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

2nd
Aug 2016
jmcrawford
@jmcrawford
Aug 02 2016 00:18
Just getting started with all this, so I've got lots of dumb questions. I've got a Medtronic 754, shows ver 2.7A 1.1. There is no PC-Connect option under the Connect Devices. Will this pump work?
Brandon Faloona
@bfaloona
Aug 02 2016 00:20
@jmcrawford I think it MUST have the PC-Connect option. :(
Dana Lewis
@danamlewis
Aug 02 2016 00:21
is it a European pump, @jmcrawford?
we know European x54 with 2.6 works…worth trying to see if this one works if it’s non-US
Chris Oattes
@cjo20
Aug 02 2016 00:34
@bfaloona "PC Connect" means that it won't work with openaps looping. We need pumps without that option
Brandon Faloona
@bfaloona
Aug 02 2016 00:34
Oh! Thanks for correcting me.
jmcrawford
@jmcrawford
Aug 02 2016 00:42
@danamlewis not sure of it's origin, bought online. Long story, but as a recent T1D dx, current political climate dictates ~2yr wait for a new pump. So we've gone a bit rogue.
@cjo20 That was my understanding also, just wanted to confirm. Thx
John Dodds
@johners97
Aug 02 2016 00:54
@danamlewis How many new loopers now?
jmcrawford
@jmcrawford
Aug 02 2016 00:58
We're in Canada, I wonder if the pumps here are the same as Europe. The main difference I'm aware of is the units (mmol/l vs mg/Dl).
John Dodds
@johners97
Aug 02 2016 01:06
In the UK we use mmol/L too, although our pumps are the same as those in mainland Europe, and they use mg/dL? Canada uses mmol yeah?
Let's just design our own pump... if I had the tech know-how I would
jmcrawford
@jmcrawford
Aug 02 2016 01:11
So aside from units and language, what other differences could there be?
John Dodds
@johners97
Aug 02 2016 01:15
hmm good question
Hichame Yessou
@hichameyessou
Aug 02 2016 01:15
@jmcrawford Maybe the frequencies of the sensor/Carelink? 868Mhz instead of 915Mhz? Or it's only a matter of the US version?
Brandon Faloona
@bfaloona
Aug 02 2016 01:17
@scottleibrand @bewest Thanks for spending that time helping me. :clap: I think we solved a few major glitches, but I'm still testing! :)
John Dodds
@johners97
Aug 02 2016 03:01
In the 2/3 days i've been looping, i've been using 29% less basal insulin. Which is great because too much plasma insulin is just almost as bad for health as high glucose levels. Anyone else using less insulin?
Brandon Faloona
@bfaloona
Aug 02 2016 04:12
I'm debugging this crash now ...
pi@openaps-lp:~/openaps-lp $ openaps report invoke oref0-monitor/profile.json
get-profile://text/shell/oref0-monitor/profile.json

