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

24th
Jan 2017
jburke41
@jburke41
Jan 24 2017 00:32
Does anyone know what the green light on the Raspberry Pi 3 is for? What is the normal flashing pattern?
RedCyclist1
@RedCyclist1
Jan 24 2017 00:36
@danamlewis Hey there, we tried to flash the Explorer board and we couldn't get it to reboot?? We unplugged it and plugged it back in and the console screen then said we were in rescue mode, we hit control D to continue and nothing happened. We unplugged again and the PuTTY screen is blank. Does this make sense?
Brian Rabinovitz
@bdr1177
Jan 24 2017 02:39
If I want to re-run oref0-setup to select different options, do I just rerun it? Or do I need to do crontab -e and delete everything in the file first? Or do something else?
cdaccarett
@cdaccarett
Jan 24 2017 02:50
Hello i'm trying to run the "Phase C of Autotune" i followed the steps accordingly and have ran into this error. I am using the IOS Loop.
Screen Shot 2017-01-23 at 6.49.34 PM.png
Dana Lewis
@danamlewis
Jan 24 2017 02:51
@bdr1177 what options do you want differently? May be able to change them directly in preferences.json
Brian Rabinovitz
@bdr1177
Jan 24 2017 02:51
Just wanted to add AMA and autosens.
Russell Gardner
@RussGG
Jan 24 2017 02:52
@Aaybob Sorry Aaron. I checked out last night. That looks closer to what I did. I can’t remember exactly. I did forget all my devices after not pairing after a few tries, then redid the order and it paired.
Dana Lewis
@danamlewis
Jan 24 2017 02:53
@bdr1177 for AMA yes easier to run by setup script. Yes for deleting or commenting out existing cron before rerunning, and you'll want to reset your maxiob again after the setup script.
Hey @cdaccarett - was this with 1 day of data?
Russell Gardner
@RussGG
Jan 24 2017 02:54
@danamlewis I didn’t run into that when I was pairing BT.
Dana Lewis
@danamlewis
Jan 24 2017 02:54
@jburke41 General Qs like that about hardware are good to google, since they aren't OpenAPS-specific
@RussGG k, thanks
@tim2000s still stuck? Looks like a time zone complaint but not sure where you'd be getting that from
@kmoran02 24 hours of data, and you ran it for 1 day? At this point, I'd say give it another day or two with you logging data..or if you want, you can pm me your NS URL and can take a look and see anything obvious that might be throwing t off.
cdaccarett
@cdaccarett
Jan 24 2017 02:58
@danamlewis --start-date=2017-01-22 and if i do --start-date=2017-01-23 it just seems to get stuck here //api/v1/treatments.json?find\[created_at\]\[$gte\]=2017-01-22T20:00-08:00&find\[created_at\]\[$lte\]=2017-01-23T00:00-08:00 -rw-rw-r-- 1 parallels parallels 132 Jan 23 18:55 ns-treatments.json
Brian Rabinovitz
@bdr1177
Jan 24 2017 02:58
@danamlewis So I can just go into crontab -e with nano and delete everything in it?
Dana Lewis
@danamlewis
Jan 24 2017 02:59
@cdaccarett hmmm, wonder what's tripping it up..
@bdr1177 yep
Brian Rabinovitz
@bdr1177
Jan 24 2017 02:59
@danamlewis Thanks!
cdaccarett
@cdaccarett
Jan 24 2017 03:00
@danamlewis profiles look good, wouldn't be an issue that im running the IOS Loop right since it pulls the data from my Nightscout profile?
Dana Lewis
@danamlewis
Jan 24 2017 03:00
@cdaccarett we've had several LoopKit/loopers run successfully so far. You put carb and bolus data into NS, right?
cdaccarett
@cdaccarett
Jan 24 2017 03:01
Yeah, directly from the Loop app
found the issue @danamlewis
Dana Lewis
@danamlewis
Jan 24 2017 03:02
What was it?
cdaccarett
@cdaccarett
Jan 24 2017 03:04
instead of : host=https://mynightscout.azurewebsites.net
i did : host=https://mynightscout.azurewebsites.net/
the / made it have // when executed
Dana Lewis
@danamlewis
Jan 24 2017 03:05
Ah, interesting
Might be worth PRing a note to the "why isn't it working" section to make note to not include a / on the URL
cdaccarett
@cdaccarett
Jan 24 2017 03:12
@danamlewis Okay will do
Dana Lewis
@danamlewis
Jan 24 2017 03:20
Thx!
kmoran02
@kmoran02
Jan 24 2017 03:29
@danamlewis thanks! I think there's something wrong, because Chris said when he ran my data it had recommendations. I'll pm you my NS URL.
Dana Lewis
@danamlewis
Jan 24 2017 03:32
Ok. A screenshot of what you're seeing would be good, too, along with showing what command you're using to run it
kmoran02
@kmoran02
Jan 24 2017 03:32
Ok let me hop on my pc
running this
oref0-autotune --dir=~/myopenaps --ns-host=https://kellytype1d.azurewebsites.net --start-date=2017-01-22
and I see this:
Parameter      | Current  | Autotune 
-------------------------------------
ISF            | 70.000   | 0.000    
CSF            | n/a      | 0.000    
Carb Ratio     | 8.000    | 0.000    
Basal Profile  |          |          
  00:00        | 0.1      |          
  00:30        |          |          
  01:00        |          |          
  01:30        |          |          
  02:00        |          |          
  02:30        |          |          
  03:00        |          |          
  03:30        |          |          
  04:00        |          |          
  04:30        |          |          
  05:00        |          |          
  05:30        |          |          
  06:00        |          |          
  06:30        |          |          
  07:00        |          |          
  07:30        |          |          
  08:00        |          |          
  08:30        |          |          
  09:00        |          |          
  09:30        |          |          
  10:00        |          |          
  10:30        |          |          
  11:00        |          |          
  11:30        |          |          
  12:00        |          |          
  12:30        |          |          
  13:00        |          |          
  13:30        |          |          
  14:00        |          |          
  14:30        |          |          
  15:00        |          |          
  15:30        |          |          
  16:00        |          |          
  16:30        |          |          
  17:00        |          |          
  17:30        |          |          
  18:00        |          |          
  18:30        |          |          
  19:00        |          |          
  19:30        |          |          
  20:00        |          |          
  20:30        |          |          
  21:00        |          |          
  21:30        |          |          
  22:00        |          |          
  22:30        |          |          
  23:00        |          |          
  23:30        |          |          
goldenowner@instance-1:~$
Dana Lewis
@danamlewis
Jan 24 2017 03:43
Is you basal really .1?
kmoran02
@kmoran02
Jan 24 2017 03:44
yep. lol! 24/7. but trust me its wrong.
it used to be .025! But 0.1 is as far as we've gotten so far, a tiny bit at a time. Its ridiculous.
Dana Lewis
@danamlewis
Jan 24 2017 03:46
And you've checked all 3 files exist? Profile.json, pump profile.json, and autotune.json?
kmoran02
@kmoran02
Jan 24 2017 03:47
I believe that's what chris had me do this morning.
He was having me paste things he wrote in.
Dana Lewis
@danamlewis
Jan 24 2017 03:48
Do nano preferences.json and copy and paste that here
Sorry
I meant profile.json
nano profile.json
kmoran02
@kmoran02
Jan 24 2017 03:51
it brought me to a black blank page
Dana Lewis
@danamlewis
Jan 24 2017 03:51
Are you in myOpenAPS directory? If not, cd myopenaps first
kmoran02
@kmoran02
Jan 24 2017 03:53
still just a black page

