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

6th
Oct 2017
cameronrenwick
@cameronrenwick
Oct 06 2017 00:02
I guess too that everyone saw this eh? http://www.animaspatientsupport.com/
bye bye animas..
Brian Rabinovitz
@bdr1177
Oct 06 2017 00:54
Is there a consensus on whether the exponential curves are superior to the bi-linear curves for predicting insulin action?
Steve Bernacki
@sbernacki
Oct 06 2017 01:41
Hi everyone; I've got my wife looping with OpenAPS, her old Medtronic pump and Dexcom. She was reading https://openaps.org/2016/12/31/2016-update-from-the-openaps-community/ where they describe advanced meal assist. Links from that page just go to the autosens documentation and I can't find anything specific about AMA. Was it folded into another feature or am I just missing something obvious? She's curious to try it but neither of us can figure out what we need to do in order to do so.
Dana Lewis
@danamlewis
Oct 06 2017 02:02
@sbernacki hi! AMA is now on by default for everyone
So you've already been using it probably :smile:
Steve Bernacki
@sbernacki
Oct 06 2017 02:03
Ah hah! Thank you, Dana :) I was just reading through more of the documentation and was starting to come to that conclusion, just didn't see it stated outright so I was questioning my sanity lol
Dana Lewis
@danamlewis
Oct 06 2017 02:11
:smile:
Dana Lewis
@danamlewis
Oct 06 2017 02:33
Note: New PR for 0.6.0-dev - it's now https://github.com/openaps/oref0/pull/694#issue-263319617 (PR 694 instead of 650)
Scott Leibrand
@scottleibrand
Oct 06 2017 02:42

@CaptainBalou if you want to dig deeper than that you'll have to start looking at the code, and probably uncommenting some debug lines. that particular warning comes from lib/iob/index.js (https://github.com/openaps/oref0/blob/dev/lib/iob/index.js#L21-L23):

~/src/oref0 $ grep -r "please pass clock-zoned" *
lib/iob/index.js:        console.error("Warning: clock input " + inputs.clock + " is unzoned; please pass clock-zoned.json instead");

as you can see from the line right before that, there is already a console.error(inputs.clock); line that's commented out. if you uncomment that locally on your rig (in ~/src/oref0/lib/iob/index.js) , you should be able to re-run openaps report invoke monitor/meal.json and see the additional context

@bdr1177 everyone who's tried them so far thinks exponential curves are better than bilinear, yes. (as they certainly should be: the bilinear curves were a really rough approximation I did while sitting at a McDonald's in Everett back when first building DIYPS before we knew much of anything about what would actually work). we'll probably switch the default before too long, but would like everyone to test them out and confirm they're better in all cases (or can be tweaked to be)
LilDucky
@LilDucky
Oct 06 2017 03:33
@tazitoo tried that with =1 as per most docs. Will try again
Tim Street
@tim2000s
Oct 06 2017 05:31
@bdr1177 The consensus is that the exponential curves are across both Loop and openaps. I certainly wouldn't go back. When using Fiasp they are a must.
Scott Leibrand
@scottleibrand
Oct 06 2017 06:03
I think you _ a word. ;-)
Marcus Lindfors
@Mackwe
Oct 06 2017 08:47
After installing the latest dev, i get this in the log: Old settings refresh at Module.load (module.js:356:32) ed. glucose.json newer than temp_basal.json. Temp refresh at Module.load (module.js:356:32) ed Could not parse input data: [SyntaxError: /root/myopenaps/monitor/iob.json: Unexpected end of input] and iob.json never seems to become populated. Eaasy fix, or do I need to re-flash/re-install?
Marco
@CaptainBalou
Oct 06 2017 09:01

@CaptainBalou if you want to dig deeper than that you'll have to start looking at the code, and probably uncommenting some debug lines. that particular warning comes from lib/iob/index.js (https://github.com/openaps/oref0/blob/dev/lib/iob/index.js#L21-L23):

~/src/oref0 $ grep -r "please pass clock-zoned" *
lib/iob/index.js:        console.error("Warning: clock input " + inputs.clock + " is unzoned; please pass clock-zoned.json instead");

as you can see from the line right before that, there is already a console.error(inputs.clock); line that's commented out. if you uncomment that locally on your rig (in ~/src/oref0/lib/iob/index.js) , you should be able to re-run openaps report invoke monitor/meal.json and see the additional context

Perfect. Thanks for that. I look into it this evening.

After installing the latest dev, i get this in the log: Old settings refresh at Module.load (module.js:356:32) ed. glucose.json newer than temp_basal.json. Temp refresh at Module.load (module.js:356:32) ed Could not parse input data: [SyntaxError: /root/myopenaps/monitor/iob.json: Unexpected end of input] and iob.json never seems to become populated. Eaasy fix, or do I need to re-flash/re-install?

I saw this as well after an update a few days ago. But "it fixed itself" after some time. Cannot say why. So probably wait a bit...