/usr/local/lib/node_modules/oref0/lib/profile/basal.js:10
    var basalRate = basalprofile_data[basalprofile_data.length-1].rate
                                                                 ^
TypeError: Cannot read property 'rate' of undefined
    at Object.basalLookup (/usr/local/lib/node_modules/oref0/lib/profile/basal.js:10:66)
    at generate (/usr/local/lib/node_modules/oref0/lib/profile/index.js:32:33)
    at Object.<anonymous> (/usr/local/lib/node_modules/oref0/bin/oref0-get-profile.js:104:19)
    at Module._compile (module.js:456:26)
    at Object.Module._extensions..js (module.js:474:10)
    at Module.load (module.js:356:32)
    at Function.Module._load (module.js:312:12)
    at Function.Module.runMain (module.js:497:10)
    at startup (node.js:119:16)
    at node.js:906:3
reporting oref0-monitor/profile.json
The report cli is: openaps use get-profile shell settings/settings.json settings/bg_targets.json settings/insulin-sensitivities.json settings/selected-basal-profile.json preferences.json
Eric Jensen
@elnjensen
Aug 02 2016 04:22
I'm stepping through the walkthrough in the docs, and trying to be diligent about following each step and understanding what it does. But I'm wrestling with the mixed use of what seems to be newer and older ways of doing things, which sometimes clash. In particular, there's a note at the start of 'phase-2/Using-oref0-tools.html' that says "Here are some of the newer reports copied from the newer style walkthrough" which then gives some oref0 commands to create reports and devices (but then also says that they are inconsistent with the rest of the walkthrough). Is this "newer style walkthrough" available somewhere? If not, any suggestions about what path to follow through the docs?
Dana Lewis
@danamlewis
Aug 02 2016 04:24
it’s now toward the top part in the read the docs flow (aka “guide to understanding openaps”): http://openaps.readthedocs.io/en/latest/
if you’re looking on github, there’s a link at the top of the nav of the outline in the readme that links to it, but the readme needs a PR to reflect the flow as outlined in the RTD version (auto generated)
Martin Haeberli
@mhaeberli
Aug 02 2016 04:28
@danamlewis thx - I just yesterday got lost in the github docs again ...
Dana Lewis
@danamlewis
Aug 02 2016 04:29
PR PR PR :)
Eric Jensen
@elnjensen
Aug 02 2016 04:33
@danamlewis Thanks, but I don't quite follow. I see the "guide to understanding openaps" link (and I am indeed in readthedocs). It looks like that set of links explains a bunch of things nicely. But is one not supposed to go on to the Walkthrough section after that? Since there is no 'next' link at the bottom of any of the doc pages I've been using the left-hand Navigation bar to try to figure out where to go next.
Dana Lewis
@danamlewis
Aug 02 2016 04:35
yes. the context is, there used to not be that “understanding” guide to help you get introduced to the toolset (openaps). Ben added that guide, and moved it to the top because it seemed to help people get their brain wrapped around the toolkit before they started applying it to building an implementation (which is what the walkthrough docs are).
Whether using RTD or using github, I usually keep a tab open with the highest level outline/TOC (in github, it’s the readme page) and use that to help me navigate in another tab as I move through, so I can see which section I’m in and how it relates to the overall flow.
Martin Haeberli
@mhaeberli
Aug 02 2016 04:37
@danamlewis :+1:
Eric Jensen
@elnjensen
Aug 02 2016 04:38
Cool. I feel like I totally get openaps. But mixed in with the openaps stuff (pretty early on in the walkthrough) is some oref0 stuff that is much more cryptic. For example, in the middle of the pretty gentle walkthrough of adding devices, in Phase 1, are these two lines that kind of come out of nowhere:

Create bunch of reports:

oref0 template mint reports medtronic-pump | openaps import
and then that adds about 25 reports to your openaps.ini, including with subdirectories that aren't created yet, devices (units) that don't exist yet, etc.

