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

2nd
Jun 2017
cameronrenwick
@cameronrenwick
Jun 02 2017 00:12
@danamlewis sorry to re-raise the question...but any further thoughts regarding my wifi problem with network name/username/pw? Regrettably the doc info doesn't let me connect and google isn't helping..
Cameron Chunn
@applehat
Jun 02 2017 00:13
Is it normal for the explorer board to have the Could not get sug_rfpsy state when it first starts? This was working yesterday reading my 754, and now that I have the 722 configured i haven't seen it successfully read yet - tho Im probably being excited and impatient.
Dana Lewis
@danamlewis
Jun 02 2017 00:14
haven’t done that myself, @cameronrenwick - but @ELUTE is the one who figured that out so she might have some ideas on what else to try? (or @Kdisimone , depending on who is around)
Cameron Chunn
@applehat
Jun 02 2017 00:14
@cameronrenwick I wasn't around - whats your wifi problem?
cameronrenwick
@cameronrenwick
Jun 02 2017 00:16
so my prob is, I'm at a university for the weekend. trying to connect my rig to the wifi. My comp and phone are on the wifi so the network name, user name and password I have are good. My rig won't connect. I've followed the docs regarding adding these 3 things but no go. I'm trying to figure out how to connect
katie disimone
@Kdisimone
Jun 02 2017 00:16
@cameronrenwick can you copy paste what you have for the network from your wpa supplicant?
cameronrenwick
@cameronrenwick
Jun 02 2017 00:17
here's the code from the docs
network={ scan_ssid=1 ssid="network name" password="wifi password" identity="wifi username" key_mgmt=WPA-EAP pairwise=CCMP TKIP group=CCMP TKIP WEP104 WEP40 eap=TTLS PEAP TLS priority=1 }
sorry - spacing didn't work
network={
scan_ssid=1
ssid="network name"
password="wifi password"
identity="wifi username"
key_mgmt=WPA-EAP
pairwise=CCMP TKIP
group=CCMP TKIP WEP104 WEP40
eap=TTLS PEAP TLS
priority=1
}
there is how it should look. I've replaced the pw, un, network but no go
the log just says error cycling network
Cameron Chunn
@applehat
Jun 02 2017 00:21
could be a different parwise group or eap
your phone and your computer negotiate all of that automatically
katie disimone
@Kdisimone
Jun 02 2017 00:21
Oh I know what the docs have lol...I was hoping to double check your rig's entry. Check for simple typos. Sometimes a second set of eyes is a good place to start when something that "should" work isn't working.
cameronrenwick
@cameronrenwick
Jun 02 2017 00:22
ya I think it's there and I actually drilled down in the university network stuff and it says ccmp for pairwise and TTSL for eap
Cameron Chunn
@applehat
Jun 02 2017 00:22
try removing the TKIP from pairwise and PEAP TLS from eap
cameronrenwick
@cameronrenwick
Jun 02 2017 00:22
lol... yes Katie "we've" been thru that ; ) but no, my entries are typo free (wish they weren't!)
rsilvers129
@rsilvers129
Jun 02 2017 00:29
@applehat I have had lots of those radio errors. Had them with a friend’s setup last night. I had him put the Edison on his other Explorer board, and it worked right away. I had the same problem with mine, and my other Explorer worked with the same Edison.
cameronrenwick
@cameronrenwick
Jun 02 2017 00:34
ya I'm gonna have to dig futher on this later.. I'll just wifi tether for now... too many things going on!!
thx for everyone's help tho!
Cameron Chunn
@applehat
Jun 02 2017 00:35
@rsilvers129 completely removing power fixed it
rsilvers129
@rsilvers129
Jun 02 2017 00:35
@applehat I will have him try that some time. Thanks.
rsilvers129
@rsilvers129
Jun 02 2017 00:46
@danamlewis Just wonderinf if you think it is best to full bolus or semi-bolus with SMB. What have you found to be the best SMB strategery for pizza or pasta? I just ate 107 grams carb of ravioli (460 grams of food before cooking - four “servings.”) Also had 22 grams of cheese and 10 of olive oil. My bolus calculation said I should need 14 units of insulin. I was at 130 BG level and dropping, but I did an “eating soon” for about 15 minutes. I also told the rig that I was going to eat 105 grams of carb. I saw SMB give me 0.4 units. Once I actually started eating, I did a manual bolus of 10 units (about 60% of what I think I needed). That was 70 minutes ago and I am at 107 BG now. My plan is to let SMB fill in the difference. For pizza I am thinking just bolus 50% to avoid a post-prandial low and let SMB do the rest.
Scott Leibrand
@scottleibrand
Jun 02 2017 00:54
for anything that will take more than DIA hours to absorb, you'll want to do a partial bolus and let the rest get filled in later by SMB or AMA
Dana Lewis
@danamlewis
Jun 02 2017 01:16
YDMV. I bolus based on knowing my carbs absorption is about 30g/hour, so I usually do no more than about 2 hours worth of carbs since insulin will be peaking before more absorb. also might be more/less depending on how I have SMB trigger ahead of time.
also been experimenting with pre-entering carbs by a little bit (i.e. when ordering food) to let SMB start ramping up, before doing my partial meal bolus once I start eating.
rsilvers129
@rsilvers129
Jun 02 2017 01:33
I think you need a t-shirt that says YDMV
Yes I have had meals in the last week that seemed under control, but 4 hours later, I went up and up.
garykidd
@garykidd
Jun 02 2017 01:37
what's the latest version of jubilinux that's recommended/stable to flash...
do I want .... kernel-v0.1.1.zip ?
garykidd
@garykidd
Jun 02 2017 01:44
thinking...this is right..... jubilinux-v0.1.1.zip
actually...
Dana Lewis
@danamlewis
Jun 02 2017 01:52
@applehat I just saw your new twitter bio. Related to this morning's convo I assume? Lol. (And delayed lol to this morning)
Cameron Chunn
@applehat
Jun 02 2017 01:52
My ugly purple 722? Yes =)
oh wait. my bio.
garykidd
@garykidd
Jun 02 2017 01:53
I love purple Cam...
Cameron Chunn
@applehat
Jun 02 2017 01:54
I don't even remember what we talked about this morning @danamlewis -- Been ignoring work and catching up on oref0 docs all day to get prepped
Dana Lewis
@danamlewis
Jun 02 2017 01:54
Something something I'm evil something? :smile: or whenever that convo was.
Cameron Chunn
@applehat
Jun 02 2017 01:55
lol, :shrugs:
Note to self: check that pump fits into case you print before super gluing only belt clip to case.
garykidd
@garykidd
Jun 02 2017 01:57
@scottleibrand was wondering if hooking up a few LEDs to GPIO pins might be helpful---and then I thought---well---maybe this has already been addressed as well?
Cameron Chunn
@applehat
Jun 02 2017 01:57
i'd like to see an eInk screen with the latest loop status on it personally.
garykidd
@garykidd
Jun 02 2017 01:58
I'm thinking we'll get pointed to watches...
@applehat NAIL POLISH REMOVER....WORKS WONDERS..
live4sw
@live4sw
Jun 02 2017 01:58
Silly question, but I'm trying to set up a stationary explorer board in a place not close to an outlet. Reasonable to just get a high capacity power bank and have that connected to the explorer with the lipo plugged in, and periodically recharge the power bank?
Dana Lewis
@danamlewis
Jun 02 2017 01:59
Yep
live4sw
@live4sw
Jun 02 2017 01:59
i figure the 20,100mAH powerbank should last at least a couple days at a time
thanks!
garykidd
@garykidd
Jun 02 2017 01:59
or a long extension cord
Cameron Chunn
@applehat
Jun 02 2017 01:59
@garykidd nail polish remover works wonders for what? removing super glue?
garykidd
@garykidd
Jun 02 2017 02:00
maybe plugged into your neighbors house while you're at it
@applehat roger that
yes
i've released the bond on my fingers being stuck together more than once... but also succesfully glued hang-nails down without incdedent
where's @bewest ...... not allowed to be here anymore????
@danamlewis @scottleibrand I was so excited to get a MM670g at one point...now---I'm like no way---stick with my 723 openaps...
Cameron Chunn
@applehat
Jun 02 2017 02:20
@garykidd the medtronic pump clips are ABS plastic....wanna guess how I found out? lol
managed to get it off and only slightly melted part of it.
now to scale up the case 2% and print it again.
Cameron Chunn
@applehat
Jun 02 2017 02:29
Man i am SO excited. Watching OpenAPS high-temp on a erroneous sensor reading and then correct itsself on the next one is so cool.
Cant wait to run for a bit and then try out oref1 stuff =O
Its hard to put into words how watching a tiny computer handle something thats been a daily struggle for you for 14 years feels. I'm sure you guys know.
Dana Lewis
@danamlewis
Jun 02 2017 02:34
<3 :raised_hands:
garykidd
@garykidd
Jun 02 2017 02:35
absolutely
John Benjamin
@Bender1061
Jun 02 2017 02:36
:+1: to that!
Cameron Chunn
@applehat
Jun 02 2017 03:13
I ate carbs, bolused as I was eating, and haven't spiked yet. What happened? Where is my annoying 160s number?
Dana Lewis
@danamlewis
