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

12th
May 2017
Brian Rabinovitz
@bdr1177
May 12 2017 00:40
I received the following error. Does it mean I'm supposed to find the file '/root/natopenaps/.git/index.lock' and delete it?
"916.540", 5, -60 Starting pump-loop at Thu May 11 20:33:00 EDT 2017:
Radio ok, Listening: .Starting pump-loop at Thu May 11 20:35:14 EDT 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Old pumphistory:
Error, retrying
Listening: .No pump comms detected from other rigs
mmtune: pump://JSON/mmtune/monitor/mmtune.json
reporting monitor/mmtune.json
Traceback (most recent call last):
  File "/usr/local/bin/openaps-report", line 82, in <module>
    app( )
  File "/usr/local/lib/python2.7/dist-packages/openaps/cli/__init__.py", line 51, in __call__
    self.run(self.args)
  File "/usr/local/bin/openaps-report", line 75, in run
    output = app(args, self)
  File "/usr/local/lib/python2.7/dist-packages/openaps/cli/subcommand.py", line 52, in __call__
    return self.method.main(args, app)
  File "/usr/local/lib/python2.7/dist-packages/openaps/reports/invoke.py", line 50, in main
    repo.git.add([report.name], write_extension_data=False)
  File "/usr/local/lib/python2.7/dist-packages/git/cmd.py", line 424, in <lambda>
    return lambda *args, **kwargs: self._call_process(name, *args, **kwargs)
  File "/usr/local/lib/python2.7/dist-packages/git/cmd.py", line 873, in _call_process
    return self.execute(call, **_kwargs)
  File "/usr/local/lib/python2.7/dist-packages/git/cmd.py", line 687, in execute
    raise GitCommandError(command, status, stderr_value, stdout_value)
git.exc.GitCommandError: Cmd('git') failed due to: exit code(128)
  cmdline: git add monitor/mmtune.json
  stderr: 'fatal: Unable to create '/root/natopenaps/.git/index.lock': File exists.

If no other git process is currently running, this probably means a
git process crashed in this repository earlier. Make sure no other git
process is running and remove the file manually to continue.'
Brian Rabinovitz
@bdr1177
May 12 2017 00:55
So, the answer to my own question was apparently, yes. Created a backup of the file and then deleted the original. Rebooted and now we're looping again.
rsilvers129
@rsilvers129
May 12 2017 01:06
@elodaille01 “UAM” is a way to take care of carbs that you are absorbing but didn’t bolus for. For intentional non-bolusing, the lower the carb of the meal, the better it can work. But I intend to use it for normal meals when I simply didn’t bolus enough for the amount of carb that I had. To announce a meal, you just bolus on the pump with Bolus Wizard (either a full bolus, or program in a higher number for your carb ratio into the pump than you really have to get about 2/3 or 3/4 of a bolus). Then set carbratio_adjustmentratio in preferenses.json to tell OpenAPS that you really are a different ratio. If you want to log carbs without trying for insulin… well - I think you would only do that if you were low. In that case, I would log it as a low-treatment with an ITTT HTML POST. http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-4/ifttt-integration.html
Brian Rabinovitz
@bdr1177
May 12 2017 01:22
I tried to reinstall the master branch with curl -s https://raw.githubusercontent.com/openaps/docs/master/scripts/quick-packages.sh | bash - I expected this would install openaps 0.1.5, but instead it says openaps 0.2.0-dev Is this right?
Dana Lewis
@danamlewis
May 12 2017 01:22
fine
tepidjuice
@tepidjuice
May 12 2017 01:30
@scottleibrand just letting you know I've set "remainingCarbsFraction": 1 in preference.json after uncommenting line 304 in determine-basal.js and line 30 in index.js.
tepidjuice
@tepidjuice
May 12 2017 02:16
My pump just had an A52 error (firmware 2.4A, model 722) could this be due to openaps?
tepidjuice
@tepidjuice
May 12 2017 02:25
I think it may have been when SMB tried to bolus at the same time I tried to do a manual bolus. But this is just conjecture.
Gina
@gclyon
May 12 2017 03:00
So, I noticed Loop wasn't enacting any temps today and just now looked into it.. loop was running but no temps. I noticed that my iphone had lost the BT connection to the G5 Dexcom app. Got that back going and viola, temps again. Question: I thought loop was getting BG's from Dexcom share servers? And therefore should pick em up even from G5 receiver sending to share servers? What is actually taking place?