Mike
@libxmike
Oct 06 2017 09:08
@tim2000s which curve do you use with Fiasp?
Marcus Lindfors
@Mackwe
Oct 06 2017 09:08
@CaptainBalou Okay. Thanks. I'll have a look at it again, then. Any ideas where I can locate module.js to check what's going on?
Tim Street
@tim2000s
Oct 06 2017 09:21
Ultra-rapid, as per the instructions @libxmike
Mike
@libxmike
Oct 06 2017 09:41
thx @tim2000s, just started few days ago with SMB and Ultra-rapid+Fiasp, works very well
Marco
@CaptainBalou
Oct 06 2017 11:47
@Mackwe I am sorry but I have no clue. I can just say what happens in my setup. I am not experienced in coding. Sorry.
ejoe132
@ejoe132
Oct 06 2017 14:18
So trying to get setup again, go to clear off what is on the system to do a fresh install with cd
crontab-r
rm -rf src
rm -rf myopenaps
but it is telling me its not able to do anything cause its a read only file system. Not quite sure what to do from here as thats how Ive always cleared it out before
Marcus Lindfors
@Mackwe
Oct 06 2017 15:19
After installing the latest dev, i get this in the log: Old settings refresh at Module.load (module.js:356:32) ed. glucose.json newer than temp_basal.json. Temp refresh at Module.load (module.js:356:32) ed Could not parse input data: [SyntaxError: /root/myopenaps/monitor/iob.json: Unexpected end of input] and iob.json never seems to become populated. Easy fix, or do I need to re-flash/re-install?
(Sorry, Marco
Marco
Waiting didn't help)
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 15:42
@ejoe132 are you doing it as root?
Scott Leibrand
@scottleibrand
Oct 06 2017 15:44
@Mackwe check your preferences.json for syntax errors like missing commas, bad quotes, etc.
And if the problem isn't obvious, drill down and run the iob.json and profile.json reports manually per the troubleshooting instructions to see the full error message.
Marcus Lindfors
@Mackwe
Oct 06 2017 15:45
@scottleibrand Thx! Will do!
Marco
@CaptainBalou
Oct 06 2017 15:45

@Mackwe check your preferences.json for syntax errors like missing commas, bad quotes, etc.

OMG yes I remember. On some day I had an empty iob.json as well. Hat a copy paste error in preferences.json with wrong "