Jun 02 2017 03:16
You must've eaten something with cinnamon...or gotten your rig going ;) :) in time for dinner! :D
Cameron Chunn
@applehat
Jun 02 2017 03:16
....what does cinnamon do to blood glucose?
Dana Lewis
@danamlewis
Jun 02 2017 03:17
(I'm joking, as it's one of the internets great secrets to curing diabetes) (any time I mention cinnamon, I am 100% joking)
garykidd
@garykidd
Jun 02 2017 03:19
that's right up there with the notion that eating too much candy game me/you diabetes...
Matthias
@ceben80
Jun 02 2017 03:20
@applehat you're cured as well :stuck_out_tongue_winking_eye: :smile:
garykidd
@garykidd
Jun 02 2017 03:22
I'm guessing no...but is it possible newer edison modules run any cooler than previous models....weird that this new one is so cool (literally).
Dana Lewis
@danamlewis
Jun 02 2017 03:23
Not by anything we've done
garykidd
@garykidd
Jun 02 2017 03:27
@danamlewis Do you happen to know of anyone using GPIO pins for any kind of visual indication of what rig is doing (understanding pretty much everything you need is on NS site inmost instances)
thought it mightbe cool to see if you have internet (wifi connection and/or BT connection via LED illuminating for instance)
Dana Lewis
@danamlewis
Jun 02 2017 03:33
Not that I know of. I do know someone who's experimenting w some kind of visual display, should have a prototype done in next few weeks.
garykidd
@garykidd
Jun 02 2017 03:34
neat
if you have 2 rigs running side-by-side....should they alternate setting temps...or should only one keep setting temps...
Dana Lewis
@danamlewis
Jun 02 2017 03:36
Both will keep setting temps but depends on what they're hearing when it's their turn to listen
garykidd
@garykidd
Jun 02 2017 03:37
is it better to somehow manually offset one form the other....thinking its not necessary...
Dana Lewis
@danamlewis
Jun 02 2017 03:38
Nope
Wait for silence is already built in. That's why you see "no pump comms detected" which means green light for that rig who's asking the question to talk to the pump next
Cameron Chunn
@applehat
Jun 02 2017 03:40
@garykidd are there gpio pins on the exporer board?
Or are you talking about stacking another hat / shield / whatever dumb term Edison uses.
garykidd
@garykidd
Jun 02 2017 03:41
yes; atleast I think so...i think those are the 10 traces on botton of board...
Cameron Chunn
@applehat
Jun 02 2017 03:42
Oooh.
applehat @applehat goes and looks
garykidd
@garykidd
Jun 02 2017 03:43
need to confirm.
Cameron Chunn
@applehat
Jun 02 2017 03:53
Could use one of these with it https://www.sparkfun.com/products/13036
Tho Im not sure if the fact that it uses UART would effect the usability of the radio. Need to read more.
Cameron Chunn
@applehat
Jun 02 2017 03:59
Or the I2C block to use with a display (I like this idea)
Dana Lewis
@danamlewis
Jun 02 2017 04:08
@garykidd fyi - some of your PRs have been targeting dev docs. feel free to target master w/ those.
(you can go in and edit the existing PRs if you want to target master)
Cameron Chunn
@applehat
Jun 02 2017 04:10
Per the GPIO question above, @danamlewis (or @scottleibrand ) do you know what the pads on the bottom of the explorer are?
Dana Lewis
@danamlewis
Jun 02 2017 04:11
define which thing and which direction you’re calling bottom?
Cameron Chunn
@applehat
Jun 02 2017 04:13
Large exposed pads on the reverse side of the Edison. Looks like a spot that a surface mount chip would go, but huge.
Scott Leibrand
@scottleibrand
Jun 02 2017 04:14
Those are for attaching a cc-debugger if you need to flash the cc1110 that way vs. using ccprog
Cameron Chunn
@applehat
Jun 02 2017 04:14
Ahh.
I wonder if there would be any issue using something like an i2c block with the explorer block. I'd assume not, but im also not since they are passthru, and even if anything was already on the i2c its an addressable bus.
aeandrews
@aeandrews
Jun 02 2017 04:20
hi all, i'm trying to access my RPI3 rig via my iphone, so far without success. i'm at this part of the docs (https://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-2/accessing-your-rig.html#if-your-iphone-and-rig-are-on-the-same-wifi-network). i downloaded the termius app and have followed the instructions but when i try to connect it it says "private key is not selected" and then "invalid username/password." i just changed the password on my rig (to make sure i was using the correct one lol) but still no luck. thoughts? is there some other password i've forgotten about that i should be using? i do have the rig set up for password-less login on my computer, if that makes a difference (since it's the only other reference i've seen in the docs to a private key)
Cameron Chunn
@applehat
Jun 02 2017 04:21
Are you on the same network? I've had this issue when I forgot to connect to wifi =P
Wait no - that would fail to even connect, not reject the password.
aeandrews
@aeandrews
Jun 02 2017 04:24
According to my router they are on the same network (I checked before posting here because that was my problem last time I couldn't connect to the rig haha)
ccreek
@ccreek
Jun 02 2017 04:25
Brand new build today, on Edison/Explorer, and I think all I'm missing is how to connect to pump. Where can I find this info?
Dana Lewis
@danamlewis
Jun 02 2017 04:26
@ccreek welcome! did you finish the interactive setup script and give it your pump SN?
Aaron Michelson
@Aaybob
Jun 02 2017 04:26
How does one troubleshoot g5 receiver plugged in to v1.2 explorer. I don't know where to start but I saw here using "lsusb". That shows three lines, two devices on Bus 001, one on Bus 002. My rigs using early model explorer boards do not respond to "lsusb".
Dana Lewis
@danamlewis
Jun 02 2017 04:26
(or, what step did you last complete?)
Scott Leibrand
@scottleibrand
Jun 02 2017 04:26
@aeandrews what username are you using in termius?
ccreek
@ccreek
Jun 02 2017 04:27
Successful auto install of openaps 0.1.5
Dana Lewis
@danamlewis
Jun 02 2017 04:28
k, you’ll want to continue on to step 2 then
aeandrews
@aeandrews
Jun 02 2017 04:28
I'm using "root" as the username
ccreek
@ccreek
Jun 02 2017 04:29
I guess I misunderstood the docs - I saw "skip this" and skipped it all.
ccreek
@ccreek
Jun 02 2017 04:30
So I actually need to do Step 1: Pull/clone oref0
?
Dana Lewis
@danamlewis
Jun 02 2017 04:30
yep
ccreek
@ccreek
Jun 02 2017 04:30
Got it. Thanks for clearing that up.
Dana Lewis
@danamlewis
Jun 02 2017 04:31
Editing to say “skip this and go on to Step 1” to hopefully make that more clear
ccreek
@ccreek
Jun 02 2017 04:31
I'm at a bit of a disadvantage, as I'm new, not only to the maker aspect, but to T1D and all its terminology as well.
In fact, This will be the first we've powered on the second-hand Medtronic pump as well. This is very different from the Tandem T:Slim.
Dana Lewis
@danamlewis
Jun 02 2017 04:32
fair warning - we do make up words around here quite a bit, so don’t hesitate to ask if you’re unsure of something. (Also might be worth skimming over the glossary, too: http://openaps.readthedocs.io/en/latest/docs/Resources/glossary.html)
ccreek
@ccreek
Jun 02 2017 04:33
I've printed a ream of your docs, and wife and I are reading all we can.
So glad you've made these resources available.
Dana Lewis
@danamlewis
Jun 02 2017 04:34
:+1: true community effort!
Aaron Michelson
@Aaybob
Jun 02 2017 04:37
root@green:~# lsusb
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 003: ID 22a3:0047
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
root@green:~#
Does this show that Dexcom receiver is connected?
Cameron Chunn
@applehat
Jun 02 2017 04:37
@ccreek yeah, don't hesitate. And I know the feeling - even powering on my second-hand 722 today, I was confused compared to my 754 i've used for a few years. They are "the same" but feel very different.
ccreek
@ccreek
Jun 02 2017 04:38
Do I correctly understand that the system requires WiFi to operate normally, but can be directly plugged into the Dexcom G4 for use offline?
rsilvers129
@rsilvers129
Jun 02 2017 04:39
Oh. I have been skipping ”manual installation Step 1" every time, and have never successded in installing the software. I then try doing everything over and over and eventually get it to work. Also I am not sure this is true "(note this is only necessary for the dev branch, NOT for master)"
Cameron Chunn
@applehat
Jun 02 2017 04:39
@ccreek the system relies on WiFi (or bluetooth tethering), in general, to get blood glucose data from Nightscout.
But it does try to get data from the receiver (or whatever CGM you have setup) first.
ccreek
@ccreek
Jun 02 2017 04:40
Thanks!
Cameron Chunn
@applehat
Jun 02 2017 04:40
From what I have gathered, plugging a receiver up to the explorer is not all that common of a use case
Aaron Michelson
@Aaybob
Jun 02 2017 04:41
@rsilvers129 the second instruction, why would you think its needed for non dev?