Pete Schwamb
@ps2
May 12 2017 03:04
@gclyon Let’s take this over to the Loop group.
Scott Leibrand
@scottleibrand
May 12 2017 03:42
@geoffro10 your log says it's not SMB'ing because your pump and rig time are mismatched.
@tepidjuice yes, that can happen. Or your manual bolus might not go through and you might think it did due to the SMB buzz/beep. We disable SMBs on temp targets to minimize the chances of an SMB conflicting with a manual bolus.
tepidjuice
@tepidjuice
May 12 2017 03:44
@scottleibrand ok good to know. So long as it's not the pump dying :)
Scott Leibrand
@scottleibrand
May 12 2017 03:45
Could be that too. Dunno.
tepidjuice
@tepidjuice
May 12 2017 03:45
I guess I'll find out.
Dana Lewis
@danamlewis
May 12 2017 03:47
@tepidjuice were you doing easy bolus or bolus wizard during that time?
tepidjuice
@tepidjuice
May 12 2017 04:13
Yes. I think it was openAPS trying to administer an SMB while I was delivering a correction bolus.
Dana Lewis
@danamlewis
May 12 2017 04:15
but HOW were you doing the bolus - via easy bolus buttons, or utilizing the bolus wizard menu to get to the bolus screen?
tepidjuice
@tepidjuice
May 12 2017 04:16
via the bolus wizard menu
Dana Lewis
@danamlewis
May 12 2017 04:17
k. that correlates w/ a @Kdisimone experience of seeing one of those errors from bolus wizard menu. I only use easy bolus, so what happens is it buzzes mid-my other buzzes and just won’t let me hit “act” and deliver a bolus - haven’t gotten the error, which I think is what happens when you’re in bolus wizard and it SMB’s on you.
katie disimone
@Kdisimone
May 12 2017 04:18
Yup. We had that twice
tepidjuice
@tepidjuice
May 12 2017 04:18
how do you enter carbs if you only use easy bolus?
Dana Lewis
@danamlewis
May 12 2017 04:19
NS (via IFTTT and DIYPS)
tepidjuice
@tepidjuice
May 12 2017 04:19
Is there an offline ssh script to enter carbs like there is for temp targets?
Dana Lewis
@danamlewis
May 12 2017 04:19
are you iphone or android?
tepidjuice
@tepidjuice
May 12 2017 04:20
android.
I use a modified version of the temp target script via hotbutton for eating soon and exercise.
Dana Lewis
@danamlewis
May 12 2017 04:20
:+1: not sure since I don’t use android but seems plausible
katie disimone
@Kdisimone
May 12 2017 04:22
Just spitballing but...could do carbs in pump with zero bolus
tepidjuice
@tepidjuice
May 12 2017 04:24
Whenever I do a 0 unit bolus it doesn't appear in openAPS. I guess because the carbs don't get entered into the pumps history unless there's a non-zero bolus.
Pete Schwamb
@ps2
May 12 2017 04:25
The A52 error will happen when a bolus is attempted over RF and the user is in the bolus screen. We’ve been able to reproduce that fairly reliably. It appears to crash the pump.
Dana Lewis
@danamlewis
May 12 2017 04:29
@Kdisimone needs at least a .1 bolus for some models to save the carb entry
@ps2 :+1:
katie disimone
@Kdisimone
May 12 2017 04:30
Ahhhh
Dana Lewis
@danamlewis
May 12 2017 04:30
@johnnymule did you ever confirm - are they showing up in OpenAPS pill?
And do you have AMA turned on?
sanderlong
@sanderlong
May 12 2017 05:25
I'm troubleshooting my out of space issue. Is someone able to confirm if either of these files can be deleted? 297M ./.rootfs/usr/share or 322M ./.rootfs/usr/lib
Tim Street
@tim2000s
May 12 2017 05:26
@sanderlong you'd best not delete those.
Check your .git files.
@ps2 Yes, I've recreated that multiple times too. Stopped using bolus wizard as a result.
sanderlong
@sanderlong
May 12 2017 05:29
@tim2000s 8.0K ./root/myopenaps/.git Is there another one to check?