Dave Ewall
@daveewall
Aug 02 2016 04:41
@scottleibrand I'm running AMA. (sorry for the delay.)
Dana Lewis
@danamlewis
Aug 02 2016 04:41
(Yea. We’re getting to the point where there’s been a bunch of adds to the docs, but no one has taken the time to do a start to finish run through recently with a new rig setup and flag where there’s any jarring points from new adds that need smoothing. So if you could PR clarifications, or file an issue to make note of them so someone else could tackle, that would be great!)
Eric Jensen
@elnjensen
Aug 02 2016 04:41
So that's the kind of things that's hanging me up. Not sure whether I should be skipping the oref0 stuff that comes early on (well before oref0 is even mentioned) - but once I start skipping things, it gets a little tricky to figure out what to keep in and leave out...
OK, perfect. Not trying to be hypercritical here at all - just wanting to figure out if I'm missing something particular. I've done a few PRs to fix trivial things in the docs, but there are some higher-level organizational issues that I'm not sure what to do about.
Dana Lewis
@danamlewis
Aug 02 2016 04:42
you can also click in the bottom right corner of RTD and see there’s different date versions, and see if an earlier version of the docs is more helpful to you.. some people have preferred that. but generally, if you find it jarring, others will too, and it’s worth flagging in a place where it can get tracked & followed up on. Here is good for asking a Q, but it’s super helpful to the community to submit a PR to fix or log an issue
no worries. and not just saying this to you - saying this to everyone who’s got fresh eyes looking at the docs.
Eric Jensen
@elnjensen
Aug 02 2016 04:43
So if there is something that doesn't quite seem to fit, but I don't know enough to fix and PR, is the right thing to do to file an issue for that bit?
Dana Lewis
@danamlewis
Aug 02 2016 04:43
yes.
Eric Jensen
@elnjensen
Aug 02 2016 04:44
OK, will do. Thanks for being so helpful here.
Dana Lewis
@danamlewis
Aug 02 2016 04:44
:+1: thanks for flagging things, asking Qs, and being willing to file issues! helps a lot :)
danamlewis @danamlewis encourages everyone to submit PRs or flag issues at any time, don’t wait until you’re done with your setup, otherwise you won’t remember what was confusing at various points in the docs.
Dana Lewis
@danamlewis
Aug 02 2016 04:44
</soapbox about docs> :)
Eric Jensen
@elnjensen
Aug 02 2016 04:46
:thumbsup:
Dana Lewis
@danamlewis
Aug 02 2016 05:06
Thx @elnjensen for logging that as an issue!
Brandon Faloona
@bfaloona
Aug 02 2016 05:08
Well, I was unable to get a consistent reproduction the crash I pasted above. The behavior kept changing, so now I expect these errors are occurring every loop, instead of sporatically.
Aug  1 21:48:00 openaps-lp size (0) is less than 64 and not 15, which may cause an error.
Aug  1 21:48:03 openaps-lp bad zero CRC?
Aug  1 21:48:06 openaps-lp Stick transmit[TransmitPacket:ReadProfile_STD512:data:unknown] reader[ReadRadio:size:206] download_i[2] status[<LinkStatus:0x03:status:size=??LinkStatus:error:True:reason:[]:size(0)>] poll_size[0] poll_i[False] command[<LinkStatus:0x03:status:size=??LinkStatus:error:True:reason:[]:size(0)>]:download(attempts[2],expect[0],results[0]:data[0]):BAD AILING
Usually, a few Stick transmit errors don't block a report, but things are going downhill. :)
I'll reboot and keep at it.
Scott Leibrand
@scottleibrand
Aug 02 2016 12:14
@daveewall k. Sounds like you might want to look at whether your carb ratio is too high to start with. If you could adjust the carb ratio on the pump, but continue to give the same size meal bolus, AMA will be less inclined to low temp, and should even high temp somewhat to help out. Or, if you rarely see lows after meals, you could also do the higher boluses as well: then low-temping would be just as likely, but the extra bolus insulin would help prevent meal spikes.
You should also be able to look at net BG rise after meals, and whether additional insulin is usually required to correct, to get an estimate of whether your carb ratio is too high.
If it turns out your carb ratio is accurate and it's just a timing thing, you could also look at slowing the bolus snooze time back down to what it was in master (DIA/2 instead of DIA/4 IIRC). @jasoncalabrese was exploring that for similar symptoms, and found it helped. You'll want to make sure your carb ratio is well tuned first though, as otherwise bolus snooze is just a bandaid.
Chris Oattes
@cjo20
Aug 02 2016 17:11
Trying tighter targets and a higher max basal. Fingers crossed.
Brandon Faloona
@bfaloona
Aug 02 2016 20:52
Any suggestions for easy ways to validate reports after gathering. I'm getting lots of Stick transmit and other random errors. Everything works fine when those don't occur.
I currently test that warmup worked, but I'm considering testing additional reports before running the predict / enact steps.
Brandon Faloona
@bfaloona
Aug 02 2016 21:01
@bewest @scottleibrand Those report changes we made yesterday got our IOB numbers back into the believable range. Success! Are the nightscout reports for IOB a valuable way to test those values? The data looks a bit erratic.
Scott Leibrand
@scottleibrand
Aug 02 2016 22:03
cool. I have never really used those, so not sure