Dana Lewis
@danamlewis
Jun 02 2017 04:41
@applehat actually, it is. And I recommend keeping a CGM plugged in at night so you have it in case wifi and tethering go down.
Cameron Chunn
@applehat
Jun 02 2017 04:42
I don't use a CGM receiver :shrugs:
Aaron Michelson
@Aaybob
Jun 02 2017 04:42
I wish I could plug in and work that way and I used to be able to .
rsilvers129
@rsilvers129
Jun 02 2017 04:42
@Aaybob Well, I could not get a new rig working two nights ago. I tried doing things over and over. Finally I posted here that it seems things were missing. I didn’t get a response, but then I did cd ~/src/oref0 && npm run global-install, and then Master worked.
Dana Lewis
@danamlewis
Jun 02 2017 04:42
@applehat or another offline mechanism, if you have it. You don’t have to, but for those who have receivers, I think it’s worth doing.
Cameron Chunn
@applehat
Jun 02 2017 04:43
My "idea I have that Im totally going to eventually do" would handle that pretty well.
Dana Lewis
@danamlewis
Jun 02 2017 04:43
;)
ccreek
@ccreek
Jun 02 2017 04:43
So I'm about to do Step 2, and it says that the system will pull a full history from the pump... I just got this off Craigslist - Do I need a full history? Would it be better if I didn't get a history?
Dana Lewis
@danamlewis
Jun 02 2017 04:43
if the pump has been off for a while, don’t worry about it
ccreek
@ccreek
Jun 02 2017 04:44
cool.
Dana Lewis
@danamlewis
Jun 02 2017 04:44
but this is why you want to go ahead and set up the pump w/ your settings, basal, etc.
(and put temp basal type on absolute, not percent)
so that when it reads, it’ll start reading your info and using that to calculate temps
Cameron Chunn
@applehat
Jun 02 2017 04:44
^ and if your dropping low right as your setting it up, it will save you ;)
ccreek
@ccreek
Jun 02 2017 04:45
Just so you're aware, my wife is the T1D - I'm just trying to put together a more responsive system for her.
I think she's put all that in.
Dana Lewis
@danamlewis
Jun 02 2017 04:45
:+1:
Bob Duerr
@Duer0049
Jun 02 2017 04:45
Good Evening All, I am running into an issue with BT tethering to my Galaxy S8, it is paired and connects, then just disconnects again, it keeps cycling through connecting and disconnecting, anyone run into that before.
Cameron Chunn
@applehat
Jun 02 2017 04:46
@Duer0049 try removing the BT device completely and re-adding it.
Otherwise make sure no other software (xDrip sometimes) is cycling bluetooth on you.
Aaron Michelson
@Aaybob
Jun 02 2017 04:46
@rsilvers129 that is convincing. I'd say if it doesn't hurt anything run it. And yes why would dev and not master need it?
ccreek
@ccreek
Jun 02 2017 04:47
I hate to be a pest, but I want to be certain: Since I haven't been "looping for a while", I believe I wont need to do either of the following?
Cameron Chunn
@applehat
Jun 02 2017 04:47
Also be aware that without root, androids default behavior is to disable bluetooth tethering when wifi is enabled.
ccreek
@ccreek
Jun 02 2017 04:47
mkdir -p ~/src; cd ~/src && git clone -b dev git://github.com/openaps/oref0.git || (cd oref0 && git checkout dev && git pull)
cd ~/src/oref0 && npm run global-install
Dana Lewis
@danamlewis
Jun 02 2017 04:47
@ccreek right - you’ll want master for now
onward to step 2 :)
Scott Leibrand
@scottleibrand
Jun 02 2017 04:48
@Aaybob yes, 22a3:0047 is the Dexcom receiver
Cameron Chunn
@applehat
Jun 02 2017 04:48
@ccreek dont ever feel like a pest. Everyone successfully looping wants nothing more then to see others successful.
Dana Lewis
@danamlewis
Jun 02 2017 04:48
^^^ :+1: :clap:
Aaron Michelson
@Aaybob
Jun 02 2017 04:50
@scottleibrand so cool!
ccreek
@ccreek
Jun 02 2017 04:55
Do I use my herokuapp site for "Nightscout host"?
Dana Lewis
@danamlewis
Jun 02 2017 04:59
Yes
aeandrews
@aeandrews
Jun 02 2017 05:02
another question.. my rig has been working fine for the last few hours but now it's not - when I check the logs it says "AttributeError: 'Pump' object has no attribute 'model'". anyone know what that means?
Aaron Michelson
@Aaybob
Jun 02 2017 05:02
@aeandrews means not talking to pump
aeandrews
@aeandrews
Jun 02 2017 05:04
right, sorry - I meant, does anyone know what would cause that error?
Aaron Michelson
@Aaybob
Jun 02 2017 05:05
explorer board?
aeandrews
@aeandrews
Jun 02 2017 05:05
rpi3
rsilvers129
@rsilvers129
Jun 02 2017 05:05
No matter what I try, I get pump urgent battery low pushover warnings at 1.29v.
Aaron Michelson
@Aaybob
Jun 02 2017 05:06
you know the pump battery only last two days before replacement needed?
rsilvers129
@rsilvers129
Jun 02 2017 05:06
Yes. They die at 1.22 volts. I don’t want a warning until 1.25v
I never want an urgent warning
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:08
question regarding offline looping. If I select g4-upload and I don't have an internet connection, will the loop still work? Or does it need to upload in order to successfully loop?
Aaron Michelson
@Aaybob
Jun 02 2017 05:08
@aeandrews could you rule out battery and distance? Are you using CareLink?
Scott Leibrand
@scottleibrand
Jun 02 2017 05:08
@Sgoldberg89 it'll work fine offline, and upload again when it has Internet.
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:09
thanks
aeandrews
@aeandrews
Jun 02 2017 05:10
@Aaybob yup i'm using carelink, pump <1 ft from the carelink. but now reading @rsilvers129 message about batteries dying at 1.22 v, mine's at 1.21, so that's probably it
Except now without me doing anything (i.e. I didn't change the battery yet), loop seems to be working again, logs are error-free. mysterious
Tom Petrillo
@tompetrillo
Jun 02 2017 05:12
Hi Guys! I was lucky enough to purchase a paradigm 512 off craigslist tonight (with appropriate firmware). All of my reservoirs are for the 7xx series. Has anyone solved this problem with a hacksaw?
Aaron Michelson
@Aaybob
Jun 02 2017 05:15
@aeandrews that error message is from the first handshake between pump and RPi.....carelink is kind of cludgey sometimes....definitly the weakest link.
Dana Lewis
@danamlewis
Jun 02 2017 05:15
not that I've heard of - but you might be able to trade someone for the smaller reservoirs or some other combo of supplies
(Or trade your 512 for a 712)
Aaron Michelson
@Aaybob
Jun 02 2017 05:16
@redshift83 I had the same issue. cutting with knife worked....only needed to once, got some gifted then started buying them
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:16
Is there any way to actually know that I'm looping from offline data as opposed to online data without taking my rig out of wifi range and seeing if I loop?
Sorry I'm being annoying about this, I'm leaving for China on Saturday and would love to loop abroad for the first time. Would be a real game changer
Aaron Michelson
@Aaybob
Jun 02 2017 05:17
@Sgoldberg89 you can serial link and comment out supplicant.conf
@Sgoldberg89 yea, its a game changer!
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:19
There's nothing in the tail to indicate it's pulling cgm data locally?
Aaron Michelson
@Aaybob
Jun 02 2017 05:20
@Sgoldberg89 that's what I'm trying to work on now....
I'm connected to my receiver. I know because the command "lsusb" returns the device number Bus 001 Device 002: ID 22a3:0047
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:22
sweet! Ok, I'll try that when I connect it
Aaron Michelson
@Aaybob
Jun 02 2017 05:22
But the data is not getting through to the rig.
Tom Petrillo
@tompetrillo
Jun 02 2017 05:22
Is there a page suggesting a "ideal hardware setup" as of today? I'm not rich, but now that I got a compatible pump, I'm ready to blow a couple hundred on this...
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:22
I think I ran the setup script too many times so had to do a hard reset
anyone else find that "sudo shutdown" doesn't actually work on their explorer boards?
Aaron Michelson
@Aaybob
Jun 02 2017 05:23
@Sgoldberg89 its "shutdown now -h"
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:23
cool
aeandrews
@aeandrews
Jun 02 2017 05:25
yeah the carelink is a bit of a pain.. I acquired all my loop materials in early fall 2016 when RPis were still the main method, but then didn't actually get around to setting it up until last month, at which point I realized I would have been better off with an edison rig! Hoping to build one soon but wanted to see if I could use what I already had for a while first. still working out the kinks but even so, I'm loving looping so far :) I had a day last week where I was 100 percent in range all day which I don't think has ever happened before... really amazing what a difference it makes!
Tom Petrillo
@tompetrillo
Jun 02 2017 05:25
Dana: Thats bleeding edge?
(smallest)
Dana Lewis
@danamlewis
Jun 02 2017 05:25
Yes
Aaron Michelson
@Aaybob
Jun 02 2017 05:25