Tim Street
@tim2000s
May 12 2017 05:30
The one in /tmp
sanderlong
@sanderlong
May 12 2017 05:32
4.0K
I've just set up and am encountering the error
Is there a way to just start over with everything? I've been troubleshooting for days. :(
katie disimone
@Kdisimone
May 12 2017 05:37
You can reflash. Or remove myopenaps and src directories. Either of those are "starting over". One just takes you ALL the way back
sanderlong
@sanderlong
May 12 2017 05:45
Thanks, @Kdisimone. I'll try removing the directories. I definitely don't want to reflash - that was painful!
Tim Street
@tim2000s
May 12 2017 06:05
When you do 'df -h' what's the output?
@tepidjuice You can use hotbuttons to enter carbs in set sizes but you have to disable download from NS.
gkreiner
@gkreiner
May 12 2017 10:41
I'm getting started with OpenAPS. I have a pre-flashed edison + explorer board. Where should I begin in the documentation? When I plug in the rig, my computer doesn't detect all the drivers. I wasn't sure if I should install the drivers listed, because these instructions fall under flashing a new board. Can anyone direct me as to where to begin? Thanks in advance!
tepidjuice
@tepidjuice
May 12 2017 11:49
@tim2000s what's the hotbutton command for that?
i.e. ssh command
elodaille01
@elodaille01
May 12 2017 12:33
@Kdisimone is that normal that the 'maker key' showed for the example (in IFTTT doc) is not the same as the one that is typed in the value field of (maker_key) ?
blob
thank you
and 2/ should I see something different on my NS when IFTTT has been fully set up ? (even when I don't use any 'eating soon' or 'activity' mode)
Matt Walker
@walker0
May 12 2017 12:43
@elodaille01 no they should be the same, those will just be different as @Kdisimone will have used a different account for taking screenshots. 2. No nothing different will appear until you use one of the actions.
elodaille01
@elodaille01
May 12 2017 12:46
ok thank you @walker0 !! When I 'm testing the eating soon, I don't see any temp target appearing on my graph..is it possible that something has been missing ? (I didn't guess what)
kenstack
@kenstack
May 12 2017 12:47
@tepidjuice @danamlewis @bustavo and I are very familiar with that error since we use remote boluses and the buttons. If you are in process of issuing button commands while you send a remote bolus the pump errors and often reboots (a52). What I have found is that pump history is fine, but @bustavo i think has issues with the bg tables after that from the enlite (I use dexcom so it's not an issue for me). I haven't found any work around for that. I do check to see pump state right before I send remote boluses and if it's not normal I hold off i.e. It's bolusing, but Murphy's law seems to always be in play and the kid boluses at the same time your loop sends a remote bolus. Happens to us probably once every other month - pump reboots you reset time and you are back in business. @bustavo can give you more detail on the mdt cgm history
tepidjuice
@tepidjuice
May 12 2017 12:49
I use xdrip + g4 so no issues with cgm data as it's independent of openaps and the pump. Thanks for the information though.
kenstack
@kenstack
May 12 2017 12:51
@tepidjuice yes good. Again I have never seen a case where pump history gets corrupted and I've seen no damage to the pump as it's happened many many times over the years. But it's a pain to reset the pump - always happens at the worst time :)
Matt Walker
@walker0
May 12 2017 12:52
@elodaille01 go through the steps again carefully, check that you have added maker to enable and that the nightscout address is correct for the iftt recipes
elodaille01
@elodaille01
May 12 2017 12:55
all is correct yes..checked several times..'eating soon' pushed give a 'success' message on my Pebble and yes my address has been correctly entered..thank you @walker0 ;)
I will try again ;)
David
@davidkeddydb
May 12 2017 13:03
@danamlewis Just wanted to let you know that Autotune for whatever reason still doesn't output correctly to the table after re-enabling it. Not sure if this has happened to anyone else. I have since removed it but I wanted to at least show you and @scottleibrand the output.
Screenshot 2017-05-12 08.59.57.png
katie disimone
@Kdisimone
May 12 2017 13:10
@walker0 @elodaille01 oh good catch. I had to redo one of those screenshots when ifttt updated their site and I did use a different account when I redid the screenshot. Didn't notice that they weren't the same anymore. I'll fix that for clarity! Good catch!
elodaille01
@elodaille01
May 12 2017 13:15
;)
elodaille01
@elodaille01
May 12 2017 13:46
@Kdisimone instead of 'widget settings' here, I've got 'see all widgets' (android phone)
blob
katie disimone
@Kdisimone
May 12 2017 13:47
I don't have an android...
elodaille01
@elodaille01
May 12 2017 13:48
that is just fyi ;)
katie disimone
@Kdisimone
May 12 2017 15:39
@scottleibrand is there a hard code for 0.4 being the largest smb? I'm thinking there is, right? That's the largest smb I ever see. So it's 0.4 at any single smb. And the 1/3 max of insulin required? Is the max single smb the minimum (0.4 , 1/3 max insulinrequired , max iob)?
