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

16th
Aug 2018
lyntonr
@lyntonr
Aug 16 2018 00:15
And I'm awake 10:08 am here is Australia ;). @scottleibrand I have updated to latest 0.7.0-dev and will use bolus wizard and watch out for missed pump history bolus's. Unfortunately this testing will be very limited as I only have one rig and pump and they are attached to me. So perhaps if others running dev that use bolus wizard can be encouraged to update and watch out for missed readings that may help, otherwise I not sure how this can be tested as its a rare occurrence anyway.
Oh BTW I went through git pull; npm global install and run-again to update, was all that needed or would a simple git pull and a reboot been enough ?
Scott Leibrand
@scottleibrand
Aug 16 2018 00:37
In this case just git pull, no need to even reboot.
garykidd
@garykidd
Aug 16 2018 00:39
using bolus wizard for carb entries now and ready liking it. haven't experienced missed readings on 723.
lyntonr
@lyntonr
Aug 16 2018 00:47
Hmmm i tried that but the pump log was still reporting 89 commits behind, even after cron stop, cron start hence i went through the whole thing
Scott Leibrand
@scottleibrand
Aug 16 2018 00:49
I think that “commits behind” report only updates periodically.
Zach Gohr
@zgohr
Aug 16 2018 00:49
@scottleibrand thanks a lot for that update. I’m not sure I’ll be able to test it until others confirm relative success. But if it works, moving back to bolus wizard would be great for me. It seems there is enough luck with ESC ESC ESC to protect from errors
lyntonr
@lyntonr
Aug 16 2018 00:50
Ok that explains it thanks. Will know for next time now
@zgohr The ESCs don’t work if you’re dialing up/down a bolus when they get sent. I would still avoid the Bolus Wizard when you can.
Zach Gohr
@zgohr
Aug 16 2018 00:54
drat. thanks for the warning
lyntonr
@lyntonr
Aug 16 2018 04:45
Screenshot (3).png
@scottleibrand so far so good the bolus wizard results are intact and not duplicated. However some info inside the notes does not appear correct, it may not matter but here is a screen shot in case it does, the end bit referencing targets of 5000 and 6000 seem wrong, and the glucose delta odd too, I think this is from the pump targets of 5.0 and 6.0 mmol/L
alimhassam
@alimhassam
Aug 16 2018 04:50
@lyntonr good catch i think the default unit returned by the go code for mmol user is "μmol/L"
lyntonr
@lyntonr
Aug 16 2018 04:51
is this cosmetic only or does it affect caluculations ?
alimhassam
@alimhassam
Aug 16 2018 04:52
it's just cosmetic
but we should probably do something to show it in mmol instead
lyntonr
@lyntonr
Aug 16 2018 04:53
yes although there are probaly bigger fish to fry :)
alimhassam
@alimhassam
Aug 16 2018 04:53
:)
Scott Leibrand
@scottleibrand
Aug 16 2018 04:59
What does pumphistory.json show for that record? And what does it look like in the NS API?
alimhassam
@alimhassam
Aug 16 2018 05:07
i have similar behavior, the bg_target_low and bg_target_high, and sensitivity are in the wrong unit μmol/L instead of mmol/L. They should be divided by 1000 to show correctly.
{
    "timestamp": "2018-08-15T13:39:57-04:00",
    "_type": "BolusWizard",
    "bg": 0,
    "carb_input": 31,
    "bg_target_low": 5400,
    "bg_target_high": 5900,
    "sensitivity": 2800,
    "carb_ratio": 7,
    "correction_estimate": 0,
    "food_estimate": 4.4,
    "unabsorbed_insulin_total": 15.4,
    "bolus_estimate": 4.4
  },
lyntonr
@lyntonr
Aug 16 2018 05:18
heres mine no mention of the hypothetical glucose delta though
  {
    "timestamp": "2018-08-16T12:58:10+10:00",
    "_type": "BolusWizard",
    "bg": 0,
    "carb_input": 35,
    "bg_target_low": 5000,
    "bg_target_high": 6000,
    "sensitivity": 2900,
    "carb_ratio": 6.8,
    "correction_estimate": 0.4,
    "food_estimate": 5.1,
    "unabsorbed_insulin_total": 3.3,
    "bolus_estimate": 5.1
  }
Aaron Michelson
@Aaybob
Aug 16 2018 05:20

are you seeing this error....just installed, not error just fills up the log I guess. ```
Completed oref0-pump-loop at Wed 15 Aug 22:02:29 PDT 2018
/root/src/openaps-menu/node_modules/i2c-bus/node_modules/bindings/bindings.js:88
throw e
^