ejoe132
@ejoe132
Oct 06 2017 15:51
@diabetic_dude_twitter Yes logging in with my root login and password. Its odd becasue when I go into the log, I can see stuff from the last time it was running(sept. 26) then the log just goes and at the bottom says refresh and isnt continuing
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 15:52
try just rm -r
or rename the directories and install then try to remove the old ones after you are done
Marcus Lindfors
@Mackwe
Oct 06 2017 16:01
@CaptainBalou @sbernacki
@CaptainBalou @scottleibrand Just found a missing comma in preferences.jo
.json (God I hate Gitter sometimes). Sorry for mistagging above..
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 16:06
@ejoe132 to rename the folders you could also try "mv myopenaps/ backupmyopenaps/" and "mv src/ backupsrc/" without the quotes
jaylagorio
@jaylagorio
Oct 06 2017 16:28
Trying to debug this on the fly. Last night Git had another problem but this time the reset command didn't work so I blew away the whole directly, keeping my preference file. I reran my setup command and added --nogit to it (the script said it didn't recognize that or --no-git). I have my new directory, copied my preference file, and today have done two meals. Upon putting the carbs into Nightscout it's all getting picked up and it shows up in the logs, but all I'm getting is max high temping - no boluses
Screenshot:
image.png
Preference file:
{                                                                                                    
        "max_daily_safety_multiplier": 7,                                                            
        "current_basal_safety_multiplier": 7,                                                        
        "max_iob": 5.0,                                                                              
        "autosens_max": 1.2,                                                                         
        "autosens_min": 0.7,                                                                         
        "rewind_resets_autosens": true,                                                              
        "autosens_adjust_targets": true,                                                             
        "adv_target_adjustments": true,                                                              
        "maxCOB": 120,                                                                               
        "override_high_target_with_low": false,                                                      
        "skip_neutral_temps": false,                                                                 
        "bolussnooze_dia_divisor": 2,                                                                
        "min_5m_carbimpact": 5,                                                                      
        "carbratio_adjustmentratio": 1,                                                              
        "autotune_isf_adjustmentFraction": 0.5,                                                      
        "remainingCarbsFraction": 0.7,                                                               
        "remainingCarbsCap": 0,                                                                      
        "enableUAM": true,                                                                           
        "enableSMB_with_bolus": false,                                                               
        "enableSMB_with_COB": true,                                                                  
        "enableSMB_with_temptarget": true,                                                           
        "enableSMB_after_carbs": true,                                                               
        "curve": "rapid-acting",                                                                     
        "maxSMBBasalMinutes": 45                                                                     
}
Setup command: oref0-setup --dir=/root/openaps --serial=386591 --cgm=G5 --ns-host=http://xxx --api-s ecret=xxx --tty=/dev/spidev5.1 --enable=' autosens meal autotune microbolus ' --radio_locale='US' --btmac=CC:F3:A5:xx:xx:xx --max_iob=5.0 --current_basal_safety_multiplier=7 --max_daily_safe ty_multiplier=7
Anyone see what I'm doing wrong?
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 16:35
what are the logs saying about SMB?
it should tell you weather it is working or not.
jaylagorio
@jaylagorio
Oct 06 2017 16:38
I think that's this part:
Checking deliverAt: 2017-10-06T16:32:22.094Z is within 1m of current time: Fri Oct  6 12:32:22 EDT 20
17                                                                                                   
and that smb-suggested.json is less than 1m old                                                      
enact/smb-suggested.json: {"insulinReq":1.035,"bg":169,"reservoir":"113.5","temp":"absolute","snoozeB
G":211,"predBGs":{"COB":[169,179,188,196,203,209,214,218,222,225,227,230,231,232,233,234,234,233,233,
232,231,230,229,227,226,224,222,220,217,215,212,209,206,203,199,196,193,190,187,184,182,180,178,176,1
74,172,171,170],"aCOB":[169,176,182,186,190,194,196,199,200,201,202,203,203,203,203,202,202,201,200,2
00,199,198,197,196,195,194,193,192,191,190,189,188,187,187,186,185,184,183,182,181,180,179,178,176,17
5,174,173,171],"UAM":[169,179,187,194,199,203,206,208,208,208,207,205,202,198,193,188,182,176,169,162
,156,149,143,137,131,126,120,115,110,106,101,97,93,89,86,82,79,76,73,71,68,66,64,62,60,59,57,56],"IOB
":[169,178,186,191,195,198,198,198,195,192,187,181,174,167,160,152,145,138,131,125,118,112,106,100,94
