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

18th
Feb 2017
Eric
@ecc1
Feb 18 2017 00:04
my android phone has two different options, wifi hotspot and bluetooth hotspot. you need the 2nd
David
@david-t1
Feb 18 2017 00:06
this is why I wonder if I need different phone or rollback my phone to lollipop in order to see an option for bt hotspot
Eric
@ecc1
Feb 18 2017 00:10
Hmm, some Googling shows some issues w.r.t. BT tethering on Marshmallow. Maybe try: http://techiezlounge.com/enable-tethering-android-6-0-marshmallow/
(I use Nougat, so proceed at your own risk)
David
@david-t1
Feb 18 2017 00:12
tx!
dan lightner
@danimaniac
Feb 18 2017 01:47
@peterfulvi @beached @Scorponok @thebookins Today's progress on my case can be seen at the following link. I added the beginnings of a cap, and a few other small details. I still need to make some revisions to the mating surface between the case body and the cap, but have run out of time for now. I'll come back to it again over the weekend. Let me know of any suggested revisions or problems you can anticipate with the design.
danimaniac's OpenAPS Case
!https://goo.gl/photos/hmVf8tBg1pEqEbSi9
@irunemur I wonder if you would be willing to adjust some dimensions of this case once I actually have all the components assembled? Also, when making my revisions I could not round the vent corners like you did the other day. I like the rounded vents and did not intend on reverting back.
Tim Street
@tim2000s
Feb 18 2017 02:28
@danimaniac - do you have anything separating the lipo from the Edison? If not suggest you add it in. There are a few spiky bits on the exb that could damage the lipo, which you don't want.
dan lightner
@danimaniac
Feb 18 2017 02:32
@tim2000s Yes, I the case has two compartments separated by a 1mm thick printed divider. Thanks!
Tim Street
@tim2000s
Feb 18 2017 02:38
Fair enough. Also, how do the wires work in that case?
dan lightner
@danimaniac
Feb 18 2017 02:52
@tim2000s I may have to make some adjustments after I actually have my rig assembled and see the EB 2.0. It is still on backorder. Anyway that's why I created 2 tabs on the lid with a space in between instead of just one large tab. I think the wires running from the lipo to the EB will tuck down in between the 2 tabs. There should be enough space for the wires between the top of the EB / Edison and bottom of the lid ...won't know for sure until the rig is assembled.
Aaron Michelson
@Aaybob
Feb 18 2017 04:28
@Kdisimone @scottleibrand Re: The easiest way to make a PR is to...... yes start edits in the "main" and then create PR, create PR, create PR, or green button green button green button. Or if you prefer you can study software engineering before pressing the PR button.
katie disimone
@Kdisimone
Feb 18 2017 04:34
@Aaybob ??
Aaron Michelson
@Aaybob
Feb 18 2017 04:37
@Kdisimone it was something you wrote five days ago about how to do PR more easily....sounded like you updated the docs....I had written that but couldn't explain it using git terminology so maybe it wasn't in the docs yet.....remember?

Is there an update to the function of the logging
CGM-LOOP.log? my working system is just giving this:

Fri Feb 17 20:35:02 PST 2017
AttributeError: 'NoneType' object has no attribute 'iter_records'
248 2017-02-17T20:30:41-08:00