I get this with network turned off but connected to receiver:
Temp refreshed
{"carbs":0,"boluses":0,"mealCOB":0,"currentDeviation":3.21,"maxDeviation":4.65,"minDeviationSlope":-0.177}
BG data is too old (it's probably this), or clock set incorrectly. The last BG data was read at Thu Jun 01 2017 21:50:52 GMT-0700 (PDT) but your system time currently is Thu Jun 01 2017 22:18:22 GMT-0700 (PDT)

then it shows old "bg" data

Dana Lewis
@danamlewis
Jun 02 2017 05:25
@aeandrews :smile: :+1:
Tom Petrillo
@tompetrillo
Jun 02 2017 05:26
@danamlewis thanks
Dana Lewis
@danamlewis
Jun 02 2017 05:26
:+1:
aeandrews
@aeandrews
Jun 02 2017 05:27
But speaking of kinks... I was just playing around, trying to set up my dexcom receiver (got G5 about a month ago and have only been using it on my phone) and in the process accidentally stopped my sensor, so there goes my loop for the next two hours, lol. probably it's a sign it's time for bed! thanks so much (as always) for the help!
Aaron Michelson
@Aaybob
Jun 02 2017 05:28
@aeandrews 100% and you have more energy and you sleep better and you worry less. My RPi3 works great its just hard to cary around but it did always link to the receiver...which maybe I should use...
rsilvers129
@rsilvers129
Jun 02 2017 05:33
Switching to master, I am getting this error:
Screen Shot 2017-06-02 at 1.32.52 AM.png
Tim Street
@tim2000s
Jun 02 2017 05:34
@applehat An eink screen with latest status is easy to do. I drive an HTML page off my rigs that can either be accessed over wifi or via the Bluetooth tether ip address giving just that info for when the internet connection isn't good enough/doesn't work.
Aaron Michelson
@Aaybob
Jun 02 2017 05:35
@rsilvers129 pretty sure one may have to run install a few times..
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:35
@Aaybob The dex should be charging, right?
because mine isn't
Aaron Michelson
@Aaybob
Jun 02 2017 05:36
@Sgoldberg89 yup, I have to unplug and replug it to start it charging
Scott Leibrand
@scottleibrand
Jun 02 2017 05:36
yes, if your USB OTG cable is working, and you haven't purposefully disabled charging, you'll have a charging indicator on the Dexcom. try swapping the direction of your cable if re-plugging it doesn't get it working
Tim Street
@tim2000s
Jun 02 2017 05:39
@applehat With a slight change of http server on the rig, you'd be able to display the glucose.json data as a graph as well as the prediction data from enacted.
That's one more thing I need to play with (sigh).
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:39
@scottleibrand DING DING DING swapping direction!
makes sense though... helps when you think about things
root@edison3:~# lsusb
unable to initialize libusb: -99
assuming that's bad
Aaron Michelson
@Aaybob
Jun 02 2017 05:40
@scottleibrand I can't get new data from my receiver. I get :~# lsusb Bus 001 Device 002: ID 22a3:0047
@Sgoldberg89 yea I think so, all my early version explorer boards get that result
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:41
::dramatic:: "NOOOOOOO"
Scott Leibrand
@scottleibrand
Jun 02 2017 05:42
@Aaybob check your cgm-loop.log for starters
Aaron Michelson
@Aaybob
Jun 02 2017 05:42
@Sgoldberg89 you can't be in a hurry on this job
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:42
I literally have a plane to catch
Aaron Michelson
@Aaybob
Jun 02 2017 05:43
@scottleibrand log tailed shows nothing since I disabled network one hour ago
@Sgoldberg89 but you'll be out of network? can't wifi hotspot?
@scottleibrand or just repeats Thu Jun 1 22:45:02 PDT 2017
reporting cgm/cgm-glucose.json
213 2017-06-01T21:50:52-07:00
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:48
how long does it take for the lsusb to show something else if you change the configuration?
Aaron Michelson
@Aaybob
Jun 02 2017 05:49
I just used the command for the first time this hour....
@Sgoldberg89 I also just got ssh on my phone working, there's great directions in docs. The game changer will come. Be patient. We're all in the same boat...not working on rigs full time....
Scott Leibrand
@scottleibrand
Jun 02 2017 05:55
lsusb instantly reflects any usb devices that get plugged in and recognized
@Aaybob sometimes I've noticed the rig stops being able to pull data from the CGM: unplugging and replugging it fixes that...
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:57
is it possible i turned off the USB?
Aaron Michelson
@Aaybob
Jun 02 2017 05:58
@Sgoldberg89 well do you have the explorer made last year or before march?
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:59
I think I'm working on a newer one
but I can't be sure that they didn't give me an old one when I picked them up
WAIT
Scott Leibrand
@scottleibrand
Jun 02 2017 05:59
what version number is printed on the EB?
Sgoldberg89
@Sgoldberg89
Jun 02 2017 05:59
root@edison3:~# lsusb
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 003: ID 22a3:0047
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Different is good!
Aaron Michelson
@Aaybob
Jun 02 2017 06:00
does it say 2016 or 2017 on the explorer board
how did you fix it?
Scott Leibrand
@scottleibrand
Jun 02 2017 06:00
you want the one that says Explorer Block v1.2 GMD - 2017 behind the antenna
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:01
I channeled my inner Scott and used a different USB cable and may have had them plugged in improperly...
Scott Leibrand
@scottleibrand
Jun 02 2017 06:01
s/inner Scott/basic troubleshooting skills/ ;-)
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:01
that's probably really funny
Scott Leibrand
@scottleibrand
Jun 02 2017 06:02
s/something/something else/ is just search and replace syntax
so I was just fixing your statement to say "I channeled my basic troubleshooting skills" ;-)
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:04
Scott, you're in the wrong industry. We need to get you an HBO standup special, or at least Netflix
Wait, that's not actually a bad idea, I bet you could create a whole new field of standup for coders. (or it probably already exists)
Dana Lewis
@danamlewis
Jun 02 2017 06:05
You're not helping his ego ;)
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:06
Sorry, that's why I don't get on gitter very often
Aaron Michelson
@Aaybob
Jun 02 2017 06:06
@Sgoldberg89 are you getting data from receiver?
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:06
I just added rawbg to my nightscout
If I get a Rawbg line (that's a thing, right?) then I'm assuming it's working
because I can't get raw data anywhere else
Aaron Michelson
@Aaybob
Jun 02 2017 06:07
I think I need a new receiver
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:07
is yours not working?
something real weird just happened to my nightscout
where my BG usually is, it says ?NCX
Aaron Michelson
@Aaybob
Jun 02 2017 06:08
I've been trying for three months to get it to work when I go to work where there is no internet or cell phone
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:09
dang, that's so frustrating, I'm sorry. If this actually works, I'll tell you anything you want to know about my setup!
But so far I'm not sure it's working
It may just have broken my nightscout
Aaron Michelson
@Aaybob
Jun 02 2017 06:10
but i'll thry the rpi3 rig to see if the receiver just isn't outputting data, maybe I could see with the dexcom software too.
Scott Leibrand
@scottleibrand
Jun 02 2017 06:10
@danamlewis my ego thinks @Sgoldberg89 was making fun of my sense of humor, so he isn't hurting it. :)
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:10
wow, I'm hurt
Aaron Michelson
@Aaybob
Jun 02 2017 06:10
I think we're all ready for bed....gdnt
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:11
I find you both extremely entertaining while simultaneously tremendously inspiring people.
Night!
Real quick before everyone signs off... is there a way to see raw bg on nightscout?
Scott Leibrand
@scottleibrand
Jun 02 2017 06:12
@Sgoldberg89 ok, fair. hard to tell genuine compliments from dry sardonic wit sometimes. :)
yes, if you are uploading raw bg, you can add the appropriate ENABLE settings and then turn it on in settings to see the white dots in NS
Aaron Michelson
@Aaybob
Jun 02 2017 06:13
@Sgoldberg89 not with g5
Scott Leibrand
@scottleibrand
Jun 02 2017 06:13
correct, you have to be on G4
Aaron Michelson
@Aaybob
Jun 02 2017 06:13
root
wrong screen
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:14
ok, I put rawbg in the ENABLE settings, but I don't have any additional options on my nightscout page
omg just kidding, it's right there
go to sleep
<3 you guys
... too soon?
Aaron Michelson
@Aaybob
Jun 02 2017 06:16
@Sgoldberg89 good luck with business travel
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:16
thank you, good luck to you as well
Aaron Michelson
@Aaybob
Jun 02 2017 06:17
yeah getting up at 4:15 going to bet at 11:15
Dana Lewis
@danamlewis
Jun 02 2017 06:17
@Sgoldberg89 not too soon :smile: have an awesome trip!
Aaron Michelson
@Aaybob
Jun 02 2017 06:17
but loop running
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:18
WOAH
I'll be on here tomorrow to figure out how to get this working
Aaron Michelson
@Aaybob
Jun 02 2017 06:18
not offline
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:18
nice try
Tim Street
@tim2000s
Jun 02 2017 06:23
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:25
@scottleibrand I got the following message in my tail:
enact/smb-suggested.json: {"duration":30,"reason":"CGM is calibrating or in ??? state; setting current basal of 1.7 as temp. ","rate":1.7,"deliverAt":"2017-06-02T06:23:33.069Z","temp":"absolute"}
If the reciever that is plugged into the edison is calibrating but the other one isn't does that mean it's pulling data from the one that's plugged in?!
Scott Leibrand
@scottleibrand
Jun 02 2017 06:27
Guess so
Sgoldberg89
@Sgoldberg89
Jun 02 2017 06:30
haha
ok, gnight!
PieterGit
@PieterGit
Jun 02 2017 08:39
starting with merging stuff. @scottleibrand asked me to check with @Kdisimone and @danamlewis for a new text of choosing the CGM. Current question is more than 80 character long and not very readable. I improved it:
image.png
@Kdisimone / @danamlewis (or others). Can you improve the text for xdrip/ G5? . The new proposed statement is:
    echo -e "What kind of CGM are you using?\nValid options:\nG4-upload: use Dexcom G4 receiver connected to USB and upload to Nightscout\nG4-local-only: will NOT upload BGs from a plugged in receiver to Nightscout\nG5: Dexcom G5\nMDT: Medtronic CGM\nshareble: Bluetooth Dexcom G4 or G5 receiver\nxdrip: Requires xDrip Wireless Bridge (No Receiver)."
    read -p "Please selection option for your CGM: " -r
we can also add colors to highlight the CGM options better
Savek-CC
@savek-cc
Jun 02 2017 08:43
is nightsout able to handle incoming data from more than one device (all originating by the same source/dex transmitter though)?
i.e. is there some sort of de-duplication in place?
Lorenzo Sandini
@lsandini
Jun 02 2017 08:49
Forgive my ignorance with linux, but the SSH console shows some colors in dev and b/w with master. If it is a simple thing, could this be implemented also in master ?
PieterGit
@PieterGit
Jun 02 2017 08:54
@savek-cc : see https://github.com/nightscout/cgm-remote-monitor/blob/d407bab2096d739708f365eae3c78847291bc997/lib/api/entries/index.js#L611 for the code of uploading the data to nightscout. I don't know the exact workings (never tried with more than one receiver). What do you want to accomplish: e.g. use both Xdrip and Dexcom receiver? Or something else?
@lsandini Once the 0.5.0-dev will be released the colors (and much new other features) will be available on master
See openaps/oref0#424 for progress on the 0.5.x release
Savek-CC
@savek-cc
Jun 02 2017 08:57
@PieterGit see my comment on nightscout/cgm-remote-monitor#721
Basically planing to use a network of multiple wixels around house and work to catch my G4 signal. Each node would be capable of uploading the current reading to NS - but the same reading might be captured and thus uploaded by multiple nodes.
Having a reading de-duplication would be essential for this scenario.
PieterGit
@PieterGit
Jun 02 2017 09:00
@savek-cc : that's sounds like your redoing https://github.com/jamorham/python-usb-wixel-xdrip
i would not advise to insert stuff directly into mongo, but always use the nightscout api
Savek-CC
@savek-cc
Jun 02 2017 09:03
does the rest-api feature reading de-duplication?
not re-doing the usb-wixel, since I want the nodes to be a client to NS - instead of having to be pulled by an xDrip+ instance on the same network
I want the nodes to work and transmit data to nightscout without the involvement of a mobile phone with xDrip or the need to assign static IPs on a wifi I don't control.
(e.g. workplace)
PieterGit
@PieterGit
Jun 02 2017 09:06
I think Nightscout has some deduplication but not sure. If you don't want xdrip android app you need to implement the logic of the android apps that convert raw sensor readings to glucose data (calibration stuff, etc). I don't think that stuff has been done (yet).
Savek-CC
@savek-cc
Jun 02 2017 09:07
Hmm... that's a good point though ;)
but I think, only slope and intercept are required for a start - so that could be provided by xDrip (data is already visible in NS)
PieterGit
@PieterGit
Jun 02 2017 09:12
i think it would be good to be able to use https://github.com/NightscoutFoundation/xDrip as standalone java (and optionally use calibrations records from dexcom receiver). I think both are quite a big programming job. I also wanted it, but having an android phone do the logic, and using python-usb-wixel-xdrip seemed much easier
i don't use wixel/xdrip at the moment, and just use a dexcom receiver connected to usb.
i'm using xdrip as a backupdevice in case the dexcom receiver breaks
PieterGit
@PieterGit
Jun 02 2017 09:19
@savek-cc : i also think there is a flaw in your plan. Networks you don't control like workplace/hotel wifi often only support HTTP(S) and I doubt wether a mongodb connection will work from those networks. In my workplace the only way to get out of the network is via a reverse http proxy.
Savek-CC
@savek-cc
Jun 02 2017 09:20
Workplace: often have a guest network with limit to HTTP/S but no proxy auth
not talking about direct mongo upload - want to use NS REST API
PieterGit
@PieterGit
Jun 02 2017 09:21
ok, thought you wanted a direct mongodb connection like some of the uploader apps use.
PieterGit
@PieterGit
Jun 02 2017 11:22

