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

16th
Jan 2018
Thacher
@thacherT1D
Jan 16 2018 00:29
soo... intel edisons aren't being made anymore... what's the next best thing if I want to build more rigs to test on/have for back up?
Carl
@c-robertson
Jan 16 2018 00:47
Getting 'No space left on device' which prevents retrieving updated CGM data. I started removing older files from /var/logs. Anywhere else to look for things to remove? I recall this being discussed before and references to code files that can be whacked...
Also, is it generally safe to remove log files other than the most current? For instance, I have a syslog.1 that is large (already removed all the gzip'd files)
Dave Acklam
@dcacklam
Jan 16 2018 00:54
@c-robertson Anything in /var/log that has a dot-number after it can be erased
You can also configure logrotate (google it) to rotate and eliminate logs before you run out of space
all of the loop related logs are in /var/log/openaps/ if you want to save those
Carl
@c-robertson
Jan 16 2018 01:11
thx @dcacklam
Brian Cadre
@bcadre
Jan 16 2018 02:09
Successfully SMB-ing for a couple of days now. Noticed some things that I’m not sure are preferences-related or are simply how the rig is supposed to work:
1 - Entering carbs when low (BG 60) triggers an SMB response, despite the low BG. Is this normal? If so, rationale?
Brian Cadre
@bcadre
Jan 16 2018 02:35
2 - The rig treats finger stick-only BGs without sensor data being available. Daughter was in between sensors and checked her BG (207). Rig issued a SMB correction. Is this normal? What would the next data point have been? Would it keep issuing corrections until another finger stick?
3 - Any advice on minimizing the time in between reliable sensor insertions? Seems like we have a good 10-12hrs of “wonky” sensor data until sensor becomes reliable. Have considered pre-soaking but haven’t had good results. Advice/tips/tricks? Thanks in advance, everyone.
thomasnightscout
@thomasnightscout
Jan 16 2018 04:06
Hi guys! I'm having a hard time at the step "PREP STEP" for mac users to log into my rig, and changing it to root. Please advise!
Anthony Cerrone
@awcerrone_twitter
Jan 16 2018 05:56

I feel like I might be getting closer to being online again. Anyone have any advice to help me figure out where my issue is. Thanks.Listening for 40s silence before mmtuning: .No interfering pump comms detected from other rigs (this is a good thing!)
mmtune: "868.342", 5, -28 No wait required.
Unsuccessful oref0-pump-loop at Mon Jan 15 21:53:21 PST 2018

Starting oref0-pump-loop at Mon Jan 15 21:53:31 PST 2018 with 30 second wait_for_silence:
Waiting up to 4 minutes for new BG: ls: cannot access /tmp/pump_loop_completed: No such file or directory
Radio ok. Listening: .No interfering pump comms detected from other rigs (this is a good thing!)
Preflight OK. Old pumphistory-24h, waiting for 30 seconds of silence: Radio ok. Listening: .No interfering pump comms detected from other rigs (this is a good thing!)
Old pumphistory-24h refreshValueError: need more than 0 values to unpack
Couldn't refresh_old_pumphistory_24h
oref0-pump-loop failed. pump_loop_completed more than 15m old; waiting for 40s silence before mmtuning
Radio ok. Listening: .

Ebgineer
@Ebgineer
Jan 16 2018 07:55
I saw messages like this when the rig radio checked out ok but openaps was unable to retrieve pump history (event data) from the pump. If this were successful it would instead say Old pumphistory-24h refreshed Profile less than 60m old; Profile valid. Then it would go on to verify that the pump clock matches current time.
Scott Leibrand
@scottleibrand
Jan 16 2018 09:35
@bcadre what is your target?
What results have you had with soaking new sensors?
With a reasonable target (~100) SMB will be disabled below about 70. You can also enter rescue carbs with an IFTTT trigger that sets a high temp target at the same time, to disable SMB while they kick in.
Brian Cadre
@DrCadre_twitter
Jan 16 2018 14:37
@scottleibrand her targets are 90-130. Results mixed with pre-soaking sensors. Does SMB disable below a certain % of target or an absolute number? She was obviously below (60s) and it still issued boluses when she entered the carb correction. So, for now, we are correcting unannounced. I will enable the IFTTT with the high temp target to help. Additionally, is it normal for SMB to enact based solely on a fingerstick BG (i.e. without sensor data to follow)? Thanks again for your input.
Scott Leibrand
@scottleibrand
Jan 16 2018 14:44
@bcadre when current or minimum future BG is below threshold, SMB is disabled. Threshold is defined based on target, not as a straight percentage, but similar. If you’re using 90-130 for targets, you might want to narrow that to a single target number. By default, oref0 now only uses the lower number for the target, so you’ll want to raise the 90 to at least 100, which will prevent SMBs below about 70 mg/dL.
Brian Cadre
@DrCadre_twitter
Jan 16 2018 15:01
@scottleibrand thanks for the advice. I’ll change her target to a single number. Appreciate the advice.
Dana Lewis
@danamlewis
Jan 16 2018 15:14
  • I would consider bumping it up from 90, especially when experimenting with new features
#peanutgallery 😊
Tore Bjørndalen
@BjoTore_twitter
Jan 16 2018 15:17
I think there is a NS update bug when you enter a bolus right before you do a pump suspend. The rig updates the IOB correctly, but it never makes it into NS. Discovered this twice now.
latest dev
Dana Lewis
@danamlewis
Jan 16 2018 15:52
In the OpenAPS pill; or rendering the bolus on the graph?
It's a known issue that NS is sometimes slow to render (display) some boluses. That doesn't impact looping, though.
Tore Bjørndalen
@ToreBj
Jan 16 2018 16:46
The pill shows it correctly, but the bolus is not registered in NS(and not showing in the graph) Maybe this affects autotune? A 5U bolus is not registered, in my case, as treatment in NS(when I make a treatment report) I guess this is retrieved from pump history. Glitch when in suspend?
SMBs before and after are registered correctly
Dana Lewis
@danamlewis
Jan 16 2018 16:49
Only if you ran Autotune right then. It just sounds like there's a temporary delay in rendering on the graph (say 45 minutes).
Tore Bjørndalen
@ToreBj
Jan 16 2018 16:53
well it's 4 hours since this happened now, still not showing. Is there any consistency checking between NS and pump history?
LinnBee
@petrichordb
Jan 16 2018 21:14
Open APS - NS is getting sensor data. Rig is not looping all day today. Overnight last night, I switched sensor off as had not been soaked so no data for 12 hours or so. Cannot get it to Loop. New batteries. Rig sitting on top of pump. Pump hasn’t shown circle all day despite restarts. Now I have logged in it says data retrieval interrupted all the time. Pill says unknown. My son change clock by 1 min this morning on pump. Error looping says "month must be in 1-12". Looks ok on pump. Have done multiple reboots of rig and double checked pump date/time.
tzachi-dar
@tzachi-dar
Jan 16 2018 23:18
Does anyone know how to fix a pump that does not recognize the insulin?
I got a "new" pump today, 554. Put a battery inside and obviously it say "no insulin"
So I put a canola with insulin, but when I try to fill things it just throws all the insulin (like it does not fill that it has pressure any more).
If I stop it with my finger I get error 33.
Does anyone has a clue?