Scott Leibrand
@scottleibrand
May 12 2017 15:40
The maxBolus is 30m of (current) basal.
Otherwise yes, the minimum of those three.
So for a basal of 0.8U/hr, SMB would max out at 0.4U SMBs when insulinReq is 1.2U or higher (and basalIOB < maxIOB), and any bolus larger than 0.4U would be considered a manual bolus.
katie disimone
@Kdisimone
May 12 2017 15:42
That makes sense. We are usually around 0.8 to 0.9 basal. Thanks!
Scott Leibrand
@scottleibrand
May 12 2017 15:45
@davidkeddydb I would look as the code and the data files to see why you're getting that null invalid number error.
cameronrenwick
@cameronrenwick
May 12 2017 15:59
currently I have 1 basal setting only (previous amimas pump allowed for 4 settings but with the settings on my 522 they all rounded to the same). I'm going to run autotune but am wondering if this single basal will have a negative effect?
Dana Lewis
@danamlewis
May 12 2017 16:00
autotune is probably going to recommend some differing values throughout the day
run it & see what it says; you can decide how many you want to program & keep tuning from there
cameronrenwick
@cameronrenwick
May 12 2017 16:02
that's kinda what I figured would happen...which I'm happy with for sure... looking forward to see the results
Dana Lewis
@danamlewis
May 12 2017 16:02
:+1:
cameronrenwick
@cameronrenwick
May 12 2017 16:17
just to confirm @danamlewis for carb ratio and ISF these entries are only numbers (in my case carbs = 20; ISF is 79.2) no other info is inputted
Dana Lewis
@danamlewis
May 12 2017 16:18
@cameronrenwick how are you trying to run it? you’ve got a working rig, right?
I would do phase A, so just copy your existing profile per the directions - shouldn’t have to change anything (unless you normally have multiple carb/ISF and want to average/pick others to run with)
cameronrenwick
@cameronrenwick
May 12 2017 16:19
yes I want to run it iteratively... my rig is running very well +!
oh so do a one-off first then later run iteratively you're suggesting?
Dana Lewis
@danamlewis
May 12 2017 16:20
well, if you want it to run everynight just add autotune to your runagain and run it again :)
if you don’t want it going every night as part of your rig, phase A shows you how to run it as a one-off manually but pulling the info from your rig: http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-4/autotune.html#phase-a-running-autotune-in-manual-mode-on-the-command-line
cameronrenwick
@cameronrenwick
May 12 2017 16:22
I was somewhat leaning to running each night but if you feel it's better to start with simply doing a phase A I'll go on your advice
Dana Lewis
@danamlewis
May 12 2017 16:22
not recommending that - was just trying to answer your question as I thought you were asking :)
so if you are comfy running each night, just add autotune to your enables in the runagain.sh and run it again :) (and re-fix your maxiob after that) and it’ll go tonight!
cameronrenwick
@cameronrenwick
May 12 2017 16:24
sounds good +1 thx
cameronrenwick
@cameronrenwick
May 12 2017 18:03
anyone offer opinion on this for me? seems like my carb ratio is pretty good but my ISF is "off"

Parameter | Current | Autotune

ISF | 79.200 | 109.965
CSF | n/a | 4.752
Carb Ratio | 20.000 | 23.141
00:00 | 0.500 | 0.449
01:00 | | 0.467
02:00 | | 0.499
03:00 | | 0.434
04:00 | | 0.437
05:00 | | 0.412
06:00 | | 0.508
07:00 | | 0.549
08:00 | | 0.565
09:00 | | 0.514
10:00 | | 0.465
11:00 | | 0.510
12:00 | | 0.486
13:00 | | 0.447
14:00 | | 0.391
15:00 | | 0.456
16:00 | | 0.519
17:00 | | 0.600
18:00 | | 0.600
19:00 | | 0.598
20:00 | | 0.584
21:00 | | 0.506
22:00 | | 0.451
23:00 | | 0.409
Dana Lewis
@danamlewis
May 12 2017 18:04
Autotune will tend to do ISF high; so you may need to change it; but definitely wouldn’t change your manual settings all the way up to that
but you may want to think about changing it a little bit, or watching for a day or two and seeing how the loop does with the tuned ISF before deciding how much youw ant to change it on the pump
cameronrenwick
@cameronrenwick
May 12 2017 18:11
thanks @danamlewis ..your comments confirm my suspicions. I think I'll alter my ISF a tad and see how things go.