i think openaps/oref0#470 is ready for broader testing... it installs fine (tested with manual entry of all the data and with single command line). The pump is out of the house right now, so I can't test any further now. @scottleibrand can you do a (final) review. Others willing to give it a testdrive? New features: Nightscout token based authentication.

Other improvements:

  • fix --ww_ti_usb_reset, openaps/oref0#456
  • remove --config and --noop from ns-get.sh and ns-upload.sh
  • remove oref0-find-ti.sh (not used anymore, implemented in oref0_init_pump_comms.py and oref0_subg_ww_radio_parameters.py
  • remove pump-loop loop from oref0-pump-loop.sh. Cron is responsible of calling each iteration of the loop
  • log Aborted and exit code if pump loop fails
  • less mmtune in maybe_mmtune
  • convert ns-temptargets and ns-meal-carbs to use NS device instead of calling curl
Scott Leibrand
@scottleibrand
Jun 02 2017 11:44
@PieterGit I think the "new text of choosing the CGM" has since been overtaken by events: @danamlewis recently merged some changes into dev that completely re-do the formatting of the Q&A part of oref0-setup.
PieterGit
@PieterGit
Jun 02 2017 11:45
I noticed (that's why merging my stuff was quite difficult :smile: ) . i think the formatting for a default 80 characters display can be improved quite a lot. and the various cgm methods are not very well explained in the dialogue
Scott Leibrand
@scottleibrand
Jun 02 2017 11:53
I don't think anyone is limited to 80 characters
maybe if we start having people run this over a console connection we'd want to do line wrapping, though
PieterGit
@PieterGit
Jun 02 2017 11:57
e.g. default putty consoles are still 80 characters wide
Unfortuately I need to go know. I'll be back tonight or this weekend. have a good day
jtsackdex
@jtsackdex
Jun 02 2017 12:00
Question about battery life. Has anyone used 1100mAh rechargeable batteries? How successful have you been? Found these at Amazon. https://www.amazon.com/dp/B00HIW3Q0Y?th=1
cameronrenwick
@cameronrenwick
Jun 02 2017 12:01
@jtsackdex I wouldn't recommend them as they're pretty tiny. I use a 2500 and it barely makes a day for me. Granted I have to wifi tether not BT but that 1100 IMO is small
Scott Leibrand
@scottleibrand
Jun 02 2017 12:02
@PieterGit I just replied to your "pump suspended" question. I'll need to review the rest of it when I have time to really sit down and dig in.
jtsackdex
@jtsackdex
Jun 02 2017 12:03
@cameronrenwick ok. could you share which ones. batteries could expensive fast.
rsilvers129
@rsilvers129
Jun 02 2017 12:21
@jtsackdex @cameronrenwick I think he means for the pump. You would have to test them. But as far as I know, no one has found rechargable AAA that make sense to use in the pump.
skubesch
@skubesch
Jun 02 2017 12:21
Question about Medtronic battery life with OpenAPS. I've had two batteries die in less than a week (pump showing LOW BATTERY, which apparently cannot be ignored) but no alert through Nightscout. Pump status was showing 1.14v during one of them (I didn't check on the other), which seems strange. I'm using the recommended battery by Medtronic. Anything else I can try or is this expected?
jtsackdex
@jtsackdex
Jun 02 2017 12:24
@rsilvers129 Yes I meant the pump, should have clarified. Ok Thanks.
rsilvers129
@rsilvers129
Jun 02 2017 12:25
@skubesch AAA Alkaline battery life is 2.2 days for me with Duracell. Someone got as much as 2.85 days with Alkaline. If you want more life, there are Energizer Ultimate Lithium. They last 8.5 days. I would replace Alkaline every 48 hours.
skubesch
@skubesch
Jun 02 2017 12:37
Ahhh okay, thank you @rsilvers129
PopcornPam
@PopcornPam_twitter
Jun 02 2017 12:42
Hi all, I keep having problems running autotune.
I run the command cd ~/myopenaps/settings/ && cp profile.json autotune.json && cp profile.json pumpprofile.json twice
Then I run oref0-autotune --dir=~/myopenaps --ns-host= http://pippidog.herokuapp.com--start-date=2017-05-31
and get oref0-autotune <--dir=myopenaps_directory> --ns-host=https://pippidog.herokuapp.com [--start-date=YYYY-MM-DD] [--end-date=YYYY-MM-DD] [--runs=number_of_runs] [--xlsx=autotune.xlsx] [--log=(true)|false]
This seems to happen each time. Eventually I find some small thing that wasn't clear in the docs to do, and forget to pr it
Help.. what am I missing?
The error message keeps listing azure, when I use heroku, so don't understand that
rsilvers129
@rsilvers129
Jun 02 2017 12:48
@PopcornPam_twitter You need a space after your URL
PopcornPam
@PopcornPam_twitter
Jun 02 2017 12:49
i try with and without the space, same error
It should be from cd myopenaps... right?
root@myopenaps2:~/myopenaps# oref0-autotune --dir=~/myopenaps --ns-host= http://pippidog.herokuapp.com --start-date=2017-05-31
Option http://pippidog.herokuapp.com unknown
Usage: oref0-autotune <--dir=myopenaps_directory> --ns-host=https://mynightscout.azurewebsites.net [--start-date=YYYY-MM-DD] [--end-date=YYYY-MM-DD] [--runs=number_of_runs] [--xlsx=autotune.xlsx] [--log=(true)|false]
It keeps telling me http://pippidog.herokuapp.com unknown
oh.. it needed an s
now it keeps saying any date i give it is unknown
have tried 2017-05-31, 2017-06-01, no luck
rsilvers129
@rsilvers129
Jun 02 2017 12:58
Try without the —start-date. It will run the last day without any date option.
Try https instead of http
PopcornPam
@PopcornPam_twitter
Jun 02 2017 13:00
nope
no matter what I do it keeps saying my nightscout is unknown
PopcornPam
@PopcornPam_twitter
Jun 02 2017 13:07
-bash: /root/myopenaps/autotune#: No such file or directory
rsilvers129
@rsilvers129
Jun 02 2017 13:07
This:
oref0-autotune --dir=~/myopenaps --ns-host=https://pippidog.herokuapp.com
PopcornPam
@PopcornPam_twitter
Jun 02 2017 13:07
Thank you Robert!
But that looks identical to the command I had been using, so I don't understand where my error was
Scott Leibrand
@scottleibrand
Jun 02 2017 13:10
you had a space after --ns-host=
PopcornPam
@PopcornPam_twitter
Jun 02 2017 13:11
ah.. ok. Thank you for the clarification
jtsackdex
@jtsackdex
Jun 02 2017 13:11
logout
PopcornPam
@PopcornPam_twitter
Jun 02 2017 13:12
I also have a question about offline looping. I am wondering if I am looping with a g5 reciever plugged into the otg port, is there a way to give a direct command via laptop to start a temp target? I know I would have to unplug the dexcom while doing that
I will be in a dead zone conference all day Saturday
Scott Leibrand
@scottleibrand
Jun 02 2017 13:14
it's probably easier to actually change your scheduled targets on your pump
PopcornPam
@PopcornPam_twitter
Jun 02 2017 13:15
Through bolus wizard?
rsilvers129
@rsilvers129
Jun 02 2017 13:16
@PopcornPam_twitter Yes. Bolus Setup.
PopcornPam
@PopcornPam_twitter
Jun 02 2017 13:17
ok
And just to check that I am interpretting this correctly...

