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

1st
Sep 2017
Martin Haeberli
@mhaeberli
Sep 01 2017 00:08
how are things coming on: a) display for openaps (our friends in AZ? or NV? was it?); and b) alternative to Edison after end of life 12/31/2017?
Aaron Michelson
@Aaybob
Sep 01 2017 00:11
@scottleibrand no cat ~/myopenaps/monitor/glucose.json returns nothing
Scott Leibrand
@scottleibrand
Sep 01 2017 00:12
Then that's your problem: your rig isn't getting glucose data from NS. Check your NS URL?
Aaron Michelson
@Aaybob
Sep 01 2017 00:13
@scottleibrand yes thanks!
Scott Leibrand
@scottleibrand
Sep 01 2017 00:13
@mhaeberli no recent news from AZ. I'll be back in PHX a week from Monday, so I'll probably get a chance to see their latest work then. In the mean time you could ask in the openaps-hardware-dev google group.
They're making a Pi hat, so I'm assuming our next platform will be the Pi0W, but am in no hurry to switch anyone over until we have a hat shipping (probably not until December or so I'd guess)
useraps
@useraps
Sep 01 2017 00:15
Hello and sorry but I entered letters for my pump serial # during setup. Is there an easy way to edit the sn to just the numbers? I was reading through the docs and didn't see where I would edit it at?
Martin Haeberli
@mhaeberli
Sep 01 2017 00:15
@scottleibrand thx me neither just trying to get sense of direction
@useraps I believe you’ll find a file called something like “settings …” in your openaps directory. you should be able to edit that, then reboot
Scott Leibrand
@scottleibrand
Sep 01 2017 00:16
@useraps best to edit oref0-runagain and re-run it.
Martin Haeberli
@mhaeberli
Sep 01 2017 00:16
but I as always defer to @scottleibrand
Scott Leibrand
@scottleibrand
Sep 01 2017 00:17
I think it's stored in pump.ini if you wanted to edit it directly. Not sure if it's also stored anywhere else.
Martin Haeberli
@mhaeberli
Sep 01 2017 00:23
pump.ini it is!
useraps
@useraps
Sep 01 2017 00:24
I did see that. Reading the docs to figure out how to see the contents to edit. Total noob!
Martin Haeberli
@mhaeberli
Sep 01 2017 00:26
@useraps you can use nano for example - a little more obvious than vi
so cd to the directory, say nano pump.ini ask Google also for hints on nano
Aaron Michelson
@Aaybob
Sep 01 2017 00:28
@scottleibrand I hadn't typed "s" in the "http://" part of my nightscout website. So edited my saved oref0-setup string and redoing. Thanks, I had a bad feeling about typing, not copy pasteing like usual.
@useraps re-runing setup is fast.
useraps
@useraps
Sep 01 2017 00:35
Thank you. I used VI. I did try that but I wasn't in the correct directory. Now I'm off to test my loop. Wish me luck
Aaron Michelson
@Aaybob
Sep 01 2017 00:38
@useraps good scientific luck :clap: I hope the test results meet expectations?
useraps
@useraps
Sep 01 2017 00:41
One more question. Can Max_iob be set to .5 This means it will never set the temp basal above .5u/hr. Is that correct?
John Benjamin
@Bender1061
Sep 01 2017 00:51
@useraps Max_iob is the maximum amount of insulin that you can be given above what is normally given to you with your normal basil rates. for example if the rig wanted to give you 1u/h more than your normal rate, you would max your IOB in 30 mins, and then it would not be able to set a temp basal (above your normal rate) until that insulin has been used up (at like 3units/hr) To make it so that you can't get above .5u/h you would have to change Max Basal Rate on the pump itself.
Aaron Michelson
@Aaybob
Sep 01 2017 00:57
@useraps you got basal and bolus on the pump. anything above basal is iob, insulin on board. Think of it as your basal is perfect, keeps blood sugar at normal. iob will lower your blood sugar. Normally you bolus for carbs. That bolus is insulin on board to counteract your carbs on board. And the canoe is balanced and floats gently down the river.
Aaron Michelson
@Aaybob
Sep 01 2017 01:03
as the iob and cob gently drop for four hours, now that would be perfect.
useraps
@useraps
Sep 01 2017 01:06
@Bender1061
Thank you. That is what I thought. I just worded it wrong.