,88,83,78,73,68,64,60,56,52,48,45,42,39]},"IOB":3.965,"reason":"COB: 44, Dev: 81, BGI: -2.99, ISF: 58
, Target: 80, minPredBG 146, minGuardBG 168, IOBpredBG 39, COBpredBG 170, UAMpredBG 56; Eventual BG 1
70 >= 80, max_iob 5, temp 3 >~ req 2.7U/hr. ","COB":44,"eventualBG":170,"tick":"+12","deliverAt":"201
7-10-06T16:32:22.094Z"}                                                                              
No smb_enact needed. Temp refreshed: monitor/temp_basal.json: {"duration":21,"rate":3,"temp":"absolut
e"}                                                                                                  
No bolus needed (yet).
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 16:38
that is one of the sections
at the bottom (no smb_enact needed
jaylagorio
@jaylagorio
Oct 06 2017 16:38
Except we definitely disagree about no bolus needed. And when I added the carbs I was somewhere around 130, had a low temp target in place, and there was no initial SMB like normal
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 16:39
I tend to have that same issue with eating and letting openaps take care of things
jaylagorio
@jaylagorio
Oct 06 2017 16:39
I haven't - it's been great so far, I'm really not sure what's different
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 16:39
what I am trying now is putting carbs in NS 30 minutes early and watching the logs
I am also using the eating soon settings
this is what my logs currently show
 Microbolusing 1U. ","COB":95,"eventualBG":401,"suspended":false,"duration":120,"tick":-3,"deliverAt":"2017-10-06T16:35:43.768Z","received":true,"requested":{"units":1,"params":[0,40]},"temp":"absolute","predBGs":{"UA
jaylagorio
@jaylagorio
Oct 06 2017 16:40
Setup command line, preferences.json... there isn't another file where these preferences are that I just forgot about, right?
Yeah, not getting any of that here
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 16:41
this caught my attention in your file but I am not you...
"enableSMB_with_bolus": false,
jaylagorio
@jaylagorio
Oct 06 2017 16:42
Yeah, I have SMB on with temps and with COB, but I intentionally didn't want to use boluses that I perform as triggers to enable SMB
That's a setting I've had in place for weeks and it's been fine.
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 16:43
cool, mine is different but I am sure that does not mean much
are you seeing this in your logs at all?
Starting supermicrobolus pump-loop at Fri Oct 6 11:43:02 CDT 2017 with 3 second wait_for_silence:
Waiting up to 4 minutes for new BG: ......
jaylagorio
@jaylagorio
Oct 06 2017 16:44
Yup, that's all happening as normal
Starting supermicrobolus pump-loop at Fri Oct 6 12:44:02 EDT 2017 with 17 second wait_for_silence:   
Waiting up to 4 minutes for new BG: .glucose.json newer than pump_loop_completed                     
Radio ok. Listening: .No pump comms detected from other rigs                                         
Preflight
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 16:45
:( then I am no help to you, sorry my friendc
jaylagorio
@jaylagorio
Oct 06 2017 16:45
No worries! But yeah, this is super weird. Everything was going great for so long!
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 16:46
I am sure someone will have answers for you, it has been slow in there but people have been in and out all day
I am going to lunch before I go low
katie disimone
@Kdisimone
Oct 06 2017 16:49
@jaylagorio when you say you blew away the whole directory...do you mean you deleted your myopenaps directory and then rebuilt it? If you did that after midnight, you probably aren’t getting smbs because your autotune hasn’t been run yet in the new myopenaps directory. There’s a line in the pump-Loop that says “checking to see if autotune exists....”. Do you see that line?
Scott Leibrand
@scottleibrand
Oct 06 2017 17:10
heh, @Kdisimone beat me to it. that's what @diabetic_dude_twitter was asking about too I think. It's up toward the top right after "starting supermicrobolus pump-loop"
jaylagorio
@jaylagorio
Oct 06 2017 17:16
So believe it or not just before I ran out for 15 minutes I remembered that I forgot to move my autotune directory back over, so I did that a few cycles ago. The Hoorah! entry is back now
Luckily the combination of high temping and manual bolus kept me under 200 and I think will land me somewhere around 100.
I'll remember this for next time
Why does autotune need to be enabled for SMB? Just wanting to have the very best data when doing anything bolus related because it's more dangerous?
Scott Leibrand
@scottleibrand
Oct 06 2017 17:22
Yep. Having a bad CR is a good way to get post-meal lows and then have oref0 counteract your rescue carbs and make it even worse later.
jaylagorio
@jaylagorio
Oct 06 2017 17:23
CR - right, that makes sense
Dana Lewis
@danamlewis
Oct 06 2017 17:30
even more important with the adjustments to SMB in 0.6.0
jaylagorio
@jaylagorio
Oct 06 2017 17:30
If the autotune directory isn't present but the user has enabled SMB features can a log line be produced? An anti-Hoorah! log?
Scott Leibrand
@scottleibrand
Oct 06 2017 17:31
Should be already. Check your older logs
jaylagorio
@jaylagorio
Oct 06 2017 17:31
I must have missed that. I'm particularly observant today.
jaylagorio
@jaylagorio
Oct 06 2017 17:32
Very particularly observant.
jaylagorio
@jaylagorio
Oct 06 2017 17:39
Just got a microbolus. Looks like things are back on track!
ejoe132
@ejoe132
Oct 06 2017 17:40
@diabetic_dude_twitter Why rename the folders? I guess my thinking is the issue right now is that it says its in read only ?
Dana Lewis
@danamlewis
Oct 06 2017 17:40
@jaylagorio :+1:
@ejoe132 rename so you move and create a backup copy in case you need something in it
@jaylagorio also, sounded like your no-git install failed? may want to look into that, because i think you were on no-git before?
ejoe132
@ejoe132
Oct 06 2017 17:43
@danamlewis Guess Im a lil confused, shouldnt I try to just figure out how to wipe the thing and do a new fresh install? cause for some reasn now I cant really do anything on it
jaylagorio
@jaylagorio
Oct 06 2017 17:43
The --nogit command line parameter in setup produced an unrecognized argument message. I retried with --no-git and got the same message. I was not no-git before, but resetting git was working. This time it didn't, so I nuked the whole directory and started from scratch.
Dana Lewis
@danamlewis
Oct 06 2017 17:43
ok. if you want nogit, I'd recommend following the flow we've outlined here:
jaylagorio
@jaylagorio
Oct 06 2017 17:43
Well, renamed it and moved it somewhere else. Thankfully I did that because I forgot to replace my autotune data
Dana Lewis
@danamlewis
Oct 06 2017 17:44
Screen Shot 2017-10-06 at 10.43.57 AM.png
at this point, I'm thinking we're going to want to make nogit default since people installing in place (and not following these docs) have run into so many issues
jaylagorio
@jaylagorio
Oct 06 2017 17:47
I think I've done effectively what's right there. I for sure don't have a .git directory in my openaps directory.
renegadeandy
@renegadeandy
Oct 06 2017 17:57
Hi all, can anybody point me at the libre instructions to get it hooked up to nightscout / openaps?
ejoe132
@ejoe132
Oct 06 2017 18:10
Is there an easy way to wipe the edison completely? Keep trying to re run everything but keeps freezing then watchdog comes up and restarts it. Almost ready to just order new edison and board and start from scratch
Dana Lewis
@danamlewis
Oct 06 2017 18:19
@ejoe132 just reflash from scratch if you want to do that
Michael Spradling
@M1Sports20
Oct 06 2017 18:35
I have ben using 6.0 dev branch, i think i like it much better. However, after putting carbs in and take a bolus, should it be putting a 0.0/h basal in for a period of time?
if it is suppose to do this i believe it is too long and is off setting my bolus and causing me to go high
other than that I feel I actually am getting better results.
Dana Lewis
@danamlewis
Oct 06 2017 18:36
@M1Sports20 that means it thinks you've done too much bolus for your autotuned-carb ratio, and is trying to ward off a low. check and see - is the carb ratio you're manually bolusing on different from the autotune carb ratio?
Michael Spradling
@M1Sports20
Oct 06 2017 18:38
@danamlewis: Thanks. I do run autotune, but haven't updated my pump settings with the results of autotune yet. I guess I have to update the bolus wizard settings if I feel they are a better fit
However, I believe I don't have to update my basals from autotune, since openaps can do that for me with temp basals.
correct?
Dana Lewis
@danamlewis
Oct 06 2017 18:40
a couple of things: as I suggested, check and see if the carb ratio openaps is using in autotune is diverging from what you're using in bolus wizard. if it is, that's likely why you're seeing the low temps set when bolus. you can a) adjust carb ratio in bolus wizard, or b) consider trying partial bolus for meals. (or both).
I would look and see if autotune basals are being capped by the safety guards. if so, it can't tune any further, and you may want to adjust your underlying pump basals to give autotune more leeway if you feel that is correct directionally.
Michael Spradling
@M1Sports20
Oct 06 2017 18:41
My autotune Car Ratio is 9.88, my pump is set to 10. So they are not that to far off
basels are also not capped, just checked that
Dana Lewis
@danamlewis
Oct 06 2017 18:41
k and k
I'd try partial bolus for the meal, then, and see if you get a different outcome
Michael Spradling
@M1Sports20
Oct 06 2017 18:41
so let smbs handle the rest then?
ejoe132
@ejoe132
Oct 06 2017 18:41
@danamlewis Cant find in the docs anymore about flashing jubilinux, am I missing it? see it saying to recommend buy edison with it pre installed but cant find the old pages with the instructions for that
Dana Lewis
@danamlewis
Oct 06 2017 18:42
@M1Sports20 yes. If you've been doing 100% bolus, I'd try something like 75% for a meal or two, observe difference, and work down from there until you find a happy medium.
Michael Spradling
@M1Sports20
Oct 06 2017 18:42
@danamlewis, will try. Thanks
Dana Lewis
@danamlewis
Oct 06 2017 18:42
I've cut out all of my post-meal lows (which I would usually get from dinner with a walk home and then a shower expediting the insulin activity further)
@ejoe132 it's in Resources: http://openaps.readthedocs.io/en/latest/docs/Resources/index.html -- there's the all types, mac, and pc flashing pages still
scroll down and pick your page of choice
Michael Spradling
@M1Sports20
Oct 06 2017 18:44
@danamlewis are you still on fiasp? Wondering how its going the longer you are on it. It seems people get resistive to it or something with more time
Dana Lewis
@danamlewis
Oct 06 2017 18:45
@M1Sports20 yes I am. about 3 months now. some people do have less optimal expereinces further on, but not all. like all things, YDMV. (same for some people who swear by novolog vs. humalog or vice versa)
Michael Spradling
@M1Sports20
Oct 06 2017 18:46
@danamlewis, exciting. Hope to try it as soon as the US supply gets here
Although, I guess its not approved for pumps in the US, I wonder if I can get a prescription then
Dana Lewis
@danamlewis
Oct 06 2017 18:49
endo's write prescriptions off label all the time
br
@brmccollum
Oct 06 2017 18:54
When I log into my rig, I have two "myopenaps" files. One within itself, "~/myopenaps/myopenaps# "
