
Dec 24 00:11:45 SammysAPS autotune.log:   23:00        |          | 1.253
Dana Lewis
@danamlewis
Dec 25 2017 03:39
autotune won't pick up the new pump settings to tune from until it's automated overnight run. so if you want it to reset sooner, you can a) manually run autoutne on the rig after you change pump; and/or b) delete the autotune stuff, so you're using the pump settings until new autotune is run off the pump profile overnight
it looks like your baseline basals are too high, and it's trying to drop them down, then compensate more with ISF
so also changing your underlying pump basals may help so ISF doesn't have to be tuned so far
Samuchco
@Samuchco
Dec 25 2017 03:45
@danamlewis Thanks Dana, but with the autotune settings (i.e. basals and ISF), he is going up !!
Can I set a new Basal Profile (Profile A) in the pump and test the recommended autotune adjustments
Scott Leibrand
@scottleibrand
Dec 25 2017 03:47
I would delete the autotune files that were calculated when you had multiple ISFs and re-run autotune with the single one.
Dana Lewis
@danamlewis
Dec 25 2017 03:47
^ that
yes, you could make profile A the autotune recommended profile, but I would clear it out and re-run it with the new settings before you try anything else
Samuchco
@Samuchco
Dec 25 2017 03:49
@danamlewis what do I delete? rm autotune?
Scott Leibrand
@scottleibrand
Dec 25 2017 03:50
cp settings/pumpprofile.json autotune/profile.json && cp settings/pumpprofile.json settings/autotune.json will do it I think
Samuchco
@Samuchco
Dec 25 2017 03:51
@danamlewis @scottleibrand Thanks, but with the autotune settings (i.e. basals and ISF), he is going up !!
Can I set a new Basal Profile (Profile A) in the pump and test the recommended autotune adjustments "
Will the rig collect the new basals on Profile A from the pump?
Scott Leibrand
@scottleibrand
Dec 25 2017 03:56
oref0 pulls whichever pump profile is active
Samuchco
@Samuchco
Dec 25 2017 03:56
Ok Thanks
can you suggest which changes to do, based on ^^
Scott Leibrand
@scottleibrand
Dec 25 2017 04:00
I would want to re-run autotune on your existing data, using a single ISF, before making any recommendations
I just sped autotune way up if you'd like me to give it a try on your data: just PM me your NS URL. :)
Jieseldeep
@Jieseldeep
Dec 25 2017 08:20

I'm getting an error when I'm trying to setup wifi over again. It gives me:

cp: error writing ‘interfaces.1514189987.bak’: No space left on device
cp: failed to extend ‘interfaces.1514189987.bak’: No space left on device