`cd myopenaps
nano profile.json
``

```

Dana Lewis
@danamlewis
Jan 24 2017 03:54
Ls, then, in your myopenaps directory?
kmoran02
@kmoran02
Jan 24 2017 03:54
I don't know what that means. I'm sorry!
cdaccarett
@cdaccarett
Jan 24 2017 03:54
@kmoran02 try this cat ~/myopenaps/settings/profile.json
kmoran02
@kmoran02
Jan 24 2017 03:56

just brought me to the next line
cd myopenaps nano profile.json cat ~/myopenaps/settings/profile.json

``

on the very top is says nano 2.6.5
2.6.3
Dana Lewis
@danamlewis
Jan 24 2017 03:59
Do you remember doing step 3 from the documentation last night?
kmoran02
@kmoran02
Jan 24 2017 03:59
let me look at it
Dana Lewis
@danamlewis
Jan 24 2017 03:59
(Especially 3E and 3F?)
kmoran02
@kmoran02
Jan 24 2017 04:01
I don't think I did either E or F
they don't look familiar
Dana Lewis
@danamlewis
Jan 24 2017 04:01
Heh, that'd do it. Do 3D too if you skipped it, then 3E and 3F
kmoran02
@kmoran02
Jan 24 2017 04:02
ok can you tell me how to get back to where I would enter those steps?
Dana Lewis
@danamlewis
Jan 24 2017 04:02
But make sure you find your original profile.json first and if not, just do all of step 3 over again.
Are you logged into your VM? Where you've been running autotune?
kmoran02
@kmoran02
Jan 24 2017 04:08
I am now
this is what I see:
goldenowner@instance-1:~$ jq . profile.json
jq: error: Could not open file profile.json: No such file or directory
goldenowner@instance-1:~$ mkdir -p ~/myopenaps/settings
goldenowner@instance-1:~$
Dana Lewis
@danamlewis
Jan 24 2017 04:09
Ok. So you may want to ls inside your myopenaps/settings directory to see if they're there already
Ok. So yes do step 3 all over then..
Chris Oattes
@cjo20
Jan 24 2017 04:10
@kmoran02 if a command gives you a blank line and nothing else (no new prompt to type stuff at), press ctrl+c then try again
kmoran02
@kmoran02
Jan 24 2017 04:11
I don't know what ls in myopenaps means.
Chris Oattes
@cjo20
Jan 24 2017 04:11
it means you want your current directory (the thing between : and $) to be myopnaps/settings
and then you run ls
so in this case, you'd run cd ~/myopenaps/settings, then ls
or, alternatively, you run ls ~/myopenaps/settings
Dana Lewis
@danamlewis
Jan 24 2017 04:12
(Anytime you see something in grey, it's a command that someone is suggesting you copy and paste and hit enter)
cjo20 @cjo20 disappears to bed, being 4am
Dana Lewis
@danamlewis
Jan 24 2017 04:14
Ah, and here I thought you were waking up! Night, @cjo20 :smile:
kmoran02
@kmoran02
Jan 24 2017 04:15
goldenowner@instance-1:~$ ls ~/myopenaps/settings
]  ],  {  }  },  autosens_max:  autosens_min:  autotune.json  carb_ratio:  endOffset:  i:  isfProfile:  minutes:  offset:  profile.json  profile.json.save  pumpprofile.json  rate:  sensitivities:  sensitivity:  start:  x:
goldenowner@instance-1:~$
should I do all of step 3 again?
Dana Lewis
@danamlewis
Jan 24 2017 04:17
Start from 3b
kmoran02
@kmoran02
Jan 24 2017 04:17
ok ty!
ok I edit my profile, and it will not let me hit control x. It brings me back to the previous step
kmoran02
@kmoran02
Jan 24 2017 04:22
this part: Make sure to exit the profile.json when done editing this file - Control-X and hit yes to save.
RedCyclist1
@RedCyclist1
Jan 24 2017 04:28
@danamlewis Will try again tomorrow.. looks like lots of stuff going on... :)
Dana Lewis
@danamlewis
Jan 24 2017 04:32
@RedCyclist1 maybe tag @Kdisimone - she just flashed yesterday and might have some ideas, I can't remember if she ran into the exact same thing or not
@kmoran02 "it brings you back to the previous step" previous step of what? Where?
kmoran02
@kmoran02
Jan 24 2017 04:38
`goldenowner@instance-1:~/myopenaps/settings$ nano profile.json goldenowner@instance-1:~/myopenaps/settings$ jq . profile.json parse error: Invalid literal at line 1, column 5 goldenowner@instance-1:~/myopenaps/settings$ jq . profile.json parse error: Invalid literal at line 1, column 5 goldenowner@instance-1:~/myopenaps/settings$

katie disimone
@Kdisimone
Jan 24 2017 04:43
@RedCyclist1 sorry…i didn’t have any problems at the flashing step. Where were you in the steps? I did have a couple places I had to go a little slower…I’d be happy to look for you
kmoran02
@kmoran02
Jan 24 2017 04:43
now its even MORE broken. I officially give up. More tomorrow.
Parameter      | Current  | Autotune 
-------------------------------------
ISF            | 0.000    | 0.000    
CSF            | n/a      | 0.000    
Carb Ratio     | 0.000    | 0.000    
Basal Profile  |          |          
  00:00        |          |          
  00:30        |          |          
  01:00        |          |          
  01:30        |          |          
  02:00        |          |          
  02:30        |          |          
  03:00        |          |          
  03:30        |          |          
  04:00        |          |          
  04:30        |          |          
  05:00        |          |          
  05:30        |          |          
  06:00        |          |          
  06:30        |          |          
  07:00        |          |          
  07:30        |          |          
  08:00        |          |          
  08:30        |          |          
  09:00        |          |          
  09:30        |          |          
  10:00        |          |          
  10:30        |          |          
  11:00        |          |          
  11:30        |          |          
  12:00        |          |          
  12:30        |          |          
  13:00        |          |          
  13:30        |          |          
  14:00        |          |          
  14:30        |          |          
  15:00        |          |          
  15:30        |          |          
  16:00        |          |          
  16:30        |          |          
  17:00        |          |          
  17:30        |          |          
  18:00        |          |          
  18:30        |          |          
  19:00        |          |          
  19:30        |          |          
  20:00        |          |          
  20:30        |          |          
  21:00        |          |          
  21:30        |          |          
  22:00        |          |          
  22:30        |          |          
  23:00        |          |          
  23:30        |          |          
goldenowner@instance-1:~/myopenaps/settings$
```
Aaron Michelson
@Aaybob
Jan 24 2017 04:53
@ejoe132 The flash of edison must be successful before serial port will be working....
Aaron Michelson
@Aaybob
Jan 24 2017 05:00
@Kdisimone I get 6 hours with bluetooth hotspot, to run wifi hotspot I need a charging battery...maybe three hours.
ejoe132
@ejoe132
Jan 24 2017 05:54
@Aaybob thank you for posting those, right now trying to get my pi back up and running as it just stopped working this weekend, keeping getting stuck at installing the tool kit :/ keeps saying "update mozilla certificate authority bundle to 2.9 then has some lines but just stops, so not sure whats happening there. But tyring to follow these videos for the edison my jubilinux files dont have the win32 folder his does so Im deleting them off my computer and gonna try to re download and see, is the second video just a copy of the first?
ejoe132
@ejoe132
Jan 24 2017 06:06
so hit enter a few times on my pi install and it came up with something odd, says "--michael shuler michael@pbandjelly.org Fri, 18 Nov 2016 09:09:47 -0600" I have no clue why it would be giving me some random email?
Sandra Keßler
@SandraK82
Jan 24 2017 06:09
hm finally able to run autotune for nearly two months of data.. completely changes my basal..
Screen Shot 2017-01-24 at 07.07.33.png
Scott Leibrand
@scottleibrand
Jan 24 2017 06:10
Did you disable the 20% cap?
Sandra Keßler
@SandraK82
Jan 24 2017 06:11
no i only reduced the limit for carb Ratio to 2, as i started with 3
Scott Leibrand
@scottleibrand
Jan 24 2017 06:12
Hmm... autosens_max should limit it to setting basals to 1.2x your pump basals
Sandra Keßler
@SandraK82
Jan 24 2017 06:12
nohup /usr/local/bin/oref0-autotune --dir=~/myopenaps --ns-host=http://localhost:1337 --start-date=2016-12-01 > autotune.log &
pulled dev yesterday and changed line 54 oref0-autotune-prep.js to bail out only < 2
Scott Leibrand
@scottleibrand
Jan 24 2017 06:14
Part of the reason for the 20% limit is because, unless you are already using a closed loop (where it essentially is already picking an appropriate basal) autotune can only really tell that "your basals should be higher" - you'll have to run with slightly higher basals for awhile to really be able to tell if they should be raised further than that.
I believe that limit of 3 is because that's as low as MDT can go.
Dana Lewis
@danamlewis
Jan 24 2017 06:14
@SandraK82 what is autosens min and max in your preferences.json?
Sandra Keßler
@SandraK82
Jan 24 2017 06:14
but why is the 20% cap failing?
cat preferences.json 
{
    "max_iob": 10,
    "type": "current",
    "max_daily_safety_multiplier": 3,
    "current_basal_safety_multiplier": 4,
    "autosens_max": 1.7,
    "autosens_min": 0.5,
    "autosens_adjust_targets": true,
    "override_high_target_with_low": false,
    "skip_neutral_temps": true,
    "bolussnooze_dia_divisor": 2,
    "min_5m_carbimpact": 3,
    "carbratio_adjustmentratio": 1
}
Scott Leibrand
@scottleibrand
Jan 24 2017 06:15
Oh, 1.7
Those really should be closer to the defaults of 1.2 and 0.7
Sandra Keßler
@SandraK82
Jan 24 2017 06:16
hm maybe, but they do really fine for me
Scott Leibrand
@scottleibrand
Jan 24 2017 06:16
For autotune I mean, not for autosens
There's also a chance that autotune adjusting the basals too much means that other stuff doesn't get tuned quite right.
Sandra Keßler
@SandraK82
Jan 24 2017 06:18
i wonder about the changed carb Ration, i use 3 and it set it to 5. Carb ratio i did Testing a lot
Scott Leibrand
@scottleibrand
Jan 24 2017 06:18
If you were using autotune in your loop, the limits you have for autosens might end up being ok, since it is an iterative closed loop that will be able to self correct. But for retrospective I would recommend the tighter ones.
3 seems incredibly low.
Medtronic pumps literally will not got below 3
It's the very edge of the physiologically possible range.
Sandra Keßler
@SandraK82
Jan 24 2017 06:19
in Germany i was taugt to use BE as basis for calculations (1 BE = 12grams carbs) and I would give 4 units per BE, therefore 12 / 4 gives me 3 for carb Ratio and that seems to work fine
Dana Lewis
@danamlewis
Jan 24 2017 06:20
If you change your preferences.json to be normal defaults (1.2 and .7), run autotune and get the basal output with the caps, then you can go back and change it in preferences to what you want for realtime looping and autosens.
Sandra Keßler
@SandraK82
Jan 24 2017 06:20
yeah my 522 complains about the setting too
Dana Lewis
@danamlewis
Jan 24 2017 06:21
@ryanpacker similarly, if you ended up running w/out caps would be curious to how far it tuned yours..
Scott Leibrand
@scottleibrand
Jan 24 2017 06:21
Anyway, probably best to re-run autotune and see if it still recommends a slightly higher CR