Martin Haeberli
@mhaeberli
Sep 01 2017 01:38
In the unlikely event that anyone has a seriously broken formerly - loopable Medtronic pumps that they can donate - I am hoping to figure out how to make more 'unicorn' pumps out of existing newer Medtronic pumps - please pm me ... thx
Jacob H
@jdhigh
Sep 01 2017 03:07
@mhaeberli Not sure what you mean by seriously broken and formerly loopable?
Katie Aldridge
@kcrcgm
Sep 01 2017 03:07
Good evening, everybody. I'm trying to work out problems I've had with monitoring OpenAPS offline with the offline webpage. I have gotten my rig to broadcast the offline webpage, and I can view it both in a browser and in the Android app HTTP Widget. But it's not updating. I get maybe one or two refreshed webpages in a 24 hr period. I can't figure out why it's not running every 5 minutes like it's supposed to. I can't figure out why it runs when it does. It seems random. This is the cron line from the docs that is supposed to make it run regularly. Does it look right? */5 * * * * (bash /root/http.sh) 2>&1 | tee -a /var/log/openaps/http.log
Scott Leibrand
@scottleibrand
Sep 01 2017 03:09
@Bender1061 not sure what you mean by insulin getting used up at 3u/h. IOB decays over time, not at a constant rate per hour like it can be infused...
Martin Haeberli
@mhaeberli
Sep 01 2017 03:17
@jdhigh I mean out of warrantee, preferably early enough to be loopable, but not functioning. As in dropped, cracked, not repairable. But not run over by a steamroller. Non loopable newer pumps might be ok too
Jacob H
@jdhigh
Sep 01 2017 03:41
Hi all,
I'm seeing a lot of these: Timed out or other comms error - Received an error response Timeout - retrying: 1 of 3
What does it mean?
katie disimone
@Kdisimone
Sep 01 2017 04:01
@jdhigh normal message. Doesn't mean anything is wrong. They show on localhost in papertrail all the time.
Jacob H
@jdhigh
Sep 01 2017 04:03
Thanks! I just installed papertrail, so I'm reviewing error messages a little more tonight than normal. One other question: Is it normal to have frequently (20+ times /day) "pump data stale" and "openAPS isn't looping" messages come through on pushover?
Martin Haeberli
@mhaeberli
Sep 01 2017 04:25
@kcrcgm with respect to offline web page, without revealing anything (like pump serial #) can you share the .html you use on your page? I ask because there is a possibility of setting an auto-refresh header, but I haven’t read the notes on the offline web page stuff yet
katie disimone
@Kdisimone
Sep 01 2017 05:54

@jdhigh that's a lot more often than we get, yes that's an unusual amount of instances. However, if you have one instance that goes on for a long period of time, pushover will keep alerting until you both snooze NS alarm AND acknowledge the pushover alert. So one longer time of OpenAPS being offline could lead to a lot of notifications for the same instance if the alarm/alerts aren't acknowledged.

But generally speaking, we'd go days without any significant periods of OpenAPS not looping. Our alerts were set to 20min. So usually we don't notice until it gets longer than 20 min of OpenAPS being out.

Katie Aldridge
@kcrcgm
Sep 01 2017 12:48
@mhaeberli I'm not sure! When I was able to see it in the browser, the web address was like http://192.168.44.70:1337/index.html (this isn't the real address, bc I wasn't sure that was safe). In the rig setup, it must just be a default of the Simple Http Server because I didn't set anything. I'd love your help if you have ideas! https://github.com/ceben80/http-widget/blob/master/README.md
Tim Street
@tim2000s
Sep 01 2017 14:37
@kcrcgm - what are the contents of your file and are you using oref1 or oref0?
(ie have you enabled SMB)
Jacob H
@jdhigh
Sep 01 2017 15:40
Really basic question: enableSMB_with_COB means SMB will be enabled when you've entered carbs. What are the ways to enter carbs? I have a MDT pump, and I use nightscout.
Scott Leibrand
@scottleibrand
Sep 01 2017 15:48
Either of those will work, or IFTTT etc. See the docs for details.
Martin Haeberli
@mhaeberli
Sep 01 2017 15:52
@kcrcgm Tim asked the question better than I did - to help, it would be good to know the contents of the file on your rig that makes the web page ...
@jdhigh - where are you setting up / enabling messages like ‘pump data stale’ - in Nightscout? - now I have to go read more docs...
Jacob H
@jdhigh
Sep 01 2017 15:59
Thanks @scottleibrand I think I'll create a PR to define COB in the docs. (It's not in the glossary.) @mhaeberli the pump data stale are coming through pushover. It's not something I recall setting up— it's getting really annoying.