Aaron Michelson
@Aaybob
Feb 18 2017 04:47
@Kdisimone am I being too cryptic?
katie disimone
@Kdisimone
Feb 18 2017 04:49
Perhaps. I'm not sure if you're offering comments on the new version of the how-to-do-a-PR? Or just commentary in general? There was some bug edits to that page to add screen shots and help. You don't like that?
Aaron Michelson
@Aaybob
Feb 18 2017 04:50
@Kdisimone No, I haven't looked at the docs in a while...
Aaron Michelson
@Aaybob
Feb 18 2017 04:57
You posted something and scott @ me in a reply to you saying "aaron found that out the hard way". I had PR'd the "how to do your first PR" documentation with Dana. I wanted to describe that way of PRing and wanted to leave out all the GIT theory. So if that method of PRing to the master (not local or repo or whatever its called) was not in the docs its not for lack of trying to put it there. scott gave your comment the thumbs up symbol so maybe they finaly think its ok to not study software engineering, grocking git and throwing in gratuitous references to repository forks.
Aaron Michelson
@Aaybob
Feb 18 2017 05:05
Wow big change to the doc @Kdisimone ! Were the images uploads a seperate pull request?
Aaron Michelson
@Aaybob
Feb 18 2017 05:11
I'm using the explorer board I thought I broke for "out of internet" use. I had plugged in to the battery terminal with reverse polarity and smoked the board. It now powers up through the uart port on a 10Ah battery, cgm connected to OTG port. I can only see that the pump is getting temp basals and I don't get a record. Still hoping to make Pancreabble run the smartwatch from the edison like I had previously. But it is so great to have the added health benefit at work. Wow!
My non-smoked explorer board didn't seem to accept the trace cut update. Thinking I need to order the board with the update.
katie disimone
@Kdisimone
Feb 18 2017 05:28
@Aaybob same PR but I have a second PR still outstanding to see if the RTD version will display the screenshots (I think the image paths need to be updated to make that happen)
Aaron Michelson
@Aaybob
Feb 18 2017 05:29
@Kdisimone you put in images as a reference to a png file in the directory https://github.com/openaps/docs/tree/master/docs/docs/Images how did the png files get placed in that directory?
katie disimone
@Kdisimone
Feb 18 2017 05:32
@Aaybob if you add them to your temporary branch's images folder, the PR will automatically include them in the pull. The trick is getting the path correctly written in for when they move from your repo's branch to the OpenAPS/docs repo.
Getting the path (in the docs) correctly written. It's easy to start with them pointing to your own images folder, but the final destination will need to point to OpenAPS/docs instead of my own
Aaron Michelson
@Aaybob
Feb 18 2017 05:39
tough one because the RTD version is a reference to http://openaps.readthedocs.io/en/latest/docs/introduction/understand-this-guide.html Maybe super user needs to create a directory in http://openaps.readthedocs.io/ for images.
Aaron Michelson
@Aaybob
Feb 18 2017 05:47
@ejoe132 How does one find out how much data . I know I haven't gone over my 2G limit.
Aaron Michelson
@Aaybob
Feb 18 2017 05:58
@Kdisimone I think the text there is what I wrote. I hadn't seen it there before. You added the pictures but they don't show up in the RTD version. But when you go to edit that version and then select preview changes the pictures do show up. Am I on the same page/understanding what you were saying just now?
katie disimone
@Kdisimone
Feb 18 2017 06:00
@Aaybob that's pretty much what I'm saying yes. I have a PR in that makes the file path less specific for the images. I think that will fix the rtd problem.
openaps/docs#665
And if you want to see a history of the commits, some of the language was tweaked a little, but not much.
Aaron Michelson
@Aaybob
Feb 18 2017 06:11
@Kdisimone isn't Markup supposed to automatically make a link when you write this as you did: Go to https://github.com/openaps/docs and hit Fork in.... but it didn't.
Paul Dickens
@thebookins
Feb 18 2017 10:22
Hi everyone. The OpenAPS pillbox on our nightscout site has mysteriously stopped updating, even though OpenAPS is happily looping and upload/ns-status.json is up to date. Anyone seen this before? I don't think we changed anything!
Also, everything else is updating fine on NS - glucose, temp basals etc.
dan lightner
@danimaniac
Feb 18 2017 12:52
Would anyone be willing / able to export the Explorer Board CAD files to a single .STL file? I think some of us trying to design an OpenAPS rig case could utilize the .STL much easier. Do we know if this link has the REVISED Explorer Board design?https://github.com/EnhancedRadioDevices/915MHzEdisonExplorer
lukas-ondriga
@lukas-ondriga
Feb 18 2017 13:57
Hi all, I found one interesting issue: when you go back in Bolus Wizzard it writes down the already entered data several times. I entered 60g of carbs, then I selected combined bolus, went back and selected normal bolus and at the end I had 2 entries in pump history and instead of 60g of carbs, 120g of carbs... I am not able to remove this from pump, so I just have to wait until all the carbs are virtualy absorbed and then it will be possible to use the rig again...
Darrell Wright
@beached
Feb 18 2017 16:17
@danimaniac If you cannot get that, do you have a set of calipers? They can be found for under $30 Canadian and will let you get good measurements
JELCRAWFORD
@JELCRAWFORD
Feb 18 2017 16:17
@lukas-ondriga I think this is a known issue, though I'm not sure what the resolution is - what pump model are you using? I believe that some pumps, when you enter into bolus wizard, even if the treatment recommendation in BW is not enacted, will make an entry in the pump and thus in NS/OpenAPS. I think @Kdisimone found this also, but I'm not sure the resolution.
@tim2000s can you share the specifics on the hot button setup for temp targets - we have just set up hot buttons, and I'd love to control temp targets from there too...
katie disimone
@Kdisimone
Feb 18 2017 16:23