Sandra Keßler
@SandraK82
Jan 24 2017 06:21
i have no preference.json in teh autotune setup, the autotune setup is on the same server as my ns instance
where should i put the values?
Dana Lewis
@danamlewis
Jan 24 2017 06:22
Ah. See example then in the docs - profile.json
Sandra Keßler
@SandraK82
Jan 24 2017 06:23
ah ok, thanks
Dana Lewis
@danamlewis
Jan 24 2017 06:23
And make sure to recopy to pumpprofile.json and autotune.json after you change
:+1:
Sandra Keßler
@SandraK82
Jan 24 2017 06:26
cat settings/autotune.json 
{
  "min_5m_carbimpact": 3,
  "dia": 4,
  "basalprofile": [
    {
      "i": 0,
      "start": "00:00:00",
      "rate": 1,
      "minutes": 0
    },
    ...
  ],
  "isfProfile": {
    "sensitivities": [
      {
        "i": 0,
        "start": "00:00:00",
        "sensitivity": 30,
        "x": 0,
        "offset": 0,
        "endOffset": 1440
      }
    ],
    "first": 1
  },
  "autosens_max": 1.2,
  "autosens_min": 0.7,
  "carb_ratio": 3
}
Scott Leibrand
@scottleibrand
Jan 24 2017 06:27
another thing that is interesting is that your ISF / CR is 30 mg/dL per U / 3 g/U = 10 mg/dL per gram. Based on that, I would predict your body mass is something like 40kg. :smile:
That is CSF, which tends to be fairly proportional to body mass, because it's basically just a "how much water is this sugar dissolving into" problem.
Sandra Keßler
@SandraK82
Jan 24 2017 06:28
that would be low but great :) sadly my bodyweight is much to high
Scott Leibrand
@scottleibrand
Jan 24 2017 06:30
That json looks right.
Sandra Keßler
@SandraK82
Jan 24 2017 06:31
I had lot of troubles to find a therapist who accepts how my body works in regards to Insulin needs instead of mumbling constantly something about thats abnormaly... :)
just running last two weeks of data, that should be good enough to see if it makes a difference
Dana Lewis
@danamlewis
Jan 24 2017 06:36
:+1:
Sandra Keßler
@SandraK82
Jan 24 2017 06:38
just checket some very old data from 1996 (hand written of course) at that time I had 6U of actrapid / 1 BE (12grams carbs) :/
Scott Leibrand
@scottleibrand
Jan 24 2017 06:39
At least with something like autotune, we can go by what your data says about your body, rather than worrying too much about population averages.
Sandra Keßler
@SandraK82
Jan 24 2017 06:40
:+1:
ah now I see stuff like Limiting hour 0 basal to 1.20 (which is 1.2 * pump basal of 1 ) in the output
Dana Lewis
@danamlewis
Jan 24 2017 06:43
:+1:
Sandra Keßler
@SandraK82
Jan 24 2017 06:47
@scottleibrand can you add a local check to the script? Most installations will run on a default (english) server, mine was preconfigured by my hosting company to german locale so 1.0 is printed as 1,0 default which breaks some of the parsing
Dana Lewis
@danamlewis
Jan 24 2017 06:48
Can you create an issue in the oref0 repo so we can follow up on it?
(Bedtime here :smile: )
Sandra Keßler
@SandraK82
Jan 24 2017 06:56
openaps/oref0#342
have a good night :) day just started for me here
Dana Lewis
@danamlewis
Jan 24 2017 07:00
Thank you!
Sandra Keßler
@SandraK82
Jan 24 2017 07:14
@scottleibrand @danamlewis with safeties applied it looks much better:
Screen Shot 2017-01-24 at 08.13.32.png
Dana Lewis
@danamlewis
Jan 24 2017 07:18
:+1:
Sandra Keßler
@SandraK82
Jan 24 2017 07:21
thank you very much for your Input on this matter!
Dana Lewis
@danamlewis
Jan 24 2017 07:22
:smile: thanks for the feedback, and for logging issues, too!
Tim Street
@tim2000s
Jan 24 2017 07:26
@SandraK82 -your pattern for basal is very similar to mine. 👍
Tim Street
@tim2000s
Jan 24 2017 07:31
@danamlewis Had to go to bed. Think I should probably completely start from scratch on the Pi. It's my first rig, so is about six months out of date and probably needs a whole bunch of updates at OS and framework level.
Dana Lewis
@danamlewis
Jan 24 2017 07:32
:+1:
Scott Leibrand
@scottleibrand
Jan 24 2017 07:35
:+1: @SandraK82 and thanks!
Tim Street
@tim2000s
Jan 24 2017 07:48
@danamlewis if you're running Autotune from OpenAPS, which profile is used?
Tim Street
@tim2000s
Jan 24 2017 07:54
(I think it starts with profile.json then generates profile.pump.json, which is then iteratively updated, I just wasn't clear from the docs!)
tynbendad
@tynbendad
Jan 24 2017 07:58
question about autotune basal hours... are they in GMT or your local timezone? i've been staring at results for a day or two and they didn't make sense, but if they are in GMT then maybe they make more sense...
Roumen Georgiev
@PoweRGbg
Jan 24 2017 08:19
Can someone test autotune on my site to check if my NS data is bad or my local profile is ?
Sandra Keßler
@SandraK82
Jan 24 2017 08:21
@PoweRGbg where is the problem?
Roumen Georgiev
@PoweRGbg
Jan 24 2017 08:21
@SandraK82 It connects and downloads the treatments but no result
@SandraK82 I'm using xdrip+ as a BG uploader also, and as far as I know it's problematic
Tim Street
@tim2000s
Jan 24 2017 08:23
@colinlennon running the updated cron and it's working nicely.
Sandra Keßler
@SandraK82
Jan 24 2017 08:23
@PoweRGbg i use my modified xdrip for uploads to ns and that seems not be problematic
@PoweRGbg do you have Output?
Roumen Georgiev
@PoweRGbg
Jan 24 2017 08:27
none
Matthias
@ceben80
Jan 24 2017 08:37
@colinlennon your xdrip offline documentation looks good to me, thank you :clap:
Colin Lennon
@colinlennon
Jan 24 2017 08:43
@ceben80 Thanks!
@tim2000s Thanks Tim. So does it allow you to switch between pulling CGM data from xDrip and from NS? By turning BT tethering or your WiFi hotspot on/off?
I seem to have some issues with my Edison rig wrt switching between BT tethering and WiFi. I think I'll just stick to using BT tethering 24/7 personally - it's proved to be rock-solid in terms of reliability for me.
Tim Street
@tim2000s
Jan 24 2017 08:57
@colinlennon What I've tested is having old data in the xDrip.db generating "glucose data too old errors" on the old cron, then updating it so that it runs the compare before moving to live data from NS. I can't update xDrip data at the moment as I'm using up Enlite sensors.
I can't get tethering to work at all via BT.
Colin Lennon
@colinlennon
Jan 24 2017 09:01
@tim2000s Okay, thanks. I'll won't implement that code change in the xDripAPS repo yet, I think it needs further testing. I'l leave the documentation as-is for now - with the assumption that people will use xDripAPS as the only CGM data source, 24/7. I think a few other people have also had issues auto-switching between WiFi networks on Jubilinux, need to get that issue cracked first.
Tim Street
@tim2000s
Jan 24 2017 09:23
@colinlennon I've got the "Edison wifi" code and the new oref0-online working now. They both fix my wifi issues. I've had no switching problems in ages (except for the iPhone, but that seems to be a different story). Android hotspots work more or less perfectly.
I've never had tethering to iPhone work though.
(Via BT)
Colin Lennon
@colinlennon
Jan 24 2017 09:26
@tim2000s Thanks Tim - I didn't know that oref0-online had been updated, I'll have a look. :+1:
Tim Street
@tim2000s
Jan 24 2017 10:09
Hum. Don't know what's going on with my Edi, but it seems to have lost the Bluetooth interface.
root@ed209:~# hciconfig hci0 up
Can't get device info: No such device
Tim Street
@tim2000s
Jan 24 2017 10:33
@danamlewis I've mentioned previously about Edison/Explorer board combos seemingly aligning random numbers and interfering with each other. Running my dev board this morning alongside my master, I had a 20 minute period where neither could properly comm with the pump, both seemed to be trying to run at the same time. What seems to happen is they wait for silence, then jump in when they get it. This seems to be in the gap between getting pump history and refreshing. When I turned the dev board off, the master connected immediately.
Tim Street
@tim2000s
Jan 24 2017 10:46
@colinlennon I forgot to mention, that was scenario I was interested in. I was having issues with changing wifi networks and losing a direct link between xDrip and the Edison so needing to revert back to Nightscout. My conclusions from the test today are that that has worked, although I haven't been able to test the inverse, which is restarting xDrip use.
Tim Street
@tim2000s
Jan 24 2017 11:23
Ignore my hci0 issues. I'd managed to muck up bluez installs 😳
StuArnold
@StuArnold
Jan 24 2017 11:25
@PoweRGbg have you switched from dev to autotune-mdi branch. I use xdrip and switched last and now get an output
Roumen Georgiev
@PoweRGbg
Jan 24 2017 11:28
@StuArnold nope still using dev
StuArnold
@StuArnold
Jan 24 2017 11:30
@PoweRGbg if you are on an xdrip and trying to do the option C I think you need to switch. I think the instructions are now in the new section about when it is not working
Roumen Georgiev
@PoweRGbg
Jan 24 2017 11:34
@StuArnold Actually I'm just trying to run autotune from the command line and I'm using xdrip+ as BG source to NS. But for now autotune spits out some strange errors about timezones
StuArnold
@StuArnold
Jan 24 2017 11:38
@PoweRGbg sounds like what I was having before switching. Got a few lines about reading the data then an error message which involved a file in timezone folder and some nul errors - it gave me the table but all values 0. On phone so can not copy my messages but if you scroll back to about 5pm uk last night you should see my screenshot
@PoweRGbg actually screenshot was 18:03 uk time yesterday. Is that the same error?
Roumen Georgiev
@PoweRGbg
Jan 24 2017 11:46
@StuArnold Still trying to find it :)
Tim Street
@tim2000s
Jan 24 2017 11:47
No, even going back to the correct level of bluez I have no hci0 device. Grrrr.
Roumen Georgiev
@PoweRGbg
Jan 24 2017 11:54
@StuArnold Found it - exactly the same error
StuArnold
@StuArnold
Jan 24 2017 12:01
@PoweRGbg so think you need to switch to mdi branch as well
Roumen Georgiev
@PoweRGbg
Jan 24 2017 12:02
did it and runs with no errors but no output also :)
StuArnold
@StuArnold
Jan 24 2017 12:03
Do you have treatments in careportal?
Roumen Georgiev
@PoweRGbg
Jan 24 2017 12:03
as many as you like
Maybe I should try a fresh install of mdi
StuArnold
@StuArnold
Jan 24 2017 12:04
I think there was also a question last night over whether treatments from other devices were in same format as Careportal website entries
Tim Street
@tim2000s
Jan 24 2017 12:10
No, don't ignore my hci0 issues. Somehow, hci0 seems not to exist anymore. Anyone any ideas?
Chris Oattes
@cjo20
Jan 24 2017 12:19
Is Bluetoothd running?
DunePlodder
@DunePlodder
Jan 24 2017 12:22
Somewhat to my surprise I now have Autotune working on VirtualBox/Ubuntu. Never used Linux before - great instructions @danamlewis. I'm MDI & I'm unsure of how to enter my basal rate in my profile. I currently inject 36 units of Levemir in the evening & 18 units in the morning.
scarral
@scarral
Jan 24 2017 12:30
@taylorfowler can you please run autotune on my website and let me know what you get? oref0-autotune --dir=~/myopenaps --ns-host=https://scarralnew.azurewebsites.net --start-date=2017-01-23
I get the following:

Parameter | Current | Autotune

ISF | 225.000 | 225.000
CSF | n/a | 7.500
Carb Ratio | 30.000 | 30.000
Basal Profile | |
00:00 | 0.050 | 0.050
00:30 | |
01:00 | | 0.050
01:30 | |
02:00 | | 0.050
02:30 | |
03:00 | 0.050 | 0.050
03:30 | |
04:00 | 0.100 | 0.100
04:30 | |
05:00 | 0.100 | 0.100
05:30 | |
06:00 | | 0.100
06:30 | |
07:00 | | 0.100
07:30 | |
08:00 | | 0.100
08:30 | |
09:00 | 0.100 | 0.100
09:30 | |
10:00 | | 0.100
10:30 | |
11:00 | | 0.100
11:30 | |
12:00 | | 0.100
12:30 | |
13:00 | | 0.100
13:30 | |
14:00 | | 0.100
14:30 | |
15:00 | | 0.100
15:30 | |
16:00 | | 0.100
16:30 | |
17:00 | 0.175 | 0.175
17:30 | |
18:00 | | 0.175
18:30 | |
19:00 | 0.250 | 0.250
19:30 | |
20:00 | 0.750 | 0.750
20:30 | |
21:00 | 0.550 | 0.550
21:30 | |
22:00 | 0.225 | 0.225
22:30 | |
23:00 | 0.150 | 0.150
23:30 | |
Roumen Georgiev
@PoweRGbg
Jan 24 2017 12:30
@StuArnold It works after the fresh install - Thank you!
Tim Street
@tim2000s
Jan 24 2017 12:34
@cjo20 Yes.
@DunePlodder you need the autotune_mdi install
DunePlodder
@DunePlodder
Jan 24 2017 12:38
@tim2000s Thank you.
scarral
@scarral
Jan 24 2017 12:40
And here's my profile.json:

{
"min_5m_carbimpact": 3,
"dia": 4,
"basalprofile": [
{
"i": 0,
"start": "00:00:00",
"minutes": 0,
"rate": 0.05
},
{
"i": 1,
"start": "03:00:00",
"minutes": 180,
"rate": 0.05
},
{
"i": 2,
"start": "04:00:00",
"minutes": 240,
"rate": 0.1
},
{
"i": 3,
"start": "05:00:00",
"minutes": 300,
"rate": 0.1
},
{
"i": 4,
"start": "09:00:00",
"minutes": 540,
"rate": 0.1
},
{
"i": 5,
"start": "17:00:00",
"minutes": 1020,
"rate": 0.175
},
{
"i": 6,
"start": "19:00:00",
"minutes": 1140,
"rate": 0.25
},
{
"i": 7,
"start": "20:00:00",
"minutes": 1200,
"rate": 0.750
},
{
"i": 8,
"start": "21:00:00",
"minutes": 1260,
"rate": 0.55
},
{
"i": 9,
"start": "22:00:00",
"minutes": 1320,
"rate": 0.225
},
{
"i": 10,
"start": "23:00:00",
"minutes": 1380,
"rate": 0.15
}

],
"isfProfile": {
"sensitivities": [
{
"i": 0,
"start": "00:00:00",
"sensitivity": 225,
"offset": 0,
"x": 0,
"endOffset": 1440
}
]
},
"carb_ratio": 30
}