Error: The module '/root/src/openaps-menu/node_modules/i2c-bus/build/Release/i2c.node'
was compiled against a different Node.js version using
NODE_MODULE_VERSION 46. This version of Node.js requires
NODE_MODULE_VERSION 57. Please try re-compiling or re-installing
the module (for instance, using npm rebuild or npm install).
at Object.Module._extensions..node (module.js:681:18)
at Module.load (module.js:565:32)
at tryModuleLoad (module.js:505:12)
at Function.Module._load (module.js:497:3)
at Module.require (module.js:596:17)
at require (internal/module.js:11:18)
at bindings (/root/src/openaps-menu/node_modules/i2c-bus/node_modules/bindings/bindings.js:81:44)
at Object.<anonymous> (/root/src/openaps-menu/node_modules/i2c-bus/i2c-bus.js:4:28)
at Module._compile (module.js:652:30)
at Object.Module._extensions..js (module.js:663:10)

Starting oref0-pump-loop at Wed 15

MarvelousMarvin
@MarvelousMarvin
Aug 16 2018 05:38
I'm getting the following msg in the looplog. Why does it cancel the loop?
"If current system time Thu Aug 16 2018 07:34:50 GMT+0200 (CEST) is correct, then BG data is too old. The last BG data was read 16.3m ago at Thu Aug 16 2018 07:18:30 GMT+0200 (CEST). Canceling high temp basal of 1.9"
Temp refreshed: monitor/temp_basal.json: {"duration":0,"temp":"absolute","rate":1.9}
enact/smb-enacted.json: "Rate: 0 Duration: 0
I meen, it says bg data is to old. this happens randomly between the the loop cycles. Just tried to update and it still does the same.
Scott Leibrand
@scottleibrand
Aug 16 2018 05:46
what CGM source are you using?
it cancels the loop because it's not safe to keep running a high temp if the BG data is that old
MarvelousMarvin
@MarvelousMarvin
Aug 16 2018 05:49
Xdrip
And it is not skipping readings in xdrip
Scott Leibrand
@scottleibrand
Aug 16 2018 06:06
How is the data getting from xdrip to OpenAPS?
MarvelousMarvin
@MarvelousMarvin
Aug 16 2018 06:06
NS
Online
MarvelousMarvin
@MarvelousMarvin
Aug 16 2018 06:17
Do you think it is an connectivity issue? I haven't experienced it before.
Scott Leibrand
@scottleibrand
Aug 16 2018 06:29
ns-loop.log has info on NS downloads and uploads.
Sacha M
@coolestkidsever
Aug 16 2018 07:46

Hi all, trying to setup new Pi Hat rig, as EdisonExplorer rig is giving me issues.. but I'm stuck on the Openaps-install.sh.. I keep getting to the same point before it fails...

Selecting previously unselected package liberror-perl.
dpkg: unrecoverable fatal error, aborting:
files list file for package 'libpython2.7-dev:armhf' is missing final newline
E: Sub-process /usr/bin/dpkg returned an error code (2)
root@PiHat4e:~#

I've re-run the curl -s line, but every time it gets stuck here.
Does anyone know how to resolve this please? I'm a bit stuck..

lyntonr
@lyntonr
Aug 16 2018 10:35
Screenshot (6).png

@scottleibrand still a problem scott, dinner bolus wizard at 17:40 ends up as a carb correction and also a meal bolus at 17:40 and at 17:41 , here is relevant pumphistory

 {
    "timestamp": "2018-08-16T19:41:30+10:00",
    "_type": "Bolus",
    "amount": 11.7,
    "programmed": 11.7,
    "unabsorbed": 0,
    "duration": 0
  },
  {
    "timestamp": "2018-08-16T19:41:30+10:00",
    "_type": "BolusWizard",
    "bg": 0,
    "carb_input": 80,
    "bg_target_low": 5000,
    "bg_target_high": 6000,
    "sensitivity": 2900,
    "carb_ratio": 6.8,
    "correction_estimate": 0,
    "food_estimate": 11.7,
    "unabsorbed_insulin_total": 0,
    "bolus_estimate": 11.7
  },
  {
    "timestamp": "2018-08-16T19:40:18+10:00",
    "_type": "BolusWizard",
    "bg": 0,
    "carb_input": 80,
    "bg_target_low": 5000,
    "bg_target_high": 6000,
    "sensitivity": 2900,
    "carb_ratio": 6.8,
    "correction_estimate": 0,
    "food_estimate": 11.7,
    "unabsorbed_insulin_total": 0,
    "bolus_estimate": 11.7
  },

and the NS pics of each
Note that the libre sensor coincidentally stopped at exactly this time also so perhaps this prevented your work around kicking in at the next loop ?