Autotune pump profile recommendations:

Recommendations Log File: /root/myopenaps/autotune/autotune_recommendations.log

/usr/local/bin/oref0-autotune-recommends-report: line 72: printf: null: invalid number

Parameter | Current | Autotune

ISF | 60.000 | 60.000
CSF | n/a | 0.000
Carb Ratio | 13.000 | 13.000
00:00 | 0.500 | 0.500
06:00 | 0.600 | 0.600
22:00 | 0.500 | 0.500

It suggests that things are ok as is, and don't change them?
I have been running around 200-250 the past few days from about 8-3, and openaps is giving me really high temp basals and lots of SMB, but not coming down til later in the day. I don't think it is a site issue (changed it a few times), and can't seems to figure out what is going on. Was hoping autotune might have some thoughts
David
@davidkeddydb
Jun 02 2017 13:20
@scottleibrand Has this set neutral temp when snoozeBG > target_bg, and lastCOBpredBG > targe… been merged to dev or still in testing? Need guinea pigs?
Scott Leibrand
@scottleibrand
Jun 02 2017 13:24
Pam: you have the CSF 0 bug. If you're on oref0 dev, update to the latest and run again?
@davidkeddydb no, i just put it on our rigs and haven't run it through a meal yet. Will need testers once I confirm that looks good.
Or if you want to be the very first to run completely untested code. ;-)
David
@davidkeddydb
Jun 02 2017 13:30
@scottleibrand As I just made my ratios more aggressive given the zero temping, I will put the one for lunch back to where it was and give it a go. What is does is what I wanted to try, I just had not figured yet how to write it :) Will circle back later once I give it a run. Comment over there vs in here?
fidoman
@efidoman
Jun 02 2017 13:35
@rsilvers129 - what procedure did you use to add loggly? I'm using heroku and removed papertrail because it always filled up too fast.
jeremybarnum
@jeremybarnum
Jun 02 2017 13:41
I'm trying to understand how oref0 decays carbs without AMA enabled, so that I can then understand exactly how AMA works. I'm having a lot of trouble figuring this out by reading the code - anyone have any suggestions about good search terms? what are the key variables that relate to carb decay? More specifically, I can see how it's using deviations to determine carbs absorbed, but I'm having trouble seeing how the history of the deviation is being projected to create a prediction, and I can't seem to find at all how this is done in the absence of AMA - i.e., I don't see a basic carb decay function anywhere.
David
@davidkeddydb
Jun 02 2017 13:43
@scottleibrand Once the updated code is put into determine-basal.js, it is just committed correct? I don't have to reboot or anything or do I?
fidoman
@efidoman
Jun 02 2017 13:46
@davidkeddydb @scottleibrand - I'm willing to test this too when it is ready. I'm still getting too high ostprandial glucose using SMB. ( set neutral temp when snoozeBG > target_bg, and lastCOBpredBG > target )
PopcornPam
@PopcornPam_twitter
Jun 02 2017 13:53
@scottleibrand , coudl that explain the unexplained highs? or just an autotune glitch?
David
@davidkeddydb
Jun 02 2017 13:53
Screenshot 2017-06-02 09.52.31.png
It is slick when you can bolus and not enter the carbs and it goes like this:
bryce
@bryce2071
Jun 02 2017 13:53
openaps isnt working anymore - no basal-changes are done anymore ... 1 possible thing could be the text which nightscout shows if hovering OpenAPS - Text .. in this text, a "no temp" is in. I thik i havent seen that till now.
i manually started a temp basalrate directly on my pump .. was this wrong?
rsilvers129
@rsilvers129
Jun 02 2017 13:55
I have not been using it since now I just use the iPhone Termius app and log into the rig. I made a shorter alias called pll to see the pump-looplog. I also created a shortcut to log in using my rig’s IP WiFi and BT address.
fidoman
@efidoman
Jun 02 2017 14:00
@rsilvers129 :+1: Thanks!
francesc0-cgm
@francesc0-cgm
Jun 02 2017 14:00
@scottleibrand check your messenger...0.05 bolus is working. However smb with kids it is not a great idea. I have UAM enabled too. 0 temp immediately after meal leads to high...i had to switch back to loop to get it back on range
katie disimone
@Kdisimone
Jun 02 2017 14:00
@rsilvers129 the l letter will pull up pump logs. It is a shortcut already coded in.
francesc0-cgm
@francesc0-cgm
Jun 02 2017 14:01
Without smb high temping seems to counteract raise more efficiently
Tim Street
@tim2000s
Jun 02 2017 14:03
Your diabetes may vary. I find SMB with UAM way more effective than high TBR because the insulin is delivered earlier.
rsilvers129
@rsilvers129
Jun 02 2017 14:08
@Kdisimone I have alias l='ls $LS_OPTIONS -lA'
@tim2000s If a high temp is cranking out, for example, 4 units per hour, then that is 0.4 units in 6 minutes. My SMB from yesterday was 0.4 being issued every 5 minutes. That is not significantly faster.
katie disimone
@Kdisimone
Jun 02 2017 14:12
@rsilvers129 but I've had smb and high temps running at same time on large climbs. That is significantly faster.
rsilvers129
@rsilvers129
Jun 02 2017 14:13
IMG_6430.PNG
@Kdisimone This wasn’t fast enough to keep me below 200. I did a 10 unit bolus up front.
Tim Street
@tim2000s
Jun 02 2017 14:14
And that 0.4 is 0.4 at t0 versus 0.4/6, so 0.07 at t0.
francesc0-cgm
@francesc0-cgm
Jun 02 2017 14:14
@tim2000s on my son smb blocks bolus snooze so i have 0 temping. Then editing the code i have the chance to have smb of 0.05 because 0.1 with his sensitivity are too much. E.g. last lunch i had a peak with 0 temping and no smb...i had to switch to loop to get him on range
Uam and ama thought his future bg was 40
While he was raising till 190...
And i made 2.2 units for 45 grams with a ratio of 1:19
As i make every day
Tim Street
@tim2000s
Jun 02 2017 14:17
@rsilvers129 You also have a massive gap in data where it would probably have continued to SMB as absorption was likely to be greater than expected.
francesc0-cgm
@francesc0-cgm
Jun 02 2017 14:17
Without smb bolussnooze should have keep basals on and then when rising it would have high temped to keep him on range
bryce
@bryce2071
Jun 02 2017 14:17
to my problem 1 more question: is there someone wo is willing to look into the hover text of the word "OpenAPS x" in the nightscout site. The problem is, that there is a text which i havent seen till now, its "no temp" .. the text is at: "BG: 224, COB: 0, ... Target: 94: Eventual BG 189 >= 94, no temp, setting 5,65U/hr IOB: 1,52U"
francesc0-cgm
@francesc0-cgm
Jun 02 2017 14:17
In my experience smb is effective in late peaks and in jaundice
katie disimone
@Kdisimone
Jun 02 2017 14:18
@Duer0049 did you see in the docs there's an android app that helps the BT tether stay connected? Maybe try that app?
francesc0-cgm
@francesc0-cgm
Jun 02 2017 14:18
When he eats i have often a disaster also editing the code to let him smbing 0.05 per time
Tim Street
@tim2000s
Jun 02 2017 14:18
@francesc0-cgm @scottleibrand has said to you before, if oref0 works better then the use that rather than oref1. Don't use it for the sake of it.
rsilvers129
@rsilvers129
Jun 02 2017 14:19
@tim2000s That could be a factor, but I ate at 7:30 pm, and the first gap was over by 9pm. So my 10 units of initial bolus didn’t even reach peak yet. The second gap - yes, it seems like it would have given me more and brought me down faster. But I was already well over 200.
francesc0-cgm
@francesc0-cgm
Jun 02 2017 14:19
I'm just testing for people i don't want to criticize anyone or anything
I'm using my son to test smb for kids
Oref1 is for sure better looking and more stable
Imho for kids 0.05 smb could avoid lows due to sensitivity as it happened using normal smbing but with low basal ratios could lead to too much 0 temping
So my early impression are it is perfect for snacks and jaundice (night?) While it can gives some problems on large meal
Tim Street
@tim2000s
Jun 02 2017 14:24
@rsilvers129 no, it didn't, but that hours gap would likely have included SMB behaviour.
0.05 works only if the pump supports it, and it seems that the most common pumps are only 0.1 capable sadly.
rsilvers129
@rsilvers129
Jun 02 2017 14:25
With oref0, I can make it more aggressive by turing up the multiplier limits. With SMB, is there a way to make it more aggressive?
@tim2000s @francesc0-cgm My pump does 0.025 minimum bolus.
Tim Street
@tim2000s
Jun 02 2017 14:26
What are you using ?
rsilvers129
@rsilvers129
Jun 02 2017 14:26
MMT-723
Tim Street
@tim2000s
Jun 02 2017 14:26
🦄
francesc0-cgm
@francesc0-cgm
Jun 02 2017 14:27
I have the same and i can microbolus 0.05
Not 0.025
I have 2 in reality
Tim Street
@tim2000s
Jun 02 2017 14:27
And yes you can but you have to go into the code and modify the max bolus.
rsilvers129
@rsilvers129
Jun 02 2017 14:27
@francesc0-cgm Go into setup and change scroll rate to 0.025
francesc0-cgm
@francesc0-cgm
Jun 02 2017 14:27
However 0.05 is already ok for high sensitivity
rsilvers129
@rsilvers129
Jun 02 2017 14:27
I use it to let Bolus Wizard log carbs without doing a significant bolus.
francesc0-cgm
@francesc0-cgm
Jun 02 2017 14:27
My son has 210 on morning
Less than 0.05 is unuseful and if you want something aggressive is not for you :-)
rsilvers129
@rsilvers129
Jun 02 2017 14:38
If I understand this correctly, AMA and SMB can do the same things, but AMA has a way to change how aggressive it is, has a bolus snooze, and is much more resistant to CGM outages.
rsilvers129
@rsilvers129
Jun 02 2017 15:08
AMA doesn’t run into the Medtronic pump limitation where bolus-wizard can’t be open at the time of a dose.
Tim Street
@tim2000s
Jun 02 2017 15:35
Both AMA and SMB rely on the multipliers to determine how aggressive the insulin requirement is. There is then an additional layer of protection in SMB to restrict max bolus.
Diadon81
@Diadon81
Jun 02 2017 15:43

Dears I have a question

COB: 0, Dev: 281, BGI: -27.94, ISF: 2.6, Target: 6.6, minPredBG -7.0, IOBpredBG 2.2

why this target everytime changes. (Target: 6.6)
in my pump I have a range 5.5 -6.0 as a target value, how this correlated with Target above.