Is this the normal hierarchy?
br
@brmccollum
Oct 06 2017 18:59
My rig is pretty inconsistent on when it connects, and certain alias don't work like the one for autosens, and I was wondering if having two 'myopenaps' folders may be causing issues.
These are the common items that show up while watching my logs. (Please also note, that sometimes my rig will work great for 6+ hours)
  1. IndexError: bytearray index out of range
  2. glucose.json newer than pump_loop_completed
    mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?
  3. AttributeError: 'ReadPumpModel' object has no attribute 'read_status'
  4. RefreshValueError: need more than 0 values to unpack
Any suggestions by anyone would be appreciated.
ejoe132
@ejoe132
Oct 06 2017 19:02
@danamlewis thank you
Tim Street
@tim2000s
Oct 06 2017 19:12
two rigs running 0.6.0 in dev. First has this in enacted:
{"iob":4.443,"activity":0.0407,"bolussnooze":0.895,"basaliob":0.688,"netbasalinsulin":3.3,"hightempinsulin":8.5,"microBolusInsulin":5.1,"microBolusIOB":2.165,"time":"2017-10-06T19:06:42.000Z","lastBolusTime":1507311939000} and the second this:
{"iob":1.959,"activity":0.0299,"bolussnooze":0.402,"basaliob":0.481,"netbasalinsulin":3.85,"hightempinsulin":9.2,"microBolusInsulin":5.9,"microBolusIOB":1.225,"time":"2017-10-06T18:53:04.000Z","lastBolusTime":1507311939000}
ANy idea why the difference?
whilst the time is slightly different, it doesn't account for the difference in iob
garykidd
@garykidd
Oct 06 2017 19:26
Would love it if a script could be developed to handle BT tethering sequence---though I think I can almost handle it without looking at docs now.
David
@davidkeddydb
Oct 06 2017 19:28
@garykidd Funny you should say that. One of my rigs just stopped working BT tethering and I can't get the darn thing to restart bt tethering. I have removed the device, I have re-entered everything, re-paired. nothing. I always end up using the hotspot as I can't get it to go.
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 19:30
when I get BT issues I reboot my phone and that fixes my issues 70% of the time. If the phone reboot doesnt fix the issue I reboot the rig, if that doesnt work I start the hotspot
garykidd
@garykidd
Oct 06 2017 19:30
I'm stillhaving the inverse problem. connects to BT
then doesnt ever disconnect
are you using iphone or android
David
@davidkeddydb
Oct 06 2017 19:31
@diabetic_dude_twitter Will try all that. @garykidd Using an iphone
garykidd
@garykidd
Oct 06 2017 19:32
not sure it'll help but I think when I was using iphone and having a similar issue--Ihad to remove the device form the iphone
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 19:34
what happens if you just turn off your hotspot for like 15 seconds?
that is what I do when I cant get BT to disconnect
Jake Punshon
@JakePunshon
Oct 06 2017 19:35
is there an approved method for turning off pushover?
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 19:35
(that is also how I discovered that I was having connection issues with my wifi at work)
how can I fix this .....
 insulinReq 1.01. Microbolusing 0.5U. temp 1.85<3U/hr. ","COB":3,"eventualBG":150,"suspended":false,"duration":30,"tick":"+1","deliverAt":"2017-10-06T19:35:26.723Z",
I would like to see SMB giving me 1.01 units at this point
this is a more complete picture
 enact/bolused.json: {"insulinReq":1.01,"status":"normal","_type":"BolusRequest","bg":323,"bolusing":true,"reservoir":"114.9","snoozeBG":154,"timestamp":"2017-10-06T14:35:56.660166","raw":"00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000","reason":"COB: 3, Dev: 79, BGI: -12.99, ISF: 58, Target: 80, minPredBG 139, minGuardBG 145, IOBpredBG 137, COBpredBG 78, UAMpredBG 145;  insulinReq 1.01. Microbolusing 0.5U. temp 1.85<3U/hr. ","COB":3,"eventualBG":150,"suspended":false,"duration":30,"tick":"+1","deliverAt":"2017-10-06T19:35:26.723Z","received":true,"requested":{"units":0.5,"params":[0,20]},"temp":"absolute","predBGs":{"UAM":[323,322,320,317,314,309,304,299,292,286,278,271,262,253,245,237,229,222,215,209,203,198,193,188,184,180,177,173,170,167,165,162,160,158,156,155,153,152,151,150,149,148,147,147,146,146,145],"COB":[323,319,310,297,285,272,260,248,237,226,215,205,196,187,178,170,162,155,149,142,137,131,126,122,117,113,110,106,103,100,98,95,93,91,89,88,86,85,84,83,82,81,80,80,79,79,79,78],"aCOB":[323,317,309,299,286,274,261,250,238,227,217,207,197,188,180,171,164,157,150,144,138,133,128,123,119,115,111,108,105,102,99,97,95,93,91,89,88,87,85,84,83,83,82,81,81,80],"IOB":[323,322,320,317,313,308,302,296,289,281,273,264,254,245,237,229,221,214,207,201,195,190,185,180,176,172,168,165,162,159,156,154,152,150,148,146,145,144,142,141,141,140,139,139,138,138,137]},"rate":3,"units":0.5,"IOB":4.321}
Settings less than 15 minutes old. Edison battery at 69% is charged (>= 98%) or likely charging (60-70%). pumphistory-24h refreshed
Settings refreshed. Ref
Radio ok. Listening: .No pump comms detected from other rigs
Refreshed pumphistoryRemoved duplicate bolus/carb entries:4
 and meal.json
IOB: 4.796
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 19:40
my meal test went very badly, SMB started out very aggressive then after carbs started hitting it was weak on the corrections
        "max_iob": 30,
        "max_daily_safety_multiplier": 4,
        "current_basal_safety_multiplier": 5,
        "autosens_max": 1.2,
        "autosens_min": 0.7,
        "rewind_resets_autosens": true,
        "adv_target_adjustments": true,
        "unsuspend_if_no_temp": true,
        "enableSMB_with_bolus": true,
        "enableSMB_with_COB": true,
        "enableSMB_with_temptarget": true,
