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

18th
Jun 2018
Zakm0n
@Zakm0n
Jun 18 2018 06:40
ok boys and girls, I got one hell of a conundrum for you. Fakemeter is crashing my rig again. I've yet to get it to work at all, but I just fresh rebooted, reflashed my radio, and got a successful mmtune
Then, I ran a manual fakemeter send and got no response to wakeup, when I try to mmtune again, i get the "Could not get subg_rfspy state or version." errors.
skubesch
@skubesch
Jun 18 2018 08:12
Okay I have two rigs giving the same Json value error. One is an Edison and the other a pi with a reformatted sd card backed up with a known working version. So I think the issue is my Nightscout. Not an MLab issue and I deleted and compressed anyway. Other things to look at please?
( Could be an MLab issues but not a size issue )
Martin Haeberli
@mhaeberli
Jun 18 2018 08:38
Cn u post some excerpt s?
Martin Haeberli
@mhaeberli
Jun 18 2018 09:05
SMB somehow turned off after master/0.6.2 install, even though I copied my old preferences in.
getting:
No smb_enact needed. Temp refreshed: monitor/temp_basal.json: {"duration":27,"rate":4.95,"temp":"absolute"}
SMB disabled (!microBolusAllowed)
and that smb-suggested.json is less than 1m old
in logs...
tried a fresh install in case I missed something in interaction, and in case a new preferences value is implicated, but somehow didn’t see anything obvious
Martin Haeberli
@mhaeberli
Jun 18 2018 09:32
profile.json reads:
"max_iob": 8,
"max_daily_safety_multiplier": 3,
"current_basal_safety_multiplier": 7,
"autosens_max": 1.2,
"autosens_min": 0.7,
"rewind_resets_autosens": true,
"adv_target_adjustments": false,
"exercise_mode": false,
"unsuspend_if_no_temp": true,
"enableSMB_with_COB": true,
"enableSMB_with_temptarget": true,
"enableUAM": true,
"curve": "rapid-acting",
"offline_hotspot": false,
"carbsReqThreshold": 1,
"high_temptarget_raises_sensitivity": true,
"low_temptarget_lowers_sensitivity": false,
"sensitivity_raises_target": false,
"resistance_lowers_target": false,
"autosens_adjust_targets": false,
"skip_neutral_temps": false,
"bolussnooze_dia_divisor": 2,
"min_5m_carbimpact": 8,
"carbratio_adjustmentratio": 1,
"maxCOB": 120,
"autotune_isf_adjustmentFraction": 0.5,
"wide_bg_target_range": false,
"A52_risk_enable": true,
"maxSMBBasalMinutes": 30,
"allowSMB_with_high_temptarget": true,
"enableSMB_always": true,
Martin Haeberli
@mhaeberli
Jun 18 2018 09:48
looks like I maybe broke autotune in the process ...
Martin Haeberli
@mhaeberli
Jun 18 2018 09:58
(I HAD had a custom openaps directory, but moved to the new default; in the process, my autotune data wasn’t moved over…)
Martin Haeberli
@mhaeberli
Jun 18 2018 10:34
ok, likely fixed now...
Andy Sharrow
@dramageek
Jun 18 2018 13:03
@mhaeberli Extra comma on the end of your preferences?
Martin Haeberli
@mhaeberli
Jun 18 2018 13:10
Turns out, no- that was not an exact quote... long story
thisistw
@thisistw
Jun 18 2018 13:52
Could someone offer some advice - been looking for a solution and so far haven't found it...
Im using Xdrip+ and its having a "REST-API problem: Unable to do REST API UPload: Internal Server Error url:** marking record: failed"
It was working until about 10hours ago. I have tried restarting Heroku dynos, and tried Mongodb - repair database. (size of which is currently 366.72Mb)
N3FM
@N3FM
Jun 18 2018 16:27
Hi all! Apologizing in advance for a long post. I have several 'issues' with my installation that are in the way of looping - and I'm blindly determined to plow through them!
I'm using a Medtronic 722 with Enlite Sensors/BG meter, Edison with Explorer, and mongo/Azure NS.
N3FM
@N3FM
Jun 18 2018 16:37