@lukas-ondriga yes...basically when you enter a carb, you enter a carb in the pump history. Backing out of a bolus in bolus wizard can yield a double entry. We've also had a smaller set of instances where we've seen double cob even when NOT backed out of bolus wizard. Still working on solution. I will say though, that this seems to happen very rarely for us.

openaps/oref0#387

John Benjamin
@Bender1061
Feb 18 2017 16:42
so something that I've been using to keep my Edison and explorer separated from the battery is https://www.amazon.com/gp/product/B009IILEVY/ref=oh_aui_search_detailpage?ie=UTF8&psc=1 I heat shrink this on the Edison and explorer and then use a knife to cut out the holes for the micro usb. this works pretty good then I can use a rubber band around it and the battery, I also keep it in one of these boxes https://www.amazon.com/gp/product/B019GFLNV4/ref=oh_aui_search_detailpage?ie=UTF8&psc=1 I have to open it up to charge it, and it's a little bigger than it needs to be, but it's been working real well for me for the time being, well until one of you comes out with a perfect 3D printed box and I'll go to that, but the heat shrink have been a great protector for the whole assembly and I've been using that since December.
JELCRAWFORD
@JELCRAWFORD
Feb 18 2017 16:47
@Kdisimone @lukas-ondriga we've also had difficulty with errant COB - @scottleibrand has done a change to sorting of treatments. We are testing this week and then I believe, presuming things continue to go well, he will merge it with dev. Thus far this week, we have not seen the large COB numbers that were a result of the sorting issue and we have not found any other difficulty from the change. I'm not sure if that change will resolve your other troubles @Kdisimone
@Bender1061 that's an interesting solution - I assume it also secures your wires in place? We've had numerous wire issues - particularly with our wixel set up to transmit dex data to x-drip - I wonder if this would make it more robust....
John Benjamin
@Bender1061
Feb 18 2017 16:50
heat shrink does wonders for keeping wires secure. and with it bouncing around all day in a pocket or bag anything could help
JELCRAWFORD
@JELCRAWFORD
Feb 18 2017 16:51
yes, bouncing around in the belt of a 10yo, even inside the tic-tac box does not readily leave wires attached for long periods of time :)
John Benjamin
@Bender1061
Feb 18 2017 16:53
but that layer of protection, cause you don't want to puncture your lipo battery with the explorer board, does help a lot in that side too
Scorponok
@Scorponok
Feb 18 2017 17:09
@danimaniac Looking good! Let me know when you have it "finished" and I will get a friend of mine to 3d print it so we can give it a try
peterfulvi
@peterfulvi
Feb 18 2017 17:49
Someone help me with this one"reason":"BG data is too old (it's probably this), or clock set incorrectly. The last BG data was read at Sat Feb 18 2017 08:33:00 GMT-0500 (EST) but your system time currently is Sat Feb 18 2017 12:47:57 GMT-0500 (EST)"}
Waiting for ntpd to synchronize... No!
ntpd did not synchronize.
Restarting ntp (via systemctl): ntp.service.
Waiting for ntpd to synchronize... No!
ntpd did not synchronize.
Setting system time to later of Sat Feb 18 12:48:11 EST 2017 or "2017-02-18T12:47:17-05:00":
(date +%s; date -d 2017-02-18T12:47:17-05:00 +%s) | sort -g | tail -1 | while read line; do sudo date -s @$line; done;
Sat Feb 18 12:48:11 EST 2017
lukas-ondriga
@lukas-ondriga
Feb 18 2017 18:21
If you are interested in setting temporary targets without internet connection I prepared this README: https://github.com/lukas-ondriga/openaps-share/blob/master/README.md
Tim Street
@tim2000s
Feb 18 2017 18:22
@JELCRAWFORD read @lukas-ondriga readme.
Works nicely, but you'll need to disable wifi on Edison as you can no longer use NS to set temp targets when using hot buttons.
lukas-ondriga
@lukas-ondriga
Feb 18 2017 18:27
@tim2000s you do not need to disable wifi, you just need to remove "openaps ns-temptargets" from ns-loop, everything else from nightscout can still work
Tim Street
@tim2000s
Feb 18 2017 18:28
Yes, but if you don't disable wifi, you can't post to your rig. When i flip from bt to wifi on the rig, I can't get to it, as DNS doesn't work on andreoid
and I have to use IP addressing.
lukas-ondriga
@lukas-ondriga
Feb 18 2017 18:32
yes, you are right
Scorponok
@Scorponok
Feb 18 2017 18:36
@peterfulvi "BG data is too old" means one of two things when I see it:
1) Your CGM isn't sending updated data to your pump (or wherever)
2) You need to calibrate your pump to your CGM
@peterfulvi Do I remember right that you're using a medtronic CGM? Normally that happens when the pump decides it's just not going to tell you any new BG data until you calibrate it
peterfulvi
@peterfulvi
Feb 18 2017 18:38
How do you disable wifi. I'm doing this to night. I want to access my rig and send temp commands to it by Bluetooth tethering. I'm working tonight on that. My rig just ironed itself out with that issued
Tim Street
@tim2000s
Feb 18 2017 18:39
@Scorponok It can also mean, when using MDT CGM, that the data retrieval cycle hasn't run correctly for >10mins so the glucose.json file is out of date.
@peterfulvi You just comment out all the entries in the wpa_supplicant file.
peterfulvi
@peterfulvi
Feb 18 2017 18:48
Thanks for the help. Far as the Bluetooth I never installed ns so there is nothing I have to disable right?
jmcrawford
@jmcrawford
Feb 18 2017 18:53
@Bender1061 What does that heat shrink do to your heat dissipation?
John Benjamin
@Bender1061
Feb 18 2017 18:56
I've never had a single issue with heat on any of my 4 explorer boards. Granted because I take it out of the case when I charge, maybe that's helping. but the heat shrink has not caused any issues for me.
John Benjamin
@Bender1061
Feb 18 2017 19:11
Heat-shrink also has the use of me not touching any of the components on the board, less change of static discharge burning out any components. currently I'm rolling around on a wheelchair, and I'm about to attach a chain to it to drag along the floor cause I'm discharging against anything I touch. Yes the Dog and the Wife are not happy about that!
RBJervey
@RBJervey
Feb 18 2017 20:43
Hi guys,