Kate Farnsworth
@ELUTE
Jun 02 2017 15:51
@Diadon81 sounds like you have "autosens_adjust_targets": true enabled in preferences
Diadon81
@Diadon81
Jun 02 2017 15:57
yes you are right, but what is the reason for this function, explanation in documentation is very short
autosens_adjust_targets:
This is used to allow autosens to adjust BG targets, in addition to ISF and basals.
what is the practical meaning of this function?
David
@davidkeddydb
Jun 02 2017 16:00
Correct me if I have this wrong but isn't that the whole basis of how autosens works? It has to adjust the the target bg up or down based on its calculated ISF/basal to get you to your true target
Diadon81
@Diadon81
Jun 02 2017 16:02
hm... that explanation doesn't look solid, ISF changes not so often during the day to play with target value.
David
@davidkeddydb
Jun 02 2017 16:05
Here is the code:
// adjust min, max, and target BG for sensitivity, such that 50% increase in ISF raises target from 100 to 120
if (typeof autosens_data !== 'undefined' && profile.autosens_adjust_targets) {
  if (profile.temptargetSet) {
    process.stderr.write("Temp Target set, not adjusting with autosens; ");
  } else {
    // with a target of 100, default 0.7-1.2 autosens min/max range would allow a 93-117 target range
    min_bg = round((min_bg - 60) / autosens_data.ratio) + 60;
    max_bg = round((max_bg - 60) / autosens_data.ratio) + 60;
    new_target_bg = round((target_bg - 60) / autosens_data.ratio) + 60;
    if (target_bg == new_target_bg) {
        process.stderr.write("target_bg unchanged: "+new_target_bg+"; ");
    } else {
        process.stderr.write("target_bg from "+target_bg+" to "+new_target_bg+"; ");
    }
    target_bg = new_target_bg;
Diadon81
@Diadon81
Jun 02 2017 16:10
but if we look at example, why my target bg should raise if ISF raises, at final I want to get my 5.5 bg no matter of ISF.
Diadon81
@Diadon81
Jun 02 2017 16:16
another problem that I see with dev branch, i am always see "COB":0, for example 1 hour ago I ate approx 60 carbs and give a bolus by bolus wizard on my pump, but in the pump-log I don't see any carb change. It always 0.
Cameron Chunn
@applehat
Jun 02 2017 16:19
Is there a big list of "things I can put into NS to enable veiws" somewhere? I see people with much more feature rich NS displays then I have, but haven't found anything in the docs.
garykidd
@garykidd
Jun 02 2017 16:32
Should snoozeBG never be higher than EventualBG?
@danamlewis oops---sorry bout that---will target master accordingly....thx
Tim Street
@tim2000s
Jun 02 2017 16:35
@Diadon81 Autosens is working out You sensitivity. When you are more sensitive than your pump setting, you need less insulin to reduce You levels by the same amount. When this happens, OpenAPS adjusts ISF and raises target slightly and likewise when you are less sensitive, ISF value is lower tha your pump and target is set lower. In both cases the case Brent iob is to get you back to your target and changing target allows more or less insulin to be delivered to do that.
@applehat it's all listed in the Nightscout dev docs.
Diadon81
@Diadon81
Jun 02 2017 16:39
another question about COB, I am entering my carbs by bolus wizard, but in my pump log there is always "COB":0. Also I see UAM and IOB prediction curve in pump log, but never see COB prediction curve.
is it ok, or something wrong there.
My sugar growing high and fast.
enact/smb-suggested.json: {"insulinReq":null,"bg":324,"reservoir":"224.6","temp":"absolute","snoozeBG":369,"rate":null,"predBGs":{"UAM":[324,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null],"IOB":[324,324,324,324,323,323,322,322,321,320,319,317,315,315,316,319,321,323,325,326,328,329,330,331,332,332,333]},"minPredBG":333,"IOB":2.309,"reason":"COB: 0, Dev: 151, BGI: -23.22, ISF: 2.5, Target: 4.4, minPredBG 18.5, IOBpredBG 18.5; Eventual BG NaN >= 4.4,  insulinReq NaN. temp 0<NaNU/hr. ","COB":0,"eventualBG":null,"duration":30,"tick":"+2","deliverAt":"2017-06-02T16:38:54.424Z"}
I ate approx. 1 hour ago, 60 carbs.
it is not enough time to absorb all carbs
why I see COB:0
Matthias
@ceben80
Jun 02 2017 16:58
wtf, really BG 324 @Diadon81 I'd recommend correct per Pen sc not per pump :wink:
Sgoldberg89
@Sgoldberg89
Jun 02 2017 17:12
@Aaybob Hey! I got it looping offline last night. I had an issue with 2 dex receivers deviating significantly from each other (one read 55 and one read 70, at one point overnight) and it stopped looping, but it was 2 am so I wasn't following the log to understand why. At any rate, I'm happy to help you get there if I can!
francesc0-cgm
@francesc0-cgm
Jun 02 2017 17:13
@tim2000s i didn't understand what you said about multipliers on ama and smb to make it more aggressive
Dana Lewis
@danamlewis
Jun 02 2017 17:25
@popcornPam_twitter did you see the response about re-pulling dev to get some recent fixes, then rerunning Autotune?
@jeremybarnum dynamic carb absorption happens if AMA is enabled
jeremybarnum
@jeremybarnum
Jun 02 2017 17:37
Thanks @danamlewis - I was actually trying to see where in the code the decay of carbs is implemented, with and without AMA, just to understand the choices being made about decay rates, projections of the recent past into the future, etc, but I couldn't find it - even for the basic case without AMA enabled.
bryce
@bryce2071
Jun 02 2017 17:47
anyone could help me out with my problem, openaps doent bole anymore?
Dana Lewis
@danamlewis
Jun 02 2017 17:49
@bryce2071 what is the problem you're seeing? What does "bole" mean?
Sgoldberg89
@Sgoldberg89
Jun 02 2017 17:57
anyone have a good contact for a service to replace g4 batteries?
Diadon81
@Diadon81
Jun 02 2017 18:03
@ceben80 Yes, because I cover only half of carbs, to check UAM, but don't see any microbolusing during my bg rise.
also in my pump log i found following error message, don't know is it tied to this issue or not.
and that pumphistory is less than 1m old.  Temp refreshed
{"carbs":0,"mealCOB":0,"reason":"carb_ratio 1.480932987782303 out of bounds"}
{"iob":-0.538,"activity":-0.0111,"bolussnooze":0,"basaliob":-0.538,"netbasalinsulin":-1.6,"hightempinsulin":0.3,"microBolusInsulin":0,"microBolusIOB":0,"time":"2017-06-02T14:44:40.000Z","lastBolusTime":0}
{"delta":-31.5,"glucose":112,"short_avgdelta":-15.33,"long_avgdelta":0.32}
{"duration":30,"rate":0,"temp":"absolute"}
Autosens adjusting basal from 1.378 to 0.75; target_bg from 94.5 to 122; sens from 27 to 48.2
Carb Impact: -34.2 mg/dL per 5m; CI Duration: 0 hours; remaining 4h+ CI: 0 mg/dL per 5m
Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: 0 hours
UAM Impact: -18 mg/dL per 5m; UAM Duration: 0 hours
minPredBG: -47 minIOBPredBG: 39 avgPredBG: -47 COB: 0 carbs: 0
bgUndershoot: -5 zeroTempEffect: 33 carbsReq: -1