I will run fine for a while, then get hit by "Checking pump clock", and the clock will suddenly be 4 minutes off. That will force set_clock (in set_device_clocks.sh), which fails for the pump with either 'Bytearray index out of range', or 'Comms Exception: is the pump out of range?' errors; the attempt to set the CGM also fails, as 'UpdateTime' is not a valid command. I have forked the code, and commented out the offending chunks of code (ugly, but let's get it running first). And here's where my lack of experience with BitHub starts to get in the way. I can't for the life of me figure out how to save/ push the MODIFIED code to the rig.
Second, the rig is tuned to the frequency fairly well by now -- yet it almost immediately starts searching. And the Preflight starts failing. Sounds like mmtune is re-initializing and not paying attention to the prior 'best choice'. I have no idea how to trace that one at present.
3) By the time I regain control, of course the BG is too old... and I've lost 1/2 hour or more of loop time. (Reboot will 'help'...)
A couple of things would certainly help me at this point: A) Help me understand the steps in getting a modified file loaded and run. B) Is there a block or functional diagram for the modules? I have not learned how they link together. C) I'm expecting this is something I am doing wrong, as I would expect others to be bringing these issues up if it was a widespread issue. D) I thought somebody mentioned having a less aggressive approach to running mmtune... I'd appreciate some guidance!

I am SO psyched about this project, and appreciative of the team in place here. I had a successful overnight loop, so I know I'm close. Any suggestions?
N3FM (Bob)

and version 0.6.2
ElwinABC
@ElwinABC
Jun 18 2018 19:02
Hi all! Im trying to set up Autotune (without OpenAPS rig = phase C) with the step-by-step manual (http://openaps.readthedocs.io/en/latest/docs/Customize-Iterate/autotune.html) and the use of Google Cloud Platform. At Step 4: "Run autotune on retrospective data from Nightscout" I typed the following command but also
eversluis@compelling-pact-207418:~$ cd ~/myopenaps/settings
eversluis@compelling-pact-207418:~/myopenaps/settings$ oref0-autotune --dir=~/myopenaps --ns-host=https://elwin.herokuap.com --start-date=2018-06-15 –end-date=2018-06-16
-bash: oref0-autotune: command not found
Tried to Google to find a solution, but could not find it. Does anybody see what I'm doing wrong?
Scott Leibrand
@scottleibrand
Jun 18 2018 19:15
that means you didn't successfully install oref0 yet
the npm run global-install or npm install -g oref0 steps would be the ones that do that
ElwinABC
@ElwinABC
Jun 18 2018 19:22
many thanks!
marafie
@marafie
Jun 18 2018 23:18
Your instance of oref0 [0.7.0-dev, 0.7.0-dev] is out-of-date by 105 commits: you may want to consider updating.
You are currently running a development branch of oref0. Such branches change frequently.
Please read the latest PR notes and update with the latest commits to dev before reporting any issues.Completed oref0-pump-loop at Tue 19 Jun 02:15:47 +03 2018
Do I need to rerun installation ?
marafie
@marafie
Jun 18 2018 23:24
How long does the carb stay active ?

My kid ate 50 carbs 4 hours and 30 mins ago

And the COB now is 27

The Openaps was giving him high basel until 1 hour ago
His BG now is 4.2 , How can I prevent this from happening in the future
marafie
@marafie
Jun 18 2018 23:31
temp":"absolute","bg":76,"tick":-7,"eventualBG":128,"insulinReq":0,"reservoir":"135.2\n","deliverAt":"2018-06-18T23:20:56.884Z","sensitivityRatio":1,"COB":27,"IOB":0.352,"duration":120,"rate":0}
"COB: 27, Dev: -1.7, BGI: -0.4, ISF: 21.4, CR: 12.55, Target: 6.5, minPredBG 1.4, minGuardBG -3.9, IOBpredBG 2.2, COBpredBG 7.1, UAMpredBG 2.2; minGuardBG -3.9<4.4"