There we go. I'm a little stuck. I lost the ability to ssh into my edison and somehow in trying to fix that I broke the wifi entirely. Now when I run ifup wlan0 I get this error:

wpa_supplicant: /sbin/wpa_supplicant daemon failed to start
run-parts: /etc/network/if-pre-up.d/wpasupplicant exited with return code 1
Failed to bring up wlan0.

Anyone have any clue?

jmcrawford
@jmcrawford
Feb 18 2017 20:52
Any errors in /etc/wpa_supplicant/wpa_supplicant.conf?
RBJervey
@RBJervey
Feb 18 2017 20:56

@jmcrawford that file currently looks like this. I did add the second network (ID's and passwords hidden below) but I don't see any errors.

ctrl_interface=DIR=/var/run/wpa_supplicant GROUP=netdev
update_config=1

network={
ssid="XXXX-XXXX-2.4"
psk="password"
}

network={
ssid="XX-XXX-01"
psk="password"
}

And it did connect to the second network before it stopped connecting to anything because it uploaded data to my Heroku website.
jmcrawford
@jmcrawford
Feb 18 2017 21:36
I don't see anything obviously wrong, and if it connected, that's not the problem. Will a reboot fix it?
Paul Dickens
@thebookins
Feb 18 2017 21:38
@peterfulvi @beached @danimaniac I have made some progress on a case design that I plan to print tomorrow. I would love any critique, particularly on the amount of venting required. (@danimaniac has included a lot more). The STL files are at https://github.com/thebookins/openaps-case.
exploded.PNG
Also, @danimaniac I tried to convert the CAD in https://github.com/EnhancedRadioDevices/915MHzEdisonExplorer to STL, but it didn't import properly. Do you know which file (if any) represents the complete assembly? I also spent some time with some calipers measuring the board - was thinking of converting this to a simple CAD.
jmcrawford
@jmcrawford
Feb 18 2017 21:45
@thebookins Looks good to me, but I'm a few days away from my printer to try it out.
RBJervey
@RBJervey
Feb 18 2017 21:51
@jmcrawford I've rebooted and restarted the wifi card and I'm not sure what else to try.
Scott Leibrand
@scottleibrand
Feb 18 2017 21:58
Maybe delete your interfaces and wpa_supplicant files and re-do them?
RBJervey
@RBJervey
Feb 18 2017 22:26
Is there a copy of the files as they should be? I'm not really sure what I'm doing here and I've just been editing the ones we start with. I'd hate to delete them and miss entire lines of code I'd need.
katie disimone
@Kdisimone
Feb 18 2017 22:38
@RBJervey my supplicant file looks similar to yours except I don't have the update_config=1 line.
And I can't tell if your GROUP=netdev is still on continuous line with the previous line, but on mine it is with a space after the wpa_supplicant