Martin Haeberli
@mhaeberli
Sep 01 2017 16:03
@jdhigh i’m guessing its a NightScout variable. because we have edison pushover set here in default post-install oref1 SMB ‘installed’ but all SMB flags false, and are getting lots of interesting pushovers but no stale. Suggest you consider sharing a (sanitized for privacy) set of NS variable settings to see if we can weigh in.
Jacob H
@jdhigh
Sep 01 2017 16:05
Yup. I found it. https://github.com/nightscout/cgm-remote-monitor I think this is it: TIMEAGO_ENABLE_ALERTS (false) - Set to true to enable stale data alarms via Pushover and IFTTT.
Martin Haeberli
@mhaeberli
Sep 01 2017 16:07
:+1:
Jacob H
@jdhigh
Sep 01 2017 16:23
I've got a little free time today. Does anyone think it would be ok for me to submit a PR to alphabetize the glossary? https://openaps.readthedocs.io/en/latest/docs/Resources/glossary.html
dan lightner
@danimaniac
Sep 01 2017 16:25
@jdhigh alphabetized glossary would make sense to me.
David
@davidkeddydb
Sep 01 2017 17:12
If this is my crontab section relative to autotune, with the 50 in front, it means it in essence is commented out to not run? 5 0 * ( oref0-autotune -d=/root/myopenaps
Curses, that 50 is the time. I need to comment it out so it won't run
Dana Lewis
@danamlewis
Sep 01 2017 17:15
@jdhigh it's always ok to submit PRs :) thanks!
Brian Rabinovitz
@bdr1177
Sep 01 2017 19:58
Pump-rig connection problem. Our child's rig is on the school wifi and works fine at school. When she comes home it successfully connects to home wifi, but then the communication between the pump and the rig breaks down, says no acknowledgement from pump on wakeup, and mmtune shows "916.636", 0, -99. If I manually restart the rig, that solves the problem and it starts looping again without issue. There doesn't seem to be any problem going the other direction from home to school wifi in the morning. Don't see anything in the troubleshooting docs that would explain this behavior. Any ideas?
Dana Lewis
@danamlewis
Sep 01 2017 19:59
:clap: to @kcrcgm and @jdhigh for PRing into the docs! Thx for the edits & additions.
btw, another place that needs some love if anyone's on a PR kick is the understanding determine-basal page (http://openaps.readthedocs.io/en/latest/docs/While%20You%20Wait%20For%20Gear/Understand-determine-basal.html); it needs to be updated to match the current content displayed in the logs/pill in NS, so if anyone wants to add in some more recent examples, that would be great.
@bdr1177 do you know how long after it switches wifi that continues?
Brian Rabinovitz
@bdr1177
Sep 01 2017 20:10
It continues indefinitely until I do a hard restart.
Martin Haeberli
@mhaeberli
Sep 01 2017 20:13
off topic but relating to post-Edison APS and AZ hardware team - probably too much, but maybe add a 433.923 MHz radio option should the omnipod team make progress on that front?
Dirk Gastaldo
@dirkgastaldo
Sep 01 2017 20:35
Hello, I think I saw someone say 0.5.3, is that the current version?
Dana Lewis
@danamlewis
Sep 01 2017 20:36
@dirkgastaldo that's in dev, and it's a release candidate to go to master here soon. More testers welcome.
details here: openaps/oref0#591
Dirk Gastaldo
@dirkgastaldo
Sep 01 2017 20:36
Ah, thanks, Dana, perfect, I shall read
Dana Lewis
@danamlewis
Sep 01 2017 20:37
@bdr1177 k. what version are you on?
wondering if that's the same thing @jsmurray1 was seeing on his Pi rig when testing 0.5.3-dev, or something else.
Dirk Gastaldo
@dirkgastaldo
Sep 01 2017 20:40
I love this: Calling out the number one beginner error #586
Dana Lewis
@danamlewis
Sep 01 2017 20:40
;)
Dana Lewis
@danamlewis
Sep 01 2017 20:46
Scott was not a fan of my long, literal branch name for that one :)
Brian Rabinovitz
@bdr1177
Sep 01 2017 20:48
@danamlewis Yes I'm on latest 0.5.3 dev. I'll take a look at that issue and see if it sounds the same.
Dirk Gastaldo
@dirkgastaldo
Sep 01 2017 20:56
I have two rigs, both with 0.5.2. I guess that doesn't make sense. I should just turn off one of them and install the DEV branch on the other. I can always turn the DEV rig off and run the 0.5.2 rig, if need be?