anyone have any ideas what to do from here?
Jieseldeep
@Jieseldeep
Dec 25 2017 08:39
Removing any existing git
Removed any existing git
Checking openaps 0.2.1 installation with --nogit support
openaps 0.2.1-dev
Checking /root/myopenaps: mkdir: cannot create directory ‘/root/myopenaps’: File exists
/usr/local/bin/nogit-openaps-init: line 9: echo: write error: No space left on device
/usr/local/bin/nogit-openaps-init: line 10: echo: write error: No space left on device
Can't init /root/myopenaps
That's the message when I run oref0-setup
alternateal
@alternateal
Dec 25 2017 10:19
@Jieseldeep - at the command prompt type df, that will show you the amount of space left on the device. I usually end up going to logs and removing all the zipped logs to create space, <cd /var/log/openaps && rm *.gz> this should create space for you
tepidjuice
@tepidjuice
Dec 25 2017 10:51
I'm consistently, every 20 mins or so, getting future entries in NS, uploaded from my rig which means the basal isn't rendered. Anyone know what might be happening?
Samuchco
@Samuchco
Dec 25 2017 15:25
@Jieseldeep I cleaned my cache by running sudo apt-get clean
Dana Lewis
@danamlewis
Dec 25 2017 16:12
@tepidjuice means rig pump CGM or something has the wrong time
Scott Leibrand
@scottleibrand
Dec 25 2017 16:20
Other possibility is NS is in the wrong TZ.
peterleimbach
@peterleimbach
Dec 25 2017 16:24
Last week I used the autotune output of around 11 days input from a Nightscout site with carbs and bolus to calculate a new basal profile.
Overall it is not a big thing but some shifts in the day profile and I gave it a try. I monitor the change carefully. If the sensor is not dying and the calibration shows no error at the Moment ist seems to work very well.
My question now is if I run autotune again (with changed profile files) from which day can I start from today as I changed the basal rate or from start of collecting the data when this basal rate was not active?
Dana Lewis
@danamlewis
Dec 25 2017 16:27
It doesn't matter what profile was active when
It tunes based on the current profile you give it at the time of the run
So you can run it on whatever time period you want
peterleimbach
@peterleimbach
Dec 25 2017 16:30
Is more Data better than less (If data quality is good)?
Scott Leibrand
@scottleibrand
Dec 25 2017 16:37
Up to a point. Beyond 30 days the old data gets almost completely drowned out by the more recent runs.
Or if things are steady, beyond about 14 days has diminishing returns.
Jeremy Cunningham
@jpcunningh
Dec 25 2017 18:46
On our second day of looping! About an hour ago, Nightscout stopped showing the temp basals. Any ideas?
NS Temp Basals Not Showing.PNG
I circled in red where the 1.2 temp basal should be showing and in the log where it issued it. I also verified that the pump is running the temp basal.
Although, I should be in the other group. Sorry! I'll post it in intend-to-bolus.
Dana Lewis
@danamlewis
Dec 25 2017 18:49
@jpcunningh you're in the right place!
Check for future data first; after that, mlab size. See http://openaps.readthedocs.io/en/latest/docs/Troubleshooting/Rig-NS-communications-troubleshooting.html for details
Jeremy Cunningham
@jpcunningh
Dec 25 2017 18:51
ok. looking now. Thanks @danamlewis !
Jeremy Cunningham
@jpcunningh
Dec 25 2017 18:57
No future treatments and mlab size is 6.55 MB and size of disk is 16.00 MB
future treatments.PNG
mlab size.PNG
it just self resolved...
Dana Lewis
@danamlewis
Dec 25 2017 19:03
:+1:
Jeremy Cunningham
@jpcunningh
Dec 25 2017 19:03

Does this possibly indicate the issue:

Dec 25 12:32:19 CaptainCAPP ns-loop.log: ns://JSON/shell/settings/temptargets.json
Dec 25 12:32:19 CaptainCAPP ns-loop.log: No JSON object could be decoded

It's below the enact/smb-enact.json line that indicates the 1.2 U/hr bolus...