scarral
@scarral
Jan 24 2017 12:47
I have the feeling that there's something missing in the settings of my azure website. Can someone please help me figure this out?
Chris Oattes
@cjo20
Jan 24 2017 12:55
What do you think is wrong with your data?
Also, when you're pasting stuff like that, please put ``` on a line on it's own at the top, and again on a line on it's own at the bottom. (Note that ` isn't the same as ')
@scarral ^
dougtakeuti
@dougtakeuti
Jan 24 2017 13:01
The loop is not running and shows the following error. Does that mean that git is currupt?
  File "/usr/local/bin/openaps-report", line 6, in <module>
    exec(compile(open(__file__).read(), __file__, 'exec'))
  File "/root/src/openaps/bin/openaps-report", line 82, in <module>
    app( )
  File "/usr/local/lib/python2.7/dist-packages/openaps/cli/__init__.py", line 52                                          , in __call__
    self.epilog( )
  File "/root/src/openaps/bin/openaps-report", line 69, in epilog
    super(ReportToolApp, self).epilog( )
  File "/usr/local/lib/python2.7/dist-packages/openaps/cli/__init__.py", line 75                                          , in epilog
    self.create_git_commit( )
  File "/usr/local/lib/python2.7/dist-packages/openaps/cli/__init__.py", line 98                                          , in create_git_commit
    self.repo.index.commit(msg)
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.1-py2.7.egg/git/ind                                          ex/base.py", line 954, in commit
    author_date=author_date, commit_date=commit_date)
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.1-py2.7.egg/git/obj                                          ects/commit.py", line 393, in create_from_tree
    repo.head.set_commit(new_commit, logmsg=message)
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.1-py2.7.egg/git/ref                                          s/symbolic.py", line 211, in set_commit
    self.set_object(commit, logmsg)
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.1-py2.7.egg/git/ref                                          s/symbolic.py", line 240, in set_object
    return self._get_reference().set_object(object, logmsg)
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.1-py2.7.egg/git/ref                                          s/reference.py", line 64, in set_object
    super(Reference, self).set_object(object, logmsg)
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.1-py2.7.egg/git/ref                                          s/symbolic.py", line 237, in set_object
    return self.set_reference(object, logmsg)
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.1-py2.7.egg/git/ref                                          s/symbolic.py", line 308, in set_reference
    fd = lfd.open(write=True, stream=True)
  File "/usr/local/lib/python2.7/dist-packages/gitdb2-2.0.0-py2.7.egg/gitdb/util                                          .py", line 314, in open
    raise IOError("Lock at %r could not be obtained" % self._lockfilepath())
IOError: Lock at '/root/myaps/.git/refs/heads/master.lock' could not be obtained
Error, retrying
Chris Oattes
@cjo20
Jan 24 2017 13:02
I think that's more likely to mean there's something else trying to use git at the same time
Colin Lennon
@colinlennon
Jan 24 2017 13:02
Yeah, what @cjo20 said. Deleting /root/myaps/.git/refs/heads/master.lock should fix it for now.
Chris Oattes
@cjo20
Jan 24 2017 13:02
I thought the loop tried to reset git if there is corruption anyway
Tim Street
@tim2000s
Jan 24 2017 13:12
Yes, it does.
The oref0-git-reset is setup to run every loop
kmoran02
@kmoran02
Jan 24 2017 13:13
Ok you lovely people. My plan today is to try to figure out how to delete everything and just start over. But I woke up sick so I need time to wake up more. I most DEFINITELY needed to walk away from my computer last night before I lost my mind.
kmoran02
@kmoran02
Jan 24 2017 13:34
since I only have one basal 24/7 right now, would I leave the comma that's there by the end of my basal line when deleteing the rest of the basal spots?
{
  "min_5m_carbimpact": 3,
  "dia": 4,
  "basalprofile": [
    {
      "i": 0,
      "start": "00:00:00",
      "minutes": 0,
      "rate": 0.1
    },
  ],
  "isfProfile": {
    "sensitivities": [
      {
        "i": 0,
        "start": "00:00:00",
        "sensitivity": your_isf,
        "offset": 0,
        "x": 0,
        "endOffset": 1440
      }
    ]
  },
  "carb_ratio": your_ic_ratio,
  "autosens_max": 1.2,
  "autosens_min": 0.7
}
As in leave it just the way you see it, or more to delete?
kmoran02
@kmoran02
Jan 24 2017 13:44
so its telling me there is an error to fix in this. Anyone see anything obvious?
{
  "min_5m_carbimpact": 3,
  "dia": 4,
  "basalprofile": [
    {
      "i": 0,
      "start": "00:00:00",
      "minutes": 0,
      "rate": 0.1
    },
  ]
  "isfProfile": {
    "sensitivities": [
      {
        "i": 0,
        "start": "00:00:00",
        "sensitivity": 70,
        "offset": 0,
        "x": 0,
        "endOffset": 1440
      }
    ]
  },
  "carb_ratio": 8,
  "autosens_max": 1.2,
  "autosens_min": 0.7
}
Roumen Georgiev
@PoweRGbg
Jan 24 2017 13:46
did you try jq . profile.json
kmoran02
@kmoran02
Jan 24 2017 13:48
yes and it says theres an error
goldenowner@instance-2:~/myopenaps/settings$ jq . profile.json 
parse error: Expected another array element at line 11, column 3
goldenowner@instance-2:~/myopenaps/settings$
I tried it wih and without the comma right above "isprofile"
Roumen Georgiev
@PoweRGbg
Jan 24 2017 13:51
try removing the comma at line 10
kmoran02
@kmoran02
Jan 24 2017 13:52
I'm terrible at this...can you be more specific?
after i:0 ?
(the second one)
Chris Oattes
@cjo20
Jan 24 2017 13:53
count 10 lines from the top
what is on that line?
kmoran02
@kmoran02
Jan 24 2017 13:53
do I include brackets as a line?
Chris Oattes
@cjo20
Jan 24 2017 13:54
yes
blank lines too. They're all lines.
kmoran02
@kmoran02
Jan 24 2017 13:54
ok thanks! I'll try that
Bingo!
HOLY CRAP!!! I did it!!!!!!!!!!!!
Parameter      | Current  | Autotune 
-------------------------------------
ISF            | 70.000   | 70.000   
CSF            | n/a      | 7.312    
Carb Ratio     | 8.000    | 9.573    
Basal Profile  |          |          
  00:00        | 0.100    | 0.099    
  00:30        |          |          
  01:00        |          | 0.100    
  01:30        |          |          
  02:00        |          | 0.100    
  02:30        |          |          
  03:00        |          | 0.100    
  03:30        |          |          
  04:00        |          | 0.100    
  04:30        |          |          
  05:00        |          | 0.100    
  05:30        |          |          
  06:00        |          | 0.100    
  06:30        |          |          
  07:00        |          | 0.100    
  07:30        |          |          
  08:00        |          | 0.100    
  08:30        |          |          
  09:00        |          | 0.100    
  09:30        |          |          
  10:00        |          | 0.100    
  10:30        |          |          
  11:00        |          | 0.100    
  11:30        |          |          
  12:00        |          | 0.100    
  12:30        |          |          
  13:00        |          | 0.100    
  13:30        |          |          
  14:00        |          | 0.100    
  14:30        |          |          
  15:00        |          | 0.100    
  15:30        |          |          
  16:00        |          | 0.100    
  16:30        |          |          
  17:00        |          | 0.100    
  17:30        |          |          
  18:00        |          | 0.100    
  18:30        |          |          
  19:00        |          | 0.100    
  19:30        |          |          
  20:00        |          | 0.100    
  20:30        |          |          
  21:00        |          | 0.100    
  21:30        |          |          
  22:00        |          | 0.099    
  22:30        |          |          
  23:00        |          | 0.099    
  23:30        |          |          
goldenowner@instance-2:~/myopenaps/settings$
kmoran02
@kmoran02
Jan 24 2017 13:59
of course that's only maybe 36 hours of data. But its WORKING!!!!!!
What is CSF?
I know its carb sensitivity factor, but what does it mean exactly?
sdneufer
@sdneufer
Jan 24 2017 14:05
It is hard to keep up with all the docs... its here: http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-4/understanding-autotune.html
You might wonder what CSF in the autotune results refers to: Carb Sensitivity Factor is the amount your blood sugar will rise for a given quantity of carbs consumed. And initial value for CSF is calculated from your ISF and carb:insulin ratio (CR), i.e., CSF = ISF / CR (e.g., for an ISF of 42mgDL/U and CR of 14g/U, CSF is 3mgDL/g.) Subsequent autotune estimates for CSF are adjusted for the actual observed post-meal BG rise (relative to what would be expected based on insulin activity) compared to the number of carbs eaten.
kmoran02
@kmoran02
Jan 24 2017 14:06
Thank you. I appreciate the link. :)
I'm absolutely shocked that its not recommending a basal rate change. However as I said, this is based on only what I have entered, which isn't much. I just started entering info the evening of the 22nd.
Tim Street
@tim2000s
Jan 24 2017 14:11
It's not clear in the docs how CSF relates to min_5m_carbimpact in preferences.json when using the closed loop. I think they're the same value, but I think @danamlewis needs to confirm that.
kmoran02
@kmoran02
Jan 24 2017 14:12
@tim2000s , so there's something I may still need to change once Dana confirms?
dougtakeuti
@dougtakeuti
Jan 24 2017 14:14
@cjo20 @colinlennon thanks! That solved it. Any idea what might cause it? cron resets git every minute...
kmoran02
@kmoran02
Jan 24 2017 14:14
I need to go, but if I do need to change something, can one of you get that info to me?
Chris Oattes
@cjo20
Jan 24 2017 14:15
@dougtakeuti lock files are there to stop two programs trying to change the same data at the same time. It usually means that you've got two openaps commands running at the same time.
@tim2000s min_5m_carbimpact is tuning for how quickly you absorb carbs
Tim Street
@tim2000s
Jan 24 2017 14:27
I know. But there has to be a relationship between the two, as your carb sensitivity factor is how much your glucose levels increase per gram of carbs and absorption is how much you glucose levels increase per gram of carbs per five minutes.
Sorry- having a thinking out loud moment. It's probably useful in explaining how the two things interact in OpenAPS if nothing else.
Chris Oattes
@cjo20
Jan 24 2017 14:28
The two are related, but as long as the numbers aren't miles off, it's probably within normal variation
Dana's data shows a CSF of 4.029, for example. Using the default of 3mg/dL/5min gives about 0.75g/5 mins. Most people are probably going to be about that.
Tim Street
@tim2000s
Jan 24 2017 14:30
Anyway. That's an unnecessary aside. @kmoran02 No, you don't need to do something else.
Chris Oattes
@cjo20
Jan 24 2017 14:32
CSF tends to be proportional to body mass (as scott said, volume of water you're dissolving sugar in to).
My gut feeling is that g/min would be proportional to lean body mass (larger body = more surface area to absorb carbs). If both CSF and absorbtion rate are linear with body size, then most people are going to end up with roughly the same mg/dL/5min
Eric Luhrs
@eluhrs
Jan 24 2017 14:42
Hi, all. I have recently installed a new Explorer rig with 3.6-dev. Would like to enable purple lines to show predicted BG. Documentation says to follow NS instructions, but I can't find instructions there. Any advice?
Tim Street
@tim2000s
Jan 24 2017 14:53
Have you set Enable to include OpenAPS in the variable view on Azure/Heroku?
kmoran02
@kmoran02
Jan 24 2017 15:02
@cjo20 wouldn't you expect my CSF to be higher then? I'm not exactly a tiny girl. Not huge, but far from small. Lol!
Eric Luhrs
@eluhrs
Jan 24 2017 15:04
Hi, @tim2000s. Was your comment meant for me or Kmoran02? I do have OpenAPS as one of my enable strings. Just no purple lines.
scarral
@scarral
Jan 24 2017 15:05
@cjo20 well I'm guessing something's wrong since I have data at nightscout (both SG and treatments) but I don't get any suggestions for modifying the basal settings.
I already installed the mdi branch since I'm uploading treatments to NS via HAPP, and that didn't make any difference.
So if we assume that my installation should work, then the problem might be either in the NS website or in how autotune interprets the data coming from NS, as in it doesn't really understand it so it ignores it.
Chris Oattes
@cjo20
Jan 24 2017 15:07
@scarral 1) It only mkes suggestions if it thinks things need to be changed, 2) Your data is really fragmented and you only ran for 1 day, so there might not be enough data there for it to be able to make suggestions
scarral
@scarral
Jan 24 2017 15:07
If I run it for the whole month I don't ge
get any suggestions either. So that's why I wanted someone else to try running it with my website to try and pin down where the problem is.
kmoran02
@kmoran02
Jan 24 2017 15:07
For anyone that can answer. I was referring to what Chris said.
scarral
@scarral
Jan 24 2017 15:08
Even when a basal is really well adjusted, I can't imagine how there wouldn't be at least small differences between the real basal and the suggestions.
Chris Oattes
@cjo20
Jan 24 2017 15:08
@scarral Is the rest of the month as patchy?
scarral
@scarral
Jan 24 2017 15:09
Yes unfortunately, because there is no real coverage in Kindergarten...
And I got a new phone recently and forgot to tell the new HAPP installation to upload treatments to NS. So from the 13th to 21st of January there are no treatments. I don't mind that the results I would get with such data won't be useful, I just want to get it running, and then gather meaningful data.
DunePlodder
@DunePlodder
Jan 24 2017 15:11

```Running 1 runs from 2017-01-23 to 2017-01-23
~/src/oref0/bin/oref0-autotune-prep.js ns-treatments.json profile.json ns-entries.2017-01-23.json > autotune.1.2017-01-23.json
Could not parse input data: [SyntaxError: Unexpected end of input]
~/src/oref0/bin/oref0-autotune-core.js autotune.1.2017-01-23.json profile.json profile.pump.json > newprofile.1.2017-01-23.json
/home/bob/src/oref0/lib/autotune/index.js:77
for (var i=0; i < basalGlucose.length; ++i) {
^

TypeError: Cannot read property 'length' of undefined
at tuneAllTheThings (/home/bob/src/oref0/lib/autotune/index.js:77:39)
at Object.<anonymous> (/home/bob/src/oref0/bin/oref0-autotune-core.js:59:27)
at Module._compile (module.js:410:26)
at Object.Module._extensions..js (module.js:417:10)
at Module.load (module.js:344:32)
at Function.Module._load (module.js:301:12)
at Function.Module.runMain (module.js:442:10)
at startup (node.js:136:18)
at node.js:966:3

Autotune pump profile recommendations:

Recommendations Log File: /home/bob/myopenaps/autotune/autotune_recommendations.log

Parameter | Current | Autotune

ISF | 27.000 | 0.000
CSF | n/a | 0.000
Carb Ratio | 8.000 | 0.000
Basal Profile | |
00:00 | 2.2 |
00:30 | |
```