Scott Leibrand
@scottleibrand
May 12 2017 18:40
also, it's generally best to let autotune run on multiple days/weeks' worth of data before considering any changes. if that can be done iteratively in your loop (where the results feed back into its decisions) even better
Dana Lewis
@danamlewis
May 12 2017 18:44
:+1:
cameronrenwick
@cameronrenwick
May 12 2017 18:48
just so I'm clear, despite the fact that in the above cut/paste of my autotune output it doesn't say basal profile above the 0:00-23:00 times, the values are the basal profile correct?
R. Dennis Steed
@rdsteed
May 12 2017 18:49
This week, the National Institute of Standards and Technology (NIST) issued a new guidance on securing wireless infusion pumps. Information is on their website at https://nccoe.nist.gov/projects/use_cases/medical_devices . It is open for public comment for the next six weeks. I am concerned that it does not seem to address the needs and concerns of the DIY community. I encourage the members of this community to look at the guidance and comment.
Dana Lewis
@danamlewis
May 12 2017 18:50
@cameronrenwick no - values on the right are what it recommend. just means you only have one basal on the left
cameronrenwick
@cameronrenwick
May 12 2017 18:52
understood.. it's just on the example fo the Understanding Autotune docs it actually says "basal profile" under the parameter column and my output doesn't
Dana Lewis
@danamlewis
May 12 2017 18:52
oh, I see what you’re saying
cameronrenwick
@cameronrenwick
May 12 2017 18:52
it's the title that's missing... I understand the values I rec'd
Dana Lewis
@danamlewis
May 12 2017 18:52
yea the format otherwise looks great
:+1:
(the big red flags are if 100% of basals, plus iSF and carb ratio had NO changes recommended at all - usually means something bonked in the processing. that looks good)
Savek-CC
@savek-cc
May 12 2017 18:57
@rdsteed the whole thing looks like a marketing excercise from cisco...
Scott Leibrand
@scottleibrand
May 12 2017 19:01
@rdsteed can you tell whether this guidance would apply to insulin pumps, or if this is limited to the kinds of infusion pumps used in hospitals etc.?
Savek-CC
@savek-cc
May 12 2017 19:02
from the best practice guide it seems to be targeted at hospitals (multiple network layers, firewalls and vpn...)
R. Dennis Steed
@rdsteed
May 12 2017 19:02
@savek-cc I agree that Cisco appears prominently in some of the sample diagrams. However, this guidance is issued by NIST, and will play a role in regulation.
Savek-CC
@savek-cc
May 12 2017 19:03
@rdsteed I tried to read the best practice guide and stopped after the 10th Cisco ASA...
when you look at the companies that provided input, I don't see a single insulin-pump company
Scott Leibrand
@scottleibrand
May 12 2017 19:04
as best I can tell after a quick skim and a few searches of the document, this applies to infusion pumps used in hospitals, not insulin pumps worn by patients
R. Dennis Steed
@rdsteed
May 12 2017 19:05
@scottleibrand I believe it will affect insulin pumps. @savek-cc my concern is that it is targeted at hospitals, but ignores the concerns and rights of individual patients to control medical devices.
Scott Leibrand
@scottleibrand
May 12 2017 19:06
@rdsteed can you point me to the parts of the document that you think apply to devices that are not connected to hospital networks, but instead used in an at-home setting by patients?
R. Dennis Steed
@rdsteed
May 12 2017 19:11
@scottleibrand I'll prepare some more detailed comments. But my main concern is that insulin pumps will be under the general umbrella of this guidance (as well as control of the pumps through the nightscout network), and the total failure to recognize the DIY community as having a stake is my concern.
Scott Leibrand
@scottleibrand
May 12 2017 19:14
yeah. perhaps we should get them to clarify whether this guidance only applies to hospital-type infusion pumps, not FDA-approved insulin pumps etc?