        "enableUAM": true,
        "curve": "ultra-rapid",
        "maxSMBBasalMinutes": 105
garykidd
@garykidd
Oct 06 2017 19:45
can someone point me to docs on latest dev version, please?
philipgo
@philipgo
Oct 06 2017 19:54
I am getting the subg_rfspy error using the Explorer Board. How do I interrupt the reboot-loop so I can at least try to fix the problem?
David
@davidkeddydb
Oct 06 2017 19:56
@diabetic_dude_twitter Are you sure your ISF is right that time of day?
@drnoname82 you can go into crontab and comment stuff out. Directions are in troubleshooting
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 19:59
@garykidd I think the main question is has it been written yet. I normally look to my email because I subscribe to the github thread and search for topics until I find what I am looking for in my email.
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 20:01
@davidkeddydb my setting on my pump that I got from autotune about a month ago WAS 47
that is a diff of around 10 :(
Dana Lewis
@danamlewis
Oct 06 2017 20:02
@diabetic_dude_twitter old autotune had a known upward bias of ISF and carb ratio. the current version in dev has improvements to resolve that. if you changed your pump ISF or carb ratio based on old autotune, I'd look at running autotune on dev and using something closer to current if you are having issues
@brmccollum looks like you did an install inside an install
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 20:02
@danamlewis trying and updating settings now
thnx
Dana Lewis
@danamlewis
Oct 06 2017 20:03
@garykidd 0.6.0 notes are in the open PR here: openaps/oref0#694
garykidd
@garykidd
Oct 06 2017 20:07
@danamlewis thx.
I'm just a little confused on the requirement for DIA to be 5 or 6.
philipgo
@philipgo
Oct 06 2017 20:08
Thanks, but I can hardly enter more than one command before it reboots agsin
Dana Lewis
@danamlewis
Oct 06 2017 20:08
@drnoname82
garykidd
@garykidd
Oct 06 2017 20:08
I've always used 4 DIA w/ novolog
Dana Lewis
@danamlewis
Oct 06 2017 20:08
sudo service cron stop to stop cron temporarily
philipgo
@philipgo
Oct 06 2017 20:11
Thank you @danamlewis That is what I tried after reading the docs, but the loop keeps running and reboots a few seconds later
Dana Lewis
@danamlewis
Oct 06 2017 20:12
check and see if you have multiple crons somehow
you may also want to killall -g oref0-pump-loop
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 20:21
@danamlewis just to confirm I only need to change my setting ISF on my pump with the updated setting. (and only for note taking on NS if I so choose to)
philipgo
@philipgo
Oct 06 2017 20:21
I tried both commands before and again now and it keeps rebooting
garykidd
@garykidd
Oct 06 2017 20:21
I guess it's just that the tail is so weak (there but not very impactful to BGI) and that's why it hasnt been an issue w/ DIA set to 4...for all these years
Dana Lewis
@danamlewis
Oct 06 2017 20:22
yes* (but autotune won't know about your new current until it runs again at midnight, so you may want to do a manual run before then if you'd like to see the tunes from the new baseline before then)
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 20:43
that is the article that made me change mine from 4 to 5 after over 10 years
(or convinced I should say)
Tim Street
@tim2000s
Oct 06 2017 20:45
Wow. :+1: @diabetic_dude_twitter
philipgo
@philipgo
Oct 06 2017 20:57
Ok, finally tackled the reboot loop by deleting the crontab temporarily
Dana Lewis
@danamlewis
Oct 06 2017 21:03
anyone with a Mac want to run this autotune PR for me, confirm it works, so we can get the dev-based autotune changes into master for all the one-off users to use? openaps/oref0#693 (see instrcutions at the bottom for pulling that branch)
dmdfreak
@dmdfreak
Oct 06 2017 21:04
```
From git://github.com/openaps/oref0
ccf11cc..8be8b33 dev -> origin/dev
  • [new branch] autotune-patch -> origin/autotune-patch
  • [new branch] blend-minGuardBG -> origin/blend-minGuardBG
  • [new branch] carbsReqBy -> origin/carbsReqBy
  • [new branch] exercise-mode -> origin/exercise-mode
  • [new branch] insulinPeakTime -> origin/insulinPeakTime
  • [new branch] no-basal-v-bolus -> origin/no-basal-v-bolus
  • [new branch] no-git-default -> origin/no-git-default
  • [new branch] offline -> origin/offline
  • [new branch] raw_safety_multiplier -> origin/raw_safety_multiplier
  • [new branch] scott-dana-dev -> origin/scott-dana-dev
  • [new branch] xDrip-warning -> origin/xDrip-warning
  • [new tag] 0.6.0-dev1 -> 0.6.0-dev1
    Updating 647cf40..8be8b33
    error: Your local changes to the following files would be overwritten by merge:
     bin/oref0_nightscout_check.py
    Please, commit your changes or stash them before you can merge.
    Aborting
Dana Lewis
@danamlewis
Oct 06 2017 21:04
git stash
dmdfreak
@dmdfreak
Oct 06 2017 21:05
was that for me @danamlewis ? I didnt mean to put that question up yet LOL
Dana Lewis
@danamlewis
Oct 06 2017 21:05
yup ;)
dmdfreak
@dmdfreak
Oct 06 2017 21:05
that was quick LOL
thank you :) Do I rerun the pull now?
Dana Lewis
@danamlewis
Oct 06 2017 21:07
yea, do git stash, then run whateve ryou tried to do before
dmdfreak
@dmdfreak
Oct 06 2017 21:08
perfect. thank you again. I'm updating my rig. I'm way behind on the dev branch.
Dana Lewis
@danamlewis
Oct 06 2017 21:08
:+1:
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 21:18
@danamlewis to confirm you would like someone to connect to their rig on a mac and run autotune?
after updating per the instructions??? to test correct?
Dana Lewis
@danamlewis
Oct 06 2017 21:18
@diabetic_dude_twitter no, just download the code and run autotune on the mac, not via the rig