Chris Oattes
@cjo20
Jan 24 2017 15:13
Please put ``` on a line on it's own before and after pastes like that
scarral
@scarral
Jan 24 2017 15:13
So assuming the problem is the data I have, how should the ideal dataset look like for autotune to make suggestions?
DunePlodder
@DunePlodder
Jan 24 2017 15:13
Sorry I sent that accidentally. I installed Autotune-mdi as suggested. I'm seeing these errors, can anyone point me in the right direction please?
Tim Street
@tim2000s
Jan 24 2017 15:14
@eluhrs You. There's "SHOW_FORECAST" that needs setting to OpenAPS, then you need to turn on OpenAPS in settings on the GUI, and click the three dots next to the different hour periods and select OpenAPS forecast to see the lines. You also need to have AMA enabled on OpenAPS and "OPENAPS_FIELDS" needs to include meal-assist.
ejoe132
@ejoe132
Jan 24 2017 15:14
Setting up my edison on work computer, does putty need itunes to work right? seems like I saw that somewhere but if I dont need it dont want it.
scarral
@scarral
Jan 24 2017 15:15
Another possible problem is that there might be some setting or some enabling I need to do in the azure website to make it work.
So please give me a hint of what I can do!
Tim Street
@tim2000s
Jan 24 2017 15:15
@ejoe132 I believe iTunes gives you bonjour and therefore makes DNS access easier, but I may be incorrect.
Chris Oattes
@cjo20
Jan 24 2017 15:19
@scarral Part of the problem might be that because it's doing everything with small boluses, you don't have any time with 0 bolus active, which means that (as far as I know) autotune would treat that as tuning boluses not basal
Eric Luhrs
@eluhrs
Jan 24 2017 15:19
@tim2000s So that's SHOW_FORECAST=openaps as a stand-alone key/value pair in my NS settings on heroku, right? And not part of my ENABLE strings?
taylorfowler
@taylorfowler
Jan 24 2017 15:19
@scarral Here's what I get when I run your profile.json and website.
Parameter      | Current  | Autotune 
-------------------------------------
ISF            | 225.000  | 25.000   
CSF            | n/a      | 3.125    
Carb Ratio     | 30.000   | 8.000    
Basal Profile  |          |          
  00:00        | 0.050    | 1.000    
  00:30        |          |          
  01:00        |          | 1.000    
  01:30        |          |          
  02:00        |          | 1.000    
  02:30        |          |          
  03:00        | 0.050    | 1.000    
  03:30        |          |          
  04:00        | 0.100    | 1.000    
  04:30        |          |          
  05:00        | 0.100    | 1.000    
  05:30        |          |          
  06:00        |          | 1.000    
  06:30        |          |          
  07:00        |          | 1.000    
  07:30        |          |          
  08:00        |          | 1.200    
  08:30        |          |          
  09:00        | 0.100    | 1.200    
  09:30        |          |          
  10:00        |          | 1.500    
  10:30        |          |          
  11:00        |          | 1.500    
  11:30        |          |          
  12:00        |          | 1.500    
  12:30        |          |          
  13:00        |          | 1.200    
  13:30        |          |          
  14:00        |          | 1.200    
  14:30        |          |          
  15:00        |          | 1.000    
  15:30        |          |          
  16:00        |          | 1.000    
  16:30        |          |          
  17:00        | 0.175    | 1.000    
  17:30        |          |          
  18:00        |          | 1.000    
  18:30        |          |          
  19:00        | 0.250    | 1.000    
  19:30        |          |          
  20:00        | 0.750    | 1.000    
  20:30        |          |          
  21:00        | 0.550    | 1.000    
  21:30        |          |          
  22:00        | 0.225    | 1.000    
  22:30        |          |          
  23:00        | 0.150    | 1.000    
  23:30        |          |
though those basal rates seem really high in comparison....
scarral
@scarral
Jan 24 2017 15:20
here's what I get for yesterday, or any other day... no matter how many days I put I always get the same:

```Parameter | Current | Autotune

ISF | 225.000 | 225.000
CSF | n/a | 7.500
Carb Ratio | 30.000 | 30.000
Basal Profile | |
00:00 | 0.050 | 0.050
00:30 | |
01:00 | | 0.050
01:30 | |
02:00 | | 0.050
02:30 | |
03:00 | 0.050 | 0.050
03:30 | |
04:00 | 0.100 | 0.100
04:30 | |
05:00 | 0.100 | 0.100
05:30 | |
06:00 | | 0.100
06:30 | |
07:00 | | 0.100
07:30 | |
08:00 | | 0.100
08:30 | |
09:00 | 0.100 | 0.100
09:30 | |
10:00 | | 0.100
10:30 | |
11:00 | | 0.100
11:30 | |
12:00 | | 0.100
12:30 | |
13:00 | | 0.100
13:30 | |
14:00 | | 0.100
14:30 | |
15:00 | | 0.100
15:30 | |
16:00 | | 0.100
16:30 | |
17:00 | 0.175 | 0.175
17:30 | |
18:00 | | 0.175
18:30 | |
19:00 | 0.250 | 0.250
19:30 | |
20:00 | 0.750 | 0.750
20:30 | |
21:00 | 0.550 | 0.550
21:30 | |
22:00 | 0.225 | 0.225
22:30 | |
23:00 | 0.150 | 0.150
23:30 | | ```
taylorfowler
@taylorfowler
Jan 24 2017 15:21
OH, your ISF is set to 225. Is that right?
scarral
@scarral
Jan 24 2017 15:21
OK so this is not he right character ```
Yes, my son is 4 years old.
So there is definitely a problem, since your result is very different from mine...
ejoe132
@ejoe132
Jan 24 2017 15:22
@tim2000s ok Ill download it then. Also does anyone know the cord needed for a dexcom tied into the explorer board setup? I believe I will need this as I wont have wifi most of the time and dont have hotspot available on my phone, is it just a micro usb to micro usb cable?
Eric Luhrs
@eluhrs
Jan 24 2017 15:24
@tim2000s assuming my SHOW_FORECAST setting is correct (see above), then the only other thing I'm not clear on is this: "OpenAPS in settings on the GUI, and click the three dots next to the different hour periods and select OpenAPS forecast".
@tim2000s I have OpenAPS checked on my NS site config, but I don't see the three dots or an option to select OpenAPS forecast.
taylorfowler
@taylorfowler
Jan 24 2017 15:25
@scarral Working now? No changes?
*I mean you didnt make any changes.
scarral
@scarral
Jan 24 2017 15:25
Same as before.
No, I don't know what to change! I'll be very glad when I get guidance on what I can change!
@taylorfowler did you run it for yesterday?
This is what I get for yesterday:

```Parameter | Current | Autotune

ISF | 225.000 | 225.000
CSF | n/a | 7.500
Carb Ratio | 30.000 | 30.000
Basal Profile | |
00:00 | 0.050 | 0.050
00:30 | |
01:00 | | 0.050
01:30 | |
02:00 | | 0.050
02:30 | |
03:00 | 0.050 | 0.050
03:30 | |
04:00 | 0.100 | 0.100
04:30 | |
05:00 | 0.100 | 0.100
05:30 | |
06:00 | | 0.100
06:30 | |
07:00 | | 0.100
07:30 | |
08:00 | | 0.100
08:30 | |
09:00 | 0.100 | 0.100
09:30 | |
10:00 | | 0.100
10:30 | |
11:00 | | 0.100
11:30 | |
12:00 | | 0.100
12:30 | |
13:00 | | 0.100
13:30 | |
14:00 | | 0.100
14:30 | |
15:00 | | 0.100
15:30 | |
16:00 | | 0.100
16:30 | |
17:00 | 0.175 | 0.175
17:30 | |
18:00 | | 0.175
18:30 | |
19:00 | 0.250 | 0.250
19:30 | |
20:00 | 0.750 | 0.750
20:30 | |
21:00 | 0.550 | 0.550
21:30 | |
22:00 | 0.225 | 0.225
22:30 | |
23:00 | 0.150 | 0.150
23:30 | | ```
Very different from yours @taylorfowler
Chris Oattes
@cjo20
Jan 24 2017 15:30
taylorfowler probably isn't running with exactly the same settings
Also, there's something niggling me about those basal rates
@scottleibrand Where did we get to with basal rates under 0.1u/hr? I can't remember where it was left (looking at scarrals midnight rate)
Also, @taylorfowler's run shows it changing the values by 20x, I thought it was meant to be limited to 1.2x
scarral
@scarral
Jan 24 2017 15:35
@cjo20 well he's got the same data, and the autotune-mdi branch like I do, I don't know which settings I could change to get it to do something! Anything!
Chris Oattes
@cjo20
Jan 24 2017 15:36
do you have to set up profile.json in the -mdi branch?
scarral
@scarral
Jan 24 2017 15:36
The midnight basal is correct, it's set at 0.05 IU per hour.
I have profile.json.
I sent it before. I can't get the ``` characters to work... should I paste it again?
Chris Oattes
@cjo20
Jan 24 2017 15:36
no
You need to put the last ``` on a line on their own too
And the ones at the top
You should have two lines, one at the top and one at the bottom, with only ``` on it
scarral
@scarral
Jan 24 2017 15:37
And in profile.json I have the basal rates that you see on the middle column of @taylorfowler s result
I see thanks for clarifying.
Chris Oattes
@cjo20
Jan 24 2017 15:43
"autosens_max": 1.2, and autosens_min: 0.7, add those to the 3 profile.json files
Tim Street
@tim2000s
Jan 24 2017 15:44
@eluhrs these three dots: https://imgur.com/PH0Td8w
Chris Oattes
@cjo20
Jan 24 2017 15:46
@scarral The fact that taylorfowlers run came back with basal rates of 1 is worrying because the default autosens_max is 1.2. That menans that it should, at most, recommend current basal * 1.2. 1.0u/hr is 20x the midnight basal
Eric Luhrs
@eluhrs
Jan 24 2017 15:56
@tim2000s Hmmm. I don't have those dots. Do you know of any docs on how to configure? I must be missing something.
DunePlodder
@DunePlodder
Jan 24 2017 15:58
Parameter      | Current  | Autotune 
-------------------------------------
ISF            | 27.000   | 27.000   
CSF            | n/a      | 2.945    
Carb Ratio     | 8.000    | 9.168    
Basal Profile  |          |          
  00:00        | 3.000    | 3.001    
  00:30        |          |          
  01:00        |          | 2.840    
  01:30        |          |          
  02:00        |          | 2.653    
  02:30        |          |          
  03:00        |          | 2.543    
  03:30        |          |          
  04:00        |          | 2.638    
  04:30        |          |          
  05:00        |          | 2.966    
  05:30        |          |          
  06:00        |          | 3.106    
  06:30        |          |          
  07:00        |          | 3.125    
  07:30        |          |          
  08:00        | 1.500    | 1.433    
  08:30        |          |          
  09:00        |          | 1.388    
  09:30        |          |          
  10:00        |          | 1.370    
  10:30        |          |          
  11:00        |          | 1.350    
  11:30        |          |          
  12:00        |          | 1.311    
  12:30        |          |          
  13:00        |          | 1.475    
  13:30        |          |          
  14:00        |          | 1.512    
  14:30        |          |          
  15:00        |          | 1.586    
  15:30        |          |          
  16:00        |          | 1.445    
  16:30        |          |          
  17:00        |          | 1.454    
  17:30        |          |          
  18:00        |          | 1.515    
  18:30        |          |          
  19:00        |          | 1.741    
  19:30        |          |          
  20:00        | 3.000    | 3.182    
  20:30        |          |          
  21:00        |          | 3.228    
  21:30        |          |          
  22:00        |          | 3.063    
  22:30        |          |          
  23:00        |          | 3.028    
  23:30        |          |
Not sure it makes sense but it did run successfully!
Chris Oattes
@cjo20
Jan 24 2017 16:00
That's quite a cliff edge in basal rate between 0730 and 0800 there
DunePlodder
@DunePlodder
Jan 24 2017 16:01
I'm MDI, 18 units in the morning & 36 in the evening. I wasn't sure how to enter that?
Chris Oattes
@cjo20
Jan 24 2017 16:01
Don't think you can any other way
taylorfowler
@taylorfowler
Jan 24 2017 16:04
@scarral sorry, I've been in a meeting. @cjo20 Yeah i was surprised at the giant basal jumps too. Some thing may be screwy in there
taylorfowler
@taylorfowler
Jan 24 2017 16:13
Here's the results when i add the auto tune min and max:
Parameter      | Current  | Autotune 
-------------------------------------
ISF            | 225.000  | 187.500  
CSF            | n/a      | 3.125    
Carb Ratio     | 30.000   | 60.000   
Basal Profile  |          |          
  00:00        | 0.050    | 0.060    
  00:30        |          |          
  01:00        |          | 0.060    
  01:30        |          |          
  02:00        |          | 0.060    
  02:30        |          |          
  03:00        | 0.050    | 0.060    
  03:30        |          |          
  04:00        | 0.100    | 0.120    
  04:30        |          |          
  05:00        | 0.100    | 0.120    
  05:30        |          |          
  06:00        |          | 0.120    
  06:30        |          |          
  07:00        |          | 0.120    
  07:30        |          |          
  08:00        |          | 0.120    
  08:30        |          |          
  09:00        | 0.100    | 0.120    
  09:30        |          |          
  10:00        |          | 0.120    
  10:30        |          |          
  11:00        |          | 0.120    
  11:30        |          |          
  12:00        |          | 0.120    
  12:30        |          |          
  13:00        |          | 0.120    
  13:30        |          |          
  14:00        |          | 0.120    
  14:30        |          |          
  15:00        |          | 0.120    
  15:30        |          |          
  16:00        |          | 0.120    
  16:30        |          |          
  17:00        | 0.175    | 0.210    
  17:30        |          |          
  18:00        |          | 0.210    
  18:30        |          |          
  19:00        | 0.250    | 0.300    
  19:30        |          |          
  20:00        | 0.750    | 0.900    
  20:30        |          |          
  21:00        | 0.550    | 0.660    
  21:30        |          |          
  22:00        | 0.225    | 0.270    
  22:30        |          |          
  23:00        | 0.150    | 0.180    
  23:30        |          |
@scarral
scarral
@scarral
Jan 24 2017 16:21
And that's for yesterday, right?
@cjo20 yes I know they were dodgy the numbers @taylorfowler got before with huge basal jumps, but I still see it as progress, since I don't get any difference between current and suggested.
so @taylorfowler what do I have to change in my setup to get those numbers?
taylorfowler
@taylorfowler
Jan 24 2017 16:24
Maybe try wiping out your myopenaps folder and starting fresh @scarral ?
scarral
@scarral
Jan 24 2017 16:24
I've deleted the /src directory and reinstalled everything already.
Didn't make any difference.
You mean just the myopenaps directory?
Tim Street
@tim2000s
Jan 24 2017 16:24
@eluhrs That's the "SHOW_FORECAST" setting that needs to be in your Azure/Heroku settings.
scarral
@scarral
Jan 24 2017 16:25
I deleted the autotune directory where all the files are being generated, and run again, and got same result.
Where did you add the min and max for the autotune?
Chris Oattes
@cjo20
Jan 24 2017 16:27
scarral: you add it to profile.json, and to all of the copies of that you make
taylorfowler
@taylorfowler
Jan 24 2017 16:28
I'd wipe src and myopenaps and start from scratch. the min and max just need to go on the end of profile .json:
"min_5m_carbimpact": 3,
"dia": 4,
"basalprofile": [
{
"i": 0,
"start": "00:00:00",
"minutes": 0,
"rate": 0.05
},
{
"i": 1,
"start": "03:00:00",
"minutes": 180,
"rate": 0.05
},
{
"i": 2,
"start": "04:00:00",
"minutes": 240,
"rate": 0.1
},
{
"i": 3,
"start": "05:00:00",
"minutes": 300,
"rate": 0.1
},
{
"i": 4,
"start": "09:00:00",
"minutes": 540,
"rate": 0.1
},
{
"i": 5,
"start": "17:00:00",
"minutes": 1020,
"rate": 0.175
},
{
"i": 6,
"start": "19:00:00",
"minutes": 1140,
"rate": 0.25
},
{
"i": 7,
"start": "20:00:00",
"minutes": 1200,
"rate": 0.750
},
{
"i": 8,
"start": "21:00:00",
"minutes": 1260,
"rate": 0.55
},
{
"i": 9,
"start": "22:00:00",
"minutes": 1320,
"rate": 0.225
},
{
"i": 10,
"start": "23:00:00",
"minutes": 1380,
"rate": 0.15
}
],
"isfProfile": {
"sensitivities": [
{
"i": 0,
"start": "00:00:00",
"sensitivity": 225,
"offset": 0,
"x": 0,
"endOffset": 1440
}
]
},
"carb_ratio": 30
,"autosens_max": 1.2
,"autosens_min": 0.1
}
scarral
@scarral
Jan 24 2017 16:28
yes? where's the end?
Can you please post again the command you used to install your instance?
taylorfowler
@taylorfowler
Jan 24 2017 16:29
Sorry had a keyboard flub.
Chris Oattes
@cjo20
Jan 24 2017 16:30
min should be 0.7
taylorfowler
@taylorfowler
Jan 24 2017 16:30
yeah fix that. fat fingered / misread
scarral
@scarral
Jan 24 2017 16:30
What do these limits mean? That autotune will not suggest anything above autosens_max or below autosens_min?
Chris Oattes
@cjo20
Jan 24 2017 16:31
no. That it won't scale it by more than those
scarral
@scarral
Jan 24 2017 16:31
Or is it a percentage? Don't suggest anything above 120% or below 70%?
Chris Oattes
@cjo20
Jan 24 2017 16:31
so if your basal rate is 0.1u/hr, and the rates are set to 0.7/1.2, it won't suggest less than 0.07u/hr or more than 0.12u/hr
scarral
@scarral
Jan 24 2017 16:31
I see.
So I have to leave now, but I'm going to reinstall everything later as per written in the docs replacing dev with autotune-mdi, right?
scarral
@scarral
Jan 24 2017 16:34
yeah, somehow the version I used yesterday didn't have the min and max, I don't think...
or maybe I miscopied.
I just ran it again with the min and max at the end, and get the same as before....
taylorfowler
@taylorfowler
Jan 24 2017 16:35

So to install i used

mkdir -p ~/src; cd ~/src && git clone -b dev git://github.com/openaps/oref0.git || (cd oref0 && git checkout autotune-mdi && git pull); cd
cd ~/src/oref0 && git checkout autotune-mdi && sudo npm run global-install