Geoff Aumaugher
@geoffro10
May 12 2017 19:14
@scottleibrand , @danamlewis The pump time was the problem. Reset it to match the rig and everything seems to be working now. Thanks again!!
R. Dennis Steed
@rdsteed
May 12 2017 19:15
@scottleibrand Does it apply to insulin pump? The definition of infusion pump references the FDA definition of website that says, "There are many types of infusion pumps, including large volume, patient-controlled analgesia (PCA), elastomeric, syringe, enteral, and insulin pumps,. Some are designed mainly for stationary use at a patient’s bedside. Others, called ambulatory infusion pumps, are designed to be portable or wearable."
Pete Schwamb
@ps2
May 12 2017 19:19
@rdsteed That actually clarifies that this is probably focused on in-hospital use insulin pumps, which are very different from consumer pumps, even the “ambulatory infusion pumps” are very different.
Imagine a large boxy device on wheeled stand.
Scott Leibrand
@scottleibrand
May 12 2017 19:20
"CADD Solis 2000" is the only ambulatory infusion pump mentioned, and it matches @ps2's description, minus the wheeled stand
Pete Schwamb
@ps2
May 12 2017 19:22
And this is the picture showing that pump attached to the wheeled stand. :) http://bridgedesign.com/wp-content/uploads/2013/08/SolisPole-copy-copy.jpg
Scott Leibrand
@scottleibrand
May 12 2017 19:22
I think a bunch of comments to the effect that "this doesn't apply to insulin pumps, right? please clarify" might be useful. If they want to regulate insulin pumps, they'd need to at least talk to insulin pump manufacturers, even if they don't talk to us...
R. Dennis Steed
@rdsteed
May 12 2017 19:22
@ps2 If the definition says insulin pumps it includes insulin pumps.
Pete Schwamb
@ps2
May 12 2017 19:24
@rdsteed not arguing with that at all; just saying that the document is talking about insulin pumps for a different market, and that the vendors and designers in that market are a different set than those that design the pumps you or I use.
My dad was in the hospital a few weeks ago, and a nurse came in at one point to check his blood sugar. The meter that she used was huge, had a full screen display, scanner for scanning a patient tag, and relayed data into the hospital’s network. There are significantly different considerations in designing for a hospital than designing for an end user out of hospital device.
R. Dennis Steed
@rdsteed
May 12 2017 19:31
As an endocrinologist who has prescribed insulin pumps since 1984, I never found that the device regulations changed when I walked in or out of the hospital.
I'll put together a blog post that better outlines my concerns. In the meantime, I encourage folks to read the info on the NIST website.
Pete Schwamb
@ps2
May 12 2017 19:57
It would be great if we could have some impact on how future consumer insulin pump development was done particularly with regard to security. I think it’s very possible to have secure and patient accessable designs, and with not a lot of work on the part of the manufacturer.
Dana Lewis
@danamlewis
May 12 2017 19:57
:+1:
Ryan Dalisky
@rdalisky
May 12 2017 20:01
Is oref1 available as part of the Dev branch?
Dana Lewis
@danamlewis
May 12 2017 20:03
Yes
Ryan Dalisky
@rdalisky
May 12 2017 20:03
Sweet. If I can figure out my Edison flashing error, I'm gonna give it a shot.
alternateal
@alternateal
May 12 2017 20:13
@scottleibrand - on #462 what changed in mdt-cgm as I am still getting the syntax error??
johnnymule
@johnnymule
May 12 2017 20:15
Reflashed my edison to Jubilinux 0.1.1 and now it can't communicate with pump at all.
mmeowlink.exceptions.CommsException: No acknowledgement from pump on wakeup. Is it out of range or is the battery too low?
Scott Leibrand
@scottleibrand
May 12 2017 20:16
alternateal
@alternateal
May 12 2017 20:16
@scottleibrand yes those files
Scott Leibrand
@scottleibrand
May 12 2017 20:17
what line of what file is the syntax error complaining about?
alternateal
@alternateal
May 12 2017 20:18
/usr/local/bin/oref0-pump-loop: line 389: syntax error near unexpected token `else'
Scott Leibrand
@scottleibrand
May 12 2017 20:20
alternateal
@alternateal
May 12 2017 20:21
yes I did
Marcus Lindfors
@Mackwe
May 12 2017 20:33
Newbie question: I've ordered (bot not gotten) the Edison parts needed and I am just curious.. Would there be a spare USB port available somewhere for connecting, let's say.. A Wixel?