Scott Leibrand
@scottleibrand
Sep 01 2017 21:01
yep
that's good practice: to keep one known-good stable rig and try out new stuff on the other
Dirk Gastaldo
@dirkgastaldo
Sep 01 2017 21:02
Thanks, Scott.
Luis Betancourt
@betluis
Sep 01 2017 22:05
Hello. I had a quiet period of about one month with my daughter running openaps 5.2 with good consistency... last night she was higher than normal and just noticed that for several hours the openaps pill is "unknown" despite loop is ok (or apparently ok) temp basals and SMB ok, just checked the pumphistory24h file... Any clue as to what could cause the openaps pill like this? BG shows fine
Dana Lewis
@danamlewis
Sep 01 2017 22:05
check for future treatments in NS; then check mlab size
Luis Betancourt
@betluis
Sep 01 2017 22:06
ok will check mlab size, how do I check future treatments in NS though?
mLab size 367.29 MB / size on disk 496.00 MB
Scott Leibrand
@scottleibrand
Sep 01 2017 22:10
size on disk 496.00 MB means it's maxed out
Luis Betancourt
@betluis
Sep 01 2017 22:11
right
Luis Betancourt
@betluis
Sep 01 2017 22:11
just realizing this
thanks Scott and Dana
Scott Leibrand
@scottleibrand
Sep 01 2017 22:12
per https://mlab.com/plans/pricing/ starts at $15/mo if anyone wants to not have to worry about this limit...
maybe we need to figure out a way to help alert people when it gets big and proactively compact or whatever...
Luis Betancourt
@betluis
Sep 01 2017 22:13
it would be usefull
there is no option to delete automatically the last records in order to let the new ones get in I suppose...
Scott Leibrand
@scottleibrand
Sep 01 2017 22:16
compacting would be the first thing to try. should get the size on disk down closer to the actual size
but that's all in the doc Dana linked
Luis Betancourt
@betluis
Sep 01 2017 22:17
yes, I see it
thanks
Brian Rabinovitz
@bdr1177
Sep 01 2017 22:37
@danamlewis The communication issue that @jsmurray1 mentions seems different because as far as I can tell our issue is only with unsuccessful mmtuning.
Luis Betancourt
@betluis
Sep 01 2017 22:46
issue fixed thanks a lot.... something else. I have 4 rigs... interestingly two rigs can't connect to the pump anymore, and both showed the same issue at the same time. I am suspecting there may be issues with the explorer boards, it looks as if the pump was not in proximity to the rigs. the other two rigs work fine with the exact same conditions. I read somewhere that the explorer board chip can be formatted but haven't manage to do this so far. the other option I am considering is to purchase replacement explorer boars... I'd appreciate any suggestion.
Dana Lewis
@danamlewis
Sep 01 2017 22:56
what is the error message?
Luis Betancourt
@betluis
Sep 01 2017 23:05
Dana, it is the same message as if the ok rig was away from the pump
while troubleshooting I can't get the rig to extract the model from the pump (basic communication troubleshooting)
"openaps use pump model" no success... while the other two rigs are fine
I wonder if we have already seen cases of failed explorer boards? or similar cases?
Scott Leibrand
@scottleibrand
Sep 01 2017 23:09
search the docs for ccprog
Luis Betancourt
@betluis
Sep 01 2017 23:12
Ok will do...
I will work on it during the weekend
Dana Lewis
@danamlewis
Sep 01 2017 23:25
@betluis did you killall before issuing the manual commands?
PieterGit
@PieterGit
Sep 01 2017 23:28
i think i found a main cause of why were seeing too many pump errors with 0.5.3, see https://github.com/openaps/oref0/pull/591#issuecomment-326703920
unfortuately not able to PR this before sunday
Brian Rabinovitz
@bdr1177
Sep 01 2017 23:37
@danamlewis
Starting pump-loop at Fri Sep 1 15:33:04 EDT 2017:
Sep 01 15:33:04 natopenaps pump-loop.log: Waiting up to 4 minutes for new BG: glucose.json newer than pump_loop_completed
Sep 01 15:33:38 natopenaps pump-loop.log: Radio ok. Listening: .No pump comms detected from other rigs
Sep 01 15:34:07 natopenaps pump-loop.log: Profile less than 60m old. glucose.json newer than temp_basal.json. Temp refreshmmeowlink.exceptions.CommsException: No acknowledgement from pump on wakeup. Is it out of range or is the battery too low?