Brian Cadre
@DrCadre_twitter
Dec 25 2017 19:59
Read the docs thoroughly. Looping for 5 days now, oref0. Good results but having trouble understanding bolussnooze. Her NS OpenAPS pill states 1.6u bolussnooze,0.1u basal. What is significance of bolussnooze insulin? I thought bolussnooze was a timeframe during which the rig wouldn’t low-temp but it’s listing it as a quantity of insulin. Please help me understand. Thanks all.
Dana Lewis
@danamlewis
Dec 25 2017 20:01
Are you still on 0.5.5? You may want to update to 0.6.0
Brian Cadre
@DrCadre_twitter
Dec 25 2017 20:05
0.6.0.
Scott Leibrand
@scottleibrand
Dec 25 2017 20:51
Then your NS is out of date.
That should read “bolus IOB” instead of “bolussnooze”. You can mentally translate it, or upgrade your NS to fix.
In oref0 0.6.x there is no longer any bolus snooze.
Brian Cadre
@DrCadre_twitter
Dec 25 2017 21:08
@scottleibrand and @danamlewis thanks for the guidance. Merry Christmas!
Dana Lewis
@danamlewis
Dec 25 2017 21:19
Same to you! 🎄
Dave Acklam
@dcacklam
Dec 25 2017 22:52
So... Merry Christmas
Good: I got a pi zero, pi-UPS-hat (lipo charge/power-switcher) & slice-of-radio for Christmas. Bad? Both of my explorer boards died (One can't CCPROG anymore, the other one has continuous subg/spidev errors) & I've been doing the pi/carelink thing since (until I get the SRF set up)...
Notes: Nightscout upload seems half-broken with the pi/carelink (I get basal rates & pump activities (bolus, etc) uploaded, but none of the pills work (OpenAPS 'unknown', IOB blank, pump pill blank) setup...
@bewest You told me to message you if I ended up using an SRF/Slice-of-Radio... What do I need to do to CCprog that, besides use a different firmware?
Jieseldeep
@Jieseldeep
Dec 25 2017 22:59
Hi all, back to working on this. Still have a full disk issue and didn't find anything in the troubleshooting section. Am I just not searching for the right terms?
and Merry Christmas!
Scott Leibrand
@scottleibrand
Dec 25 2017 23:00
@dcacklam you probably need to figure out the pinouts to ccprog it
Jeremy Cunningham
@jpcunningh
Dec 25 2017 23:01
Jieseldeep
@Jieseldeep
Dec 25 2017 23:01
Thanks guys!
Jeremy Cunningham
@jpcunningh
Dec 25 2017 23:01
:smile:
Jieseldeep
@Jieseldeep
Dec 25 2017 23:02
So basically, don't search for disk full, search for disk space ;)
Dave Acklam
@dcacklam
Dec 25 2017 23:02
@scottleibrand Would those be the same pinouts needed to JTAG it?
@Jieseldeep
Disk full on your rig?
Scott Leibrand
@scottleibrand
Dec 25 2017 23:03
@Jieseldeep please PR in an additional sentence or two that would have allowed you to find it with your search
Dave Acklam
@dcacklam
Dec 25 2017 23:03
How much stuff is in /var/log/openaps
Dana Lewis
@danamlewis
Dec 25 2017 23:04
@dcacklam might also be some old helpful stuff in the mmeowlink wiki - I would check there too
kallnap
@kallnap
Dec 25 2017 23:11
The wiki that dana mentioned is great. But how did you get a slice of radio nowadays?
Boris and Kayley Raskin
@boryanvbu
Dec 25 2017 23:14
Hi guys, been looping for over a year now (raspberry pi —> Edison) and would now like to start using oref1 SMB features. It still doesn’t work after upgrading to oref0 0.6.0 and changed the preferences options enableSMB_with_bolus, enableSMB_with_COB, enableSMB_with_temptarget, enableUAM to true. When checking the log, it still says “Starting oref0-pump-loop”. Any idea how to get the smb pump loop running so that it says “Starting supermicrobolus pump-loop”?
Jieseldeep
@Jieseldeep
Dec 25 2017 23:16
Eureka! That did it! Thanks @jpcunningh and @scottleibrand !
Dana Lewis
@danamlewis
Dec 25 2017 23:17
@boryanvbu they all say oref0-pump-loop now
(In 0.6.0 and beyond)
Boris and Kayley Raskin
@boryanvbu
Dec 25 2017 23:20
@danamlewis Oh, didn’t realize that. It still doesn’t do the smb when I check in Nightscout. I still have to bolus manually.
Dana Lewis
@danamlewis
Dec 25 2017 23:21
@boryanvbu if you use the bolus wizard, that disables SMB's for 6 hours afterward unless you turn on the a52 risk preference. But if you're not and you expect SMB's some other time - paste your logs from that time so we can take a look
Boris and Kayley Raskin
@boryanvbu
Dec 25 2017 23:23
@danamlewis that might be it! I don’t think I’ve gone more than 6hrs without using the bolus wizard. We’ll try the a52 risk preference. Thanks!
Dave Acklam
@dcacklam
Dec 25 2017 23:24

@danamlewis

Is oref1 still active/useful? Or has it all been folded back into oref0

Scott Leibrand
@scottleibrand
Dec 25 2017 23:25
oref1 was never really a thing, just a way of talking about SMB as new and different. we've mostly stopped using the oref1 language since SMBs stopped being new and different
the repo with the code has always been oref0
Dave Acklam
@dcacklam
Dec 25 2017 23:35
Ok
@ceben80
As of now the x12 setup in oref0 works just fine for oref0

I've been running oref0 straight out of the repo with no tweaks as of late, successfully.

The things that need to be tweaked for the x12 are in the supporting packages (decocare/mmeowlink)

Of course, I also managed to get a 722 now... So... Major hardware update this Christmas...