johnnymule
@johnnymule
May 12 2017 20:34
Micro USB
Scott Leibrand
@scottleibrand
May 12 2017 20:35
@alternateal ok, I can repro that: lemme debug
Marcus Lindfors
@Mackwe
May 12 2017 20:36
@johnnymule Micro USB as host? Then my rig just shrunk a little bit more. =) Using G4, a built-in Parakeet would be awesome!
johnnymule
@johnnymule
May 12 2017 20:37
You'll need an OTG cable for the edison to act as host
Scott Leibrand
@scottleibrand
May 12 2017 20:38
@alternateal @ericfeibelman wow, three syntax errors. sorry about that.
pushed fixes for all those: please try another pull?
alternateal
@alternateal
May 12 2017 20:38
@scottleibrand - trying now :-)
Marcus Lindfors
@Mackwe
May 12 2017 20:40
@johnnymule But it would be possible with a USB OTG? This I am going to try! Thanks!
johnnymule
@johnnymule
May 12 2017 20:44
I haven't tried it personally, but that's my understanding.
elodaille01
@elodaille01
May 12 2017 20:52
Good evening @danamlewis you advised me to install autotune, which I planned to do right now ... but I have several carb ratios .. is it still a good idea for me to do it in this case?
johnnymule
@johnnymule
May 12 2017 21:01
I'm not seeing my Bolus Wizard carbs in pumphistory.json. Is the 512 even capable of transmitting it, or am I barking up the wrong tree?
alternateal
@alternateal
May 12 2017 21:02
@scottleibrand @ericfeibelman - that seems to work and I have seen several SMB loops complete - thank you - I am seeing another issue though which may or may not be related - never see it before in the normal pump-loop MDT cgm data retrieved MDT cgm data reformatted Profile less than 60m old. jq: error: null and number cannot be divided date: invalid date ‘@’ pump_loop_completed less than 5m ago. RefreshWarning: clock input Invalid Date is unzoned; please pass clock-zoned.json instead
Dana Lewis
@danamlewis
May 12 2017 21:08
@johnnymule you'll need to add profile settings manually and tweak alias for 512
So wonder if something got missed, or if you're first person to 512 since our last setup script changes
johnnymule
@johnnymule
May 12 2017 21:09
Yeah, I've done all that.
The docs need changing but I'd like to make sure I've got it working before I do a PR.
If it's capable of retrieving the carbs from the bolus wizard, I'd like to iron that out. If it's a limitation of the 512, I'll just input them via NS.
Scott Leibrand
@scottleibrand
May 12 2017 21:13
@alternateal sounds like y'all have some more debugging to do to figure those out
alternateal
@alternateal
May 12 2017 21:17
@scottleibrand :+1:
rsilvers129
@rsilvers129
May 12 2017 21:27
What is the deal with official G5 and Android? I see it not listed for the USA, but listed for “international” use. What is to stop someone in the US from running the international app? Would it detect your region and not connect to the Clarity server
?
Dana Lewis
@danamlewis
May 12 2017 21:48
@johnnymule :+1: I’m not sure if it’s a limitation of the 512. I’ll try to dig up who else has been using a 512 and see if we can confirm either way, or have ideas on what to test
Dana Lewis
@danamlewis
May 12 2017 22:02
@thebookins you’re not still using an x12, are you?
elodaille01
@elodaille01
May 12 2017 22:17
@danamlewis Hi..I have 4 carbs ratio in a day..is it a contraindication to use autotune ? thank you ;)
Scott Leibrand
@scottleibrand
May 12 2017 22:19
autotune will only make recommendations for a single average carb ratio. if you want to modify that based on time of day, you'll need to do so manually.
elodaille01
@elodaille01
May 12 2017 22:21
ok Scott, thank you, I had read that...
elodaille01
@elodaille01
May 12 2017 22:26
@scottleibrand I would like to use IFTTT integration, but it's not working for me..Katie checked all my set up to be sure, and nothing seems to be wrong..When I try to test a temporary target, nothing happens on my NS graph..and on my 'eating soon' applet it is writting 'never run'whereas I try several time to run it ..I have no idea about my problem..
johnnymule
@johnnymule
May 12 2017 22:28
Are you doing the button recipe?