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 21:18
ok working on it now
on the mac not the rig
Dana Lewis
@danamlewis
Oct 06 2017 21:19
you'll follow the phase C instructions (one off) and Mac path if you haven't done it on the mac before
yep, k. thanks for testing
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 21:19
yup have to do that I have it on my macbook but not my macbook air
Dana Lewis
@danamlewis
Oct 06 2017 21:46
Anyone who currently has a rig that still has git (@jaylagorio? others?) and wants to test auto-removal of git, I've created a branch that will auto-remove git without having to do the move old directory step. openaps/oref0#699
dmdfreak
@dmdfreak
Oct 06 2017 21:48
Installed /home/.rootfs/usr/local/lib/python2.7/dist-packages/decocare-0.0.31-py2.7.egg
Processing dependencies for decocare
Finished processing dependencies for decocare
Cloning decocare 0.1.0-dev: Cloning into 'decocare'...
remote: Counting objects: 11544, done.
Receiving objects: 15% (1829/11544), 2.11 MiB | 26.00 KiB/s
appears to have stopped here....
Hasn't moved in 10 minutes
Dana Lewis
@danamlewis
Oct 06 2017 21:49
control C and do whatever you're doing again
dmdfreak
@dmdfreak
Oct 06 2017 21:50
ok thank you
Scott Leibrand
@scottleibrand
Oct 06 2017 21:52
@diabetic_dude_twitter when you have an insulinReq of 1.01, it should only microbolus 1/2 of that at most (0.5U), and then re-evaluate on the next oref0-pump-loop run and do more. when you don't have any uncovered carbs and it's making decisions based off UAM, it'll only SMB 1/3 of the insulinReq at a time. Those limitations are to avoid overreacting to any single BG data point.
@brmccollum if you have ~/myopenaps/myopenaps that is an indication you accidentally installed one instance inside the other. I would back up your preferences.json and autotune directories, then rm -rf ~/myopenaps and re-run oref0-setup to clean it up.
@M1Sports20 @danamlewis if OpenAPS is predicting lows after a full meal bolus and the resulting low temps are sending you high, it's likely your CR is too weak (high) or basals are too low, or possibly ISF is too weak (high). depending on how long you've been running 0.6.0-dev autotune, it may still be catching up with your actual settings. reducing meal bolus amounts likely will only help in that situation if you also have post-meal lows. if not, then we need to figure out why oref0 is predicting BG to drop and it actually isn't, and possibly figure out why autotune hasn't corrected for that yet.
Dana Lewis
@danamlewis
Oct 06 2017 22:00
ok openaps/oref0#699 got thumbs up (no git auto fix), so that'll make everyone's dev installs easier moving forward :) big thanks @Kdisimone for testing!
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 22:01
thanks @Kdisimone it was going to be a long time until I got there
Dana Lewis
@danamlewis
Oct 06 2017 22:01
@diabetic_dude_twitter thrilled you're doing the autotune mac testing for me :)
big :clap: for all the testers and helpers!! helps move things along into dev much more quickly :confetti_ball:
Dana Lewis
@danamlewis
Oct 06 2017 22:08
simplified the 0.6.0 testing instructions - aka, now follow the normal dev upgrade path in the docs :) https://github.com/openaps/oref0/pull/694#issue-263319617
br
@brmccollum
Oct 06 2017 22:11
Thanks @danamlewis & @scottleibrand I will follow your suggestion.
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 22:17
still trying @danamlewis getting anb error because I do not have the /src foplder on my mac
trying to move that one there now
garykidd
@garykidd
Oct 06 2017 22:27
@danamlewis wondering if runagain might be mentioned w/r/t your last post.
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 22:32
@danamlewis I am so sorry but I have to meet my wife and kids for dinner
Dana Lewis
@danamlewis
Oct 06 2017 22:32
no worries
Diabetic_Dude
@diabetic_dude_twitter
Oct 06 2017 22:32
I will test at home on my other mac that is set up
later
Dana Lewis
@danamlewis
Oct 06 2017 22:32
:+1:
iananders
@iananders
Oct 06 2017 22:49
odd small issue- just updated to 0.6.0-dev and everything is running flawlessly - but in my openAPS pill in nightscout shows "not enacted" (there's an x before the last time openaps ran) even though it is indeed enacting the recommendations, basal changes and SMB
Dana Lewis
@danamlewis
Oct 06 2017 22:50
Screen Shot 2017-10-06 at 3.50.21 PM.png
garykidd
@garykidd
Oct 06 2017 22:56
can someone point me as to how I update NS?
iananders
@iananders
Oct 06 2017 22:58
@danamlewis thanks dana!
katie disimone
@Kdisimone
Oct 06 2017 23:05
@garykidd I never put a page in OpenAPS for that, but I do have one in Loop docs. Same exact idea. Update your cgm-remote-monitor repo and the. Redeploy. Here’s the page. https://loopkit.github.io/loopdocs/nightscout/update_user/#updating-your-repository-in-github
jaylagorio
@jaylagorio
Oct 06 2017 23:23
Thanks, @danamlewis! I'm git-free now though
Dana Lewis
@danamlewis
Oct 06 2017 23:23
@jaylagorio no worries. tested and fixed for everyone and all the other rigs and good to go
garykidd
@garykidd
Oct 06 2017 23:51
hmmm nearly going crazy here....cant figure out why my rig won't disconnect from BT tether upon returning to proximity of home wifi network. Can anyone help me to troubleshoot?
garykidd
@garykidd
Oct 06 2017 23:58
rig will BT tether automatically even if it's in presence of home wifi.
I even reflashed the thing
Scott Leibrand
@scottleibrand
Oct 06 2017 23:59
@garykidd we're still waiting for you to post your logs
in this case the network.log (from wifi) to start with
garykidd
@garykidd
Oct 06 2017 23:59
Oh, OK great. missed that