  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 "/root/src/openaps/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 "/root/src/openaps/openaps/cli/__init__.py", line 75, in epilog
    self.create_git_commit( )
  File "/root/src/openaps/openaps/cli/__init__.py", line 98, in create_git_commit
    self.repo.index.commit(msg)
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.3-py2.7.egg/git/index/base.py", line 951, in commit
    tree = self.write_tree()
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.3-py2.7.egg/git/index/base.py", line 545, in write_tree
    entries = self._entries_sorted()
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.3-py2.7.egg/git/index/base.py", line 174, in _entries_sorted
    return sorted(self.entries.values(), key=lambda e: (e.path, e.stage))
  File "/usr/local/lib/python2.7/dist-packages/gitdb2-2.0.2-py2.7.egg/gitdb/util.py", line 256, in __getattr__
    self._set_cache_(attr)
  File "/usr/local/lib/python2.7/dist-packages/GitPython-2.1.3-py2.7.egg/git/index/base.py", line 120, in _set_cache_
    fd = lfd.open(write=False, stream=False)
  File "/usr/local/lib/python2.7/dist-packages/gitdb2-2.0.2-py2.7.egg/gitdb/util.py", line 333, in open
    raise IOError("Lock at %r could not be obtained" % self._lockfilepath())
IOError: Lock at '/root/myopenaps/.git/index.lock' could not be obtained
Error, retrying
Sleeping 5;
katie disimone
@Kdisimone
Jun 02 2017 18:10
@applehat the OpenAPS docs just get the basics in there. If you want to read about all the bells and whistles, this is a good place to go http://www.nightscout.info/wiki/welcome/website-features/0-9-features
@davidkeddydb I use autosens with the auto targets turned off. Still works great.
katie disimone
@Kdisimone
Jun 02 2017 18:16
@Diadon81 Your loop is having errors and not looping. A lot of the messages you had in your posts indicate that (NaN= not a number, null= missing data…all of those messages show that your loop is not calculating things properly). Did you try rebooting or rerunning your loop installation?
Diadon81
@Diadon81
Jun 02 2017 18:17
ok. i will try
Sgoldberg89
@Sgoldberg89
Jun 02 2017 18:18
@Kdisimone Many thanks
fidoman
@efidoman
Jun 02 2017 18:35
@Diadon81 - this git error happened to me yesterday and I had to remove the file and reboot to clear it up. rm /root/myopenaps/.git/index.lock. Not saying it is the same issue or if it will work or not for you though.
rsilvers129
@rsilvers129
Jun 02 2017 18:45
@tim2000s does SMB use the same multiplier variables as AMA, or a different one?
Diadon81
@Diadon81
Jun 02 2017 18:51
ok i decide to do a fresh install
so I fwill return to all of you in a hour :)
Cameron Chunn
@applehat
Jun 02 2017 19:04
Am I correct in understanding that UAM only help decide how many additional carbs may be onboard with a miss count (or no notification at all) and works independently from SMBs?
Matthias Granberry
@mgranberry
Jun 02 2017 19:15
@rsilvers129 It is possible to make SMB more aggressive by changing the calculations for maxBolus and microBolus in lib/determine-basal/determine-basal.js. Just be cautious because it can and will overshoot and the CGM delay matters here. Make any changes slowly and with higher-than-normal targets to avoid some potentially serious lows.
Dana Lewis
@danamlewis
Jun 02 2017 19:16
I highly recommend not changing the code without first testing some of these toggles: http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-3/beyond-low-glucose-suspend.html#understanding-your-preferences-json
remainingCarbsCap, remainingCarbsFraction, etc. first
Matthias Granberry
@mgranberry
Jun 02 2017 19:19
And make sure that you are running autotune and it is generating sane results. SMB magnifies any problems with incorrect basal rates or incorrect insulin sensitivity.
Dana Lewis
@danamlewis
Jun 02 2017 19:21
Also for all re autotune, ICYMI - if you’re on dev, check your autotune and if you have CSF = 0, you’ll want to re-pull the dev branch for the fix to that issue.
Cameron Chunn
@applehat
Jun 02 2017 19:34
My dexcom sensor would decide to become super noisy and jumpy after my first big meal looping. Poor thing has no idea what's going on.
Matthias Granberry
@mgranberry
Jun 02 2017 19:35
My experience is that it usually all averages out fine, but you always have the option of taking back control or setting a higher temporary target.
Cameron Chunn
@applehat
Jun 02 2017 19:36
It keeps jumping up and down 20 points. Rig keeps temp lowing and temp high over and over. So I mean, it's working I guess. Lol
Normally a sensor this new (10 days) will stabilize itself back out, but if it doesn't I'll change when I get home.
bryce
@bryce2071
Jun 02 2017 20:23
@danamlewis: i meant, that openAPS wont correct bei bgs .. its not working anymore. The dex5, explorer board /edison, Medtronic 522 are all connected (showed on my nightscout site).
everything looks like ever. But there will be no temp basalrates, which correct my bg.
The only thing i know, that i did is: i manually started a temp basalrate directly on my pump.
I think, after that change, openAPS stopped working.
The text showed on my nightscout site, if hovering the "OpenAPS"-Text is: "BG: 224, COB: 0, ... Target: 94: Eventual BG 189 >= 94, no temp, setting 5,65U/hr IOB: 1,52U".
It shows, that the BG needs to be corrected, but OpenAPS does no correction.
One thing, i think of, that it never been showed before is the "no temp".
What can i do, to get it working again?
Dana Lewis
@danamlewis
Jun 02 2017 20:24
What do your logs say @bryce2071?
Is your pump battery low so the RF comms are automatically disabled?
bryce
@bryce2071
Jun 02 2017 20:24
no, battery was changed yesterday :-(/
Dana Lewis
@danamlewis
Jun 02 2017 20:25
so what do your full logs say?
bryce
@bryce2071
Jun 02 2017 20:26
1 sec. please
Dana Lewis
@danamlewis
Jun 02 2017 20:29
(my next hunch is that you accidentally set the pump to % temp basal mode instead of absolute, if the last thing you had done was a manual temp basal. % disables looping, so we should see that referenced in the logs)
bryce
@bryce2071
Jun 02 2017 20:40
ohg
i think ive done this ..
1 sec please
yes ive changed this to percent .. this was the goal i think
Dana Lewis
@danamlewis
Jun 02 2017 20:42
that disables the ability to set temp basals - so change that back to “absolute”, and your rig should be able to successfully set temp basals again.
bryce
@bryce2071
Jun 02 2017 20:42
now i returned to rates
Dana Lewis
@danamlewis
Jun 02 2017 20:43
:+1: should set a temp after the next successful read, then
bryce
@bryce2071
Jun 02 2017 20:43
i have to wait now, if the corretions come back again!
Dana Lewis
@danamlewis
Jun 02 2017 20:43
you could also set a manual temp, slightly higher/lower than your current, and let it cancel that, so you know it’s working.
bryce
@bryce2071
Jun 02 2017 20:49
oh no: now i think, i lost the connection between the components again (this sometimes happens) .. cause, log shows:
No pump comms detected from other rigs
Profile less than 60m old
glucose.json newer than temp_basal.json
Temp refreshError, retrying
Listening: .No pump comms detected from other rigs
bash: oref0_init_pump_comms.py: command not found
mmtune: pump://JSON/mmtune/monitor/mmtune.json
reporting monitor/mmtune.json
Starting pump-loop at Fri Jun 2 22:47:44 CEST 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Profile less than 60m old
glucose.json newer than temp_basal.json
Temp refresh
how to quit the log again?!
Dana Lewis
@danamlewis
Jun 02 2017 20:50
control-c to exit, but what do you think shows an issue? It’s gotten a profile and glucose recently
bryce
@bryce2071
Jun 02 2017 20:51
strg-c doesnt work to quit log?!
do you think it works?
Dana Lewis
@danamlewis
Jun 02 2017 20:51
what is strg? if you’ve used something before that works, no reason not to do it again.
do I think what works? (Need some explanation behind which “it” you are referring to :D)
bryce
@bryce2071
Jun 02 2017 20:52
i have no ctrl .. this is a win surface. 1 sec please .. maybe "Alt Gr" + C wokrs?
you said: Its gooten a profile and glucose recently, and asked what i think - shows an issue...
acutal log:
Starting pump-loop at Fri Jun 2 22:47:44 CEST 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Profile less than 60m old
glucose.json newer than temp_basal.json
Temp refreshError, retrying
Starting pump-loop at Fri Jun 2 22:48:52 CEST 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Profile less than 60m old
glucose.json newer than temp_basal.json
Temp refreshError, retrying
Starting pump-loop at Fri Jun 2 22:49:59 CEST 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Profile less than 60m old
glucose.json newer than temp_basal.json
Temp refreshError, retrying
Starting pump-loop at Fri Jun 2 22:51:09 CEST 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Profile less than 60m old
glucose.json newer than temp_basal.json
Temp refreshError, retrying
Listening: .No pump comms detected from other rigs
bash: oref0_init_pump_comms.py: command not found
there are thinks like: command not found, or: No pump comms detected .. which i thought a hints that its not working
Dana Lewis
@danamlewis
Jun 02 2017 20:55
No pump comms detected means “there’s not anything already talking to the pump, go ahead"
If you have a keyboard without a control button, you can probably Google and find out the alternative to control-c. Control-c is basic linux, not openaps-specific
what type of rig do you have?
Diadon81
@Diadon81
Jun 02 2017 20:58
Dears do I need to install openaps-dev branch to use oref-dev?
in doc - Step 0: Dependencies - (Interested in the development repositories? See this shell script.)
Dana Lewis
@danamlewis
Jun 02 2017 20:59
@Diadon81 nope. just follow dev in Step 1
Diadon81
@Diadon81
Jun 02 2017 21:00
So the question is: do I need dev versions of dependencies to use oref-dev
ok, clear
bryce
@bryce2071
Jun 02 2017 21:01
ok: temp basals are back again. everythings working now like in the first 2 months :-)
thx dana
glucose.json newer than temp_basal.json
Temp refreshError, retrying
Starting pump-loop at Fri Jun 2 22:49:59 CEST 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Profile less than 60m old
glucose.json newer than temp_basal.json
Temp refreshError, retrying
Starting pump-loop at Fri Jun 2 22:51:09 CEST 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Profile less than 60m old
glucose.json newer than temp_basal.json
Temp refreshError, retrying
Listening: .No pump comms detected from other rigs
bash: oref0_init_pump_comms.py: command not found
mmtune: pump://JSON/mmtune/monitor/mmtune.json
reporting monitor/mmtune.json
"868.486", 5, -68 Starting pump-loop at Fri Jun 2 22:53:39 CEST 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Profile less than 60m old
glucose.json newer than temp_basal.json
Temp refreshed
{"carbs":0,"mealCOB":0,"reason":"carb_ratio 2.5199496010079803 out of bounds"}
{"ratio":1}
{"iob":0,"activity":0,"bolussnooze":0,"basaliob":0,"netbasalinsulin":0,"hightempinsulin":0,"time":"2017-06-02T20:53:12.000Z"}
{"delta":-3,"glucose":218,"short_avgdelta":-3.22,"long_avgdelta":-4.05}
{"duration":24,"rate":0.9,"temp":"absolute"}
target_bg unchanged: 100
Adjusting sens from 30.56 to 30.6
Carb Impact: -4 mg/dL per 5m; CI Duration: 0 hours
Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: 0 hours
determine-basal://text/shell/enact/suggested.json
reporting enact/suggested.json
{"temp":"absolute","bg":218,"tick":-3,"eventualBG":194,"snoozeBG":194,"predBGs":{"IOB":[218,214,211,208,205,203,201,199,198,197,196,196,196]},"COB":0,"IOB":0,"reason":"COB: 0, Dev: -24, BGI: 0, ISF: 31, Target: 100; Eventual BG 194 >= 100, temp 0.9<2.4U/hr","duration":30,"rate":2.4} pump://JSON/set_temp_basal/enact/enacted.json reporting enact/enacted.json {"bg":218,"temp":"absolute","snoozeBG":194,"recieved":true,"predBGs":{"IOB":[218,214,211,208,205,203,201,199,198,197,196,196,196]},"rate":2.375,"reason":"COB: 0, Dev: -24, BGI: 0, ISF: 31, Target: 100; Eventual BG 194 >= 100, temp 0.9<2.4U/hr","COB":0,"eventualBG":194,"timestamp":"2017-06-02T22:54:31.908425","duration":30,"tick":-3,"IOB":0} glucose.json