elodaille01
@elodaille01
May 12 2017 22:29
ThisButton do you mean ? (on my pebble)
I tried through my pebble and also through my android app IFTTT on my phone (by pressing 'check now')
johnnymule
@johnnymule
May 12 2017 22:30
And both failed?
elodaille01
@elodaille01
May 12 2017 22:30
yes @johnnymule
johnnymule
@johnnymule
May 12 2017 22:30
Did you enter your MAKER_KEY in your Nightscout config variables?
elodaille01
@elodaille01
May 12 2017 22:31
yes !
johnnymule
@johnnymule
May 12 2017 22:31
You also need to have "maker" set under enable
elodaille01
@elodaille01
May 12 2017 22:31
yes it 's done too
all is ok.. @Kdisimone checked it for me
johnnymule
@johnnymule
May 12 2017 22:32
What's the first part of your IFTTT recipe?
johnnymule
@johnnymule
May 12 2017 22:36
What's the event trigger name?
elodaille01
@elodaille01
May 12 2017 22:37
eating_soon and I also tried Eating_Soon because it is written like this in the 'body'
johnnymule
@johnnymule
May 12 2017 22:38
I'm not sure if it's case sensitive or not. I'd stick with "eating_soon"
Then what are the web request values?
elodaille01
@elodaille01
May 12 2017 22:38
not sure too..ok it is like this
Eating Soon
are you talking about the name of my applet, sorry ?
the values are all ok and have been checked ..they are all ok
URL, method, Content type, body..it is ok
johnnymule
@johnnymule
May 12 2017 22:40
The "then that" values.
elodaille01
@elodaille01
May 12 2017 22:41
yes I did it with that link too
johnnymule
@johnnymule
May 12 2017 22:41
The ones for the pebble are different and use the Maker hooks on both ends.
elodaille01
@elodaille01
May 12 2017 22:42
I use Webhooks
johnnymule
@johnnymule
May 12 2017 22:43
If you did it with that link, the first part of your recipe would use the "Button Widget" and not the maker event name like you're saying.
katie disimone
@Kdisimone
May 12 2017 22:43
Wait...@elodaille are you saying you used Maker Webhooks for the first part of you recipe? The IF?? That should be Button, not Maker Webhooks.
johnnymule
@johnnymule
May 12 2017 22:43
He mentioned a pebble though which would be Maker/Maker.
elodaille01
@elodaille01
May 12 2017 22:43
no @Kdisimone it's button for the first part
Maker Webhooks for the second part
katie disimone
@Kdisimone
May 12 2017 22:45
Sounds like we're missing some part of your setup that you did differently than the directions. I'd suggest making an applet (not using pebble yet) and getting that to run properly before moving on to pebble.
elodaille01
@elodaille01
May 12 2017 22:46
I did all once again..exactly as it is written in the docs..
johnnymule
@johnnymule
May 12 2017 22:47
The docs are missing steps.
If you're trying to set up a pebble.
rsilvers129
@rsilvers129
May 12 2017 22:47
Damn. I bought this thinking it was an Edison. Because the photo if of, well, an Edison. http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItemVersion&item=291946312746&view=all&tid=1438035923019
elodaille01
@elodaille01
May 12 2017 22:48
yes I'm trying to set up a pebble but I also try directly with my phone (button/maker, am I right..and it does not work)..you said that I have to set maker/maker for my pebble ?
johnnymule
@johnnymule
May 12 2017 22:48
You can't trigger the IFTTT button widget recipe with the Pebble ThisButton app.
elodaille01
@elodaille01
May 12 2017 22:49
yes that is what I did
and my Pebble announces to me 'success' when I press 'Eating Soon' for example
but nothing happens
johnnymule
@johnnymule
May 12 2017 22:50
It announces success no matter what.
The settings page for ThisButton says that's a bug.

So you have to have in ThisButton:

A) Your IFTTT maker key, from the Maker Settings page.
B) An event name to trigger.

Since you created a "button widget" app, there is no event to trigger.

elodaille01
@elodaille01
May 12 2017 22:52
yes I have all that
and entered the IFTTT maker key
event name is eating_soon for that applet
in my ThisButton on my Pebble app, I have a field 'IFTTT secret key' that I filled
and a part named 'Events' where I entered the name and the event
elodaille01
@elodaille01
May 12 2017 22:58
not it is not a button Widget app
eating_soon is for my ThisButton sorry