Which again may be completely wrong (cloning dev then checking out autotune-mdi)

@scarral ^^
scarral
@scarral
Jan 24 2017 16:36
I'll try that again later with the min an max in profile.json. Wipe the /src directory and repeat, right?
taylorfowler
@taylorfowler
Jan 24 2017 16:36
yeah at this point i'd start from a clean slate
scarral
@scarral
Jan 24 2017 16:37
can I leave the myopenaps directory then?
taylorfowler
@taylorfowler
Jan 24 2017 16:37
Nah, rebuild it all. start from scratch.
Eric Luhrs
@eluhrs
Jan 24 2017 16:55
@tim2000s Okay, if figured out the main problem. My NS wasn't new enough. Just updated it to 9.2 and now I get the purple line, BUT.... all of my past basals and treatments are now missing. And so is the little loop/enact icon on the OpenAPS pill.
I also don't get the three dots even though I have SHOW_FORECAST set.
Eric Luhrs
@eluhrs
Jan 24 2017 17:19
@tim2000s I'm getting there. The only thing still not correct is that I don't see temp basal values as numbers. The blue line is there, just no legend that states each temp basal rate. I suspect I just need to find the right config var/string.
katie disimone
@Kdisimone
Jan 24 2017 17:25
annaloop_·_Settings___Heroku.jpg
What are the correct settings in NS for retro OpenAPS info? I currently have this but am not succesful at seeing old data @jasoncalabrese
katie disimone
@Kdisimone
Jan 24 2017 17:42
nm…solved my problem. thanks
Scott Leibrand
@scottleibrand
Jan 24 2017 18:11
for those of you on MDI with basal injections instead of pumping short-acting insulin for basal, I would put that in as a single average basal value for the entire day. basal insulins tend to have anywhere from an 18 to 36+ hour curve, so your basal doesn't actually change very much over the course of the day.
@DunePlodder ^^
katie disimone
@Kdisimone
Jan 24 2017 18:15

So what do you all do when it looks like OpenAPS hasn't pulled a pump reading for about 22min? I set up my daughters iPhone with the app to ssh in, so I can tell her to enter a command to force a mmtune...but I don't know what that command is exactly.

And since I'm a new user, how often does this happen roughly? Usually caused by??? Phone battery good. Edison battery good. Pump battery is lithium reading 1.42v....having her change it just in case.

Chris Oattes
@cjo20
Jan 24 2017 18:17
Kdisimone: what does the openaps pill in NS say?
katie disimone
@Kdisimone
Jan 24 2017 18:32

@cjo20 it said waiting. Pump data was 22min old, as was the waiting time.

It has resolved (took 31min) Not sure if it was coincidence with the battery change. It was reading full bars on the pump. Wifi signal was good.

What is the command to force a pump read/tune?

Chris Oattes
@cjo20
Jan 24 2017 18:45
It wasn't failing to read the pump, there's an issue with the way the pump sends data sometimes
It'll do that for 30 mins or more, then go back to working. Sometimes calibrating fixes it sooner
Tim Street
@tim2000s
Jan 24 2017 19:02
@cjo20 Is that what your mmeowlink update fixes?
Talking of which, to add that in, rather than reloading the entire oref0 set up, does simply reinstalling mmeowlink fix it?
Dana Lewis
@danamlewis
Jan 24 2017 19:05
@tim2000s @cjo20 csf is not something that's tied to min_5m_carbimpact. Don't change the latter. Csf is a relationship of carbs and isf. Not meant to be used to inform anything else
Tim Street
@tim2000s
Jan 24 2017 19:05
No, agreed. I think we arrived at that earlier. 😉 My error.
Dana Lewis
@danamlewis
Jan 24 2017 19:06
@taylorfowler you have autosens min and max WAY off defaults. Is that what you're using for looping? Regardless, may want to be careful copying and pasting that for people running autotune as one-off - they need to use the example values from the example in docs
@DunePlodder great to hear re working well. Did you see @scottleibrand suggestion of entering basal?
@tim2000s k catching up on 12h of content so just wanted to make sure :smile:
katie disimone
@Kdisimone
Jan 24 2017 19:24
@cjo20 thanks
Dana Lewis
@danamlewis
Jan 24 2017 19:40
Hey all - I wasn’t expecting the flurry of interest in autotune that we got :) and I know that’s making it hard to track loop setup conversations. So let’s move autotune discussion, questions generally over to this new room: https://gitter.im/openaps/autotune cc @cjo20 @scottleibrand @tim2000s @taylorfowler @DunePlodder @RosyRed @Kdisimone @scarral @SandraK82 @PoweRGbg @StuArnold @kmoran02 (tagging a few people who I know have been working through autotune, and/or helping others troubleshoot so you’ll see the new room. Thanks!)
Sandra Keßler
@SandraK82
Jan 24 2017 19:52
Thanks @danamlewis will try to direct as many people from here to there during nightshift ;)
Dana Lewis
@danamlewis
Jan 24 2017 19:53
:) thanks, and big props to you (all) for so much troubleshooting! cc @cjo20 @tim2000s @taylorfowler too :)
Tim Street
@tim2000s
Jan 24 2017 20:04
👍
Dana Lewis
@danamlewis
Jan 24 2017 20:06
:heart: for a true :earth_americas: :earth_africa: :earth_asia: :globe_with_meridians: community :)
@eluhrs - did you get sorted on NS display, or still need some input?
ejoe132
@ejoe132
Jan 24 2017 21:25
Trying to get my edison/explorer setup running but hitting a problem. When I try to download from the provided link the two extra files needed for pc setup I am only getting the dfu-util.exe not the libusb-1.0.dll, what am I missing? I get into command prompt and do flashall.bat but get a message saying dfu-utl is not recognized as an internal or external command
PieterGit
@PieterGit
Jan 24 2017 22:39
@scottleibrand @oskarpearson @ecc1 : would it be possible/helpfull to use/test another gpio library on an edison with the TI-stick? e.g. https://pypi.python.org/pypi/spidev . The answer of Andy Shev on andy-shev/linux#8 does not look quite promising on getting the TI working on the explorer board with a recent kernel
Scott Leibrand
@scottleibrand
Jan 24 2017 22:39
wrong channel @scarral :)
PieterGit
@PieterGit
Jan 24 2017 22:41
Can anybody explain which GPIO ports we need to use on an Explorerboard and what the function of pin 23 is?
Scott Leibrand
@scottleibrand
Jan 24 2017 22:42
@PieterGit I'd be happy to help try out something to replace libmraa. Do you think spidev would work without too many code changes? I'm not an expert on our python code, so I might not be much help there...
PieterGit
@PieterGit
Jan 24 2017 22:43
i don't know but andy shev seems to have strong opionions on the code quality of libmraa :smile: . So I want to see if we get communication working with the TI on an explorer board with a newer kernel
Scott Leibrand
@scottleibrand
Jan 24 2017 22:44
would it be easier to test out spidev with a known-working jubilinux build first?
PieterGit
@PieterGit
Jan 24 2017 22:47
i never got USB OTG with a Dexcom G4 stable on a jubilinux or 3.10 kernel and I don't see a lot of activity in jubilinux OS itself. USB OTG seems stable on a recent kernel. i'll dig into it later, but would like to ask @oskarpearson and @ecc1 about their opinions of the best route of getting forward. My best guess still is upgrading to recent kernel 4.10rc5+ and trying to get USB OTG and TI-communication working with ubuntu based distribution.
Scott Leibrand
@scottleibrand
Jan 24 2017 22:52
EnhancedRadioDevices/915MHzEdisonExplorer#14
PieterGit
@PieterGit
Jan 24 2017 22:59
i don't understand why i consistently got bad results with 3.10 kernel and got a few good USB connections with a 4.10rc kernel
thanks for the link to the github issue
Eric
@ecc1
Jan 24 2017 23:01
@PieterGit I don't use Python or libmraa or TI stick or Explorer, so it won't be me. But I don't understand why the SPI code for the Explorer evidently worked with the 3.10 kernel and not the new kernel. Whoever decided on which pins to use for which connections on the Explorer board needs to look into it, I would think.
Scott Leibrand
@scottleibrand
Jan 24 2017 23:06
@PieterGit sounds like there is some register setting that might be incorrect, so a software fix might be possible: dunno yet. still waiting for more word from @mogar (and he on word from Intel) to confirm that.
PieterGit
@PieterGit
Jan 24 2017 23:12
@ecc1 makes sense. Thanks for helping me with getting the recent kernel working
@scottleibrand : i'll check later and focus on the USB OTG with a newer kernel first to see if that works. goal is to make a stable USB OTG Explorer rig
i emailed @mogar , I hope a software fix is possible, otherwise I'll go cutting or will order a new Explorer board with a newer revision if I screw up cutting wires :worried:
Scott Leibrand
@scottleibrand
Jan 24 2017 23:14
@PieterGit the Explorer board that Morgan gave me to test has been running stable with USB OTG, FWIW
PieterGit
@PieterGit
Jan 24 2017 23:15
Unfortunately I got a production based one :smile: . I'm out, going to bed.
Scott Leibrand
@scottleibrand
Jan 24 2017 23:15
'nite
Dana Lewis
@danamlewis
Jan 24 2017 23:33
@colinlennon reading the readme - nicely done! I can’t test the directions since I’m not using xdrip, but they seem to catch everything I was wondering about. Want to go ahead and create a PR and put this in a page in the docs? That way 1-2 people can test it and thumbs up the PR or make notes more easily if they have any feedback on it before it gets merged? For location in the docs..what do you think about adding it to phase 1? Can rework the offline monitoring to be an overview page of all the offline options, followed by pancreabble page and then your xdrip directions..?