Screenshot (8).png
lyntonr
@lyntonr
Aug 16 2018 10:47
Screenshot (9).png
shot with the units of insulin note the 2 overlayed 80 carbs and the single 11 units in the image.
lyntonr
@lyntonr
Aug 16 2018 11:17
I think I took a while in the wizard while I was checking carb values on the peking duck packet and that it timed out half way through, may or may not be relevant to this issue.
MarvelousMarvin
@MarvelousMarvin
Aug 16 2018 11:36
.
I found this in the loop log. Couldn't make anything of it.
No deliverAt found.
{"reason":"If current system time Thu Aug 16 2018 13:29:30 GMT+0200 (CEST) is correct, then BG data is too old. The last BG data was read 12.5m ago at Thu Aug 16 2018 13:17:02 GMT+0200 (CEST). Temp 0 <= current basal 0.95U/hr; doing nothing. "}
Couldn't smb_verify_suggested
oref0-pump-loop failed. Waiting up to 4 minutes for new BG: ............^C
MarvelousMarvin
@MarvelousMarvin
Aug 16 2018 11:44
A complete log:
Starting oref0-ns-loop at Thu 16 Aug 13:36:05 CEST 2018:
Found recent valid BG:
{"glucose":121,"dateString":"2018-08-16T13:32:02.852000+02:00"}
{"sgv":121,"dateString":"2018-08-16T13:32:02.852000+02:00"}
Latest NS temptargets: {"target":80,"duration":60,"start":"2018-08-15T15:23:18.191Z"}
Merging local temptargets: (NOT VALID JSON: empty)
Temptargets merged: {"target":80,"duration":60,"start":"2018-08-15T15:23:18.191Z"}
Refreshed carbhistory; COB: 0
{"batteryVoltage":4070.0000000000005,"battery":94}
No recent suggested.json found; last updated Aug 16 13:29
No new treatments to upload
Completed oref0-ns-loop at Thu 16 Aug 13:43:19 CEST 2018
after copy/paste found out this was a successful loop
lyntonr
@lyntonr
Aug 16 2018 12:00
@coolestkidsever this is just a guess sacha but maybe try an "sudo apt-get update" to update the package lists and then try again .
@scottleibrand I just realised that maybe this is now the expected behaviour i.e. it did a second pumpcheck and picked up the full bolus wizard and the insulin. There is still the problem of the two 80g carb entries though.
MarvelousMarvin
@MarvelousMarvin
Aug 16 2018 14:18
16:10:29 up 8:23 1 user load average: 4.77 4.62 4.74
Latest NS temptargets: {"target":120,"duration":120,"start":"2018-08-16T13:08:38 .194Z"}
Merging local temptargets: (NOT VALID JSON: empty)
Temptargets merged: {"target":120,"duration":120,"start":"2018-08-16T13:08:38.19 4Z"}
meal.json.new invalid:
{"carbs":0,"nsCarbs":0,"bwCarbs":0,"journalCarbs":0,"mealCOB":0,"currentDeviatio n":2.36,"maxDeviation":1.32,"minDeviation":-0.76,"slopeFromMaxDeviation":0,"slop eFromMinDeviation":0.767,"allDeviations":[2,1,-1,0],"lastCarbTime":0,"bwFound":f alse}
}
ns_meal_carbs failed
{"batteryVoltage":3930,"battery":77}
No recent suggested.json found; last updated Aug 16 16:06
[{"duration":30,"raw_duration":{"timestamp":"2018-08-16T16:01:09+02:00","_type":"TempBasalDuration","duration (min)":30},"timestamp":"2018-08-16T16:01:09+02:00","absolute":0,"rate":0,"raw_rate":{"timestamp":"2018-08-16T16:01:09+02:00","_type":"TempBasal","temp":"absolute","rate":0},"eventType":"Temp Basal","medtronic":"mm://openaps/mm-format-ns-treatments/Temp Basal","created_at":"2018-08-16T16:01:09+02:00","enteredBy":"openaps://medtronic/754","_id":"5b75871f4b30fc8663762001"}]
Completed oref0-ns-loop at Thu 16 Aug 16:15:59 CEST 2018
what am i looking at? is it ns_meal_carbs failed? or no suggested.json?
Scott Leibrand
@scottleibrand
Aug 16 2018 14:21
@lyntonr it sounds like you did in fact enter the carbs twice. While we try to deduplicate those, if they’re far enough apart in time we can’t necessarily do so. Did the COB show up as 80 in the OpenAPS pill, or higher?
@MarvelousMarvin what is the problem?
MarvelousMarvin
@MarvelousMarvin
Aug 16 2018 14:22
i get a lot of failed loops

No deliverAt found.
{"reason":"If current system time Thu Aug 16 2018 16:15:06 GMT+0200 (CEST) is co rrect, then BG data is too old. The last BG data was read 14m ago at Thu Aug 16 2018 16:01:08 GMT+0200 (CEST). Temp 0 <= current basal 0.95U/hr; doing nothing. "}
Couldn't smb_verify_suggested
oref0-pump-loop failed. Waiting up to 4 minutes for new BG: ......................glucose.json newer than pump_loop_completed

Unsuccessful oref0-pump-loop (BG too old) at Thu 16 Aug 16:20:14 CEST 2018

Tim Street
@tim2000s
Aug 16 2018 14:23
@lyntonr - if you enter carbs in the bolus wizard, then let it time out, then go back and do it again, I’ve found that it generally picks up both sets of carbs, as the time out is about 60 seconds.
@MarvelousMarvin - it’s telling you the glucose data is too old at 14m.
MarvelousMarvin
@MarvelousMarvin
Aug 16 2018 14:25
i know, but i am not missing readings in xdrip, you told me to look in ns-loop.log. can`t see why it is going wrong
Zach Gohr
@zgohr
Aug 16 2018 14:27
are they missing in nightscout?
MarvelousMarvin
@MarvelousMarvin
Aug 16 2018 14:29
I found some 8 min breaks in the reading between the 5 min breaks
lyntonr
@lyntonr
Aug 16 2018 15:56
@scottleibrand , @T1Dengineer I think i was just in the wizard got to entering carb value then looked at the packet to check carbs (mmm peking duck) then it timed out and i reentered the wizard and the same values were still there and then i submitted it once. Are you saying that the timed out session actually recorded the carbs as submitted ? Is the pump that foolish ? Or does the pump not care about the carbs beyond the wizard calculation even though we do ?
lyntonr
@lyntonr
Aug 16 2018 16:09
Sorry @T1Dengineer i meant to tag @tim2000s in my last post.
kelseyyearick
@kelseyyearick
Aug 16 2018 16:26
image.png
Has anyone seen this highlighted error?
kelseyyearick
@kelseyyearick
Aug 16 2018 16:31
@lyntonr I am having the same issue, do to a broken screen. Did you get this resolved?
Dana Lewis
@danamlewis
Aug 16 2018 17:08
@lyntonr yes some models of pumps record carbs even if you don't go through the wizard all the way. That's why we switched to only using records from the pump that has carbs plus bolus. Can you confirm if the OpenAPS logs were reading both bolus wizard entries? Need to see if the Go code is properly excluding the non-insulin entry. (This is separate from the de-dupe time out check)
Arnoud82
@Arnoud82
Aug 16 2018 19:30
Hello is there a fix for getting BG to NS using DEV0.7.0 and MDT
Sacha M
@coolestkidsever
Aug 16 2018 20:57

@lyntonr Thanks for that, but it didn't work...

Still getting the following error:
Selecting previously unselected package liberror-perl.
dpkg: unrecoverable fatal error, aborting:
files list file for package 'libpython2.7-dev:armhf' is missing final newline
E: Sub-process /usr/bin/dpkg returned an error code (2)
root@PiHat4e:~#

And I can't get any further along in the install..

re-running curl -s gets me to the same point every time
lyntonr
@lyntonr
Aug 16 2018 22:45
@danamlewis the logs show both entries going off to nightscout together , and also show autosens ignoring the one without insulin. But shouldnt it also be excluded from being sent to nights out ? Will cut together logs when near a laptop a bit later on, still in bed now :)
Dana Lewis
@danamlewis
Aug 16 2018 22:58
what to look for: is if the pump-loop log is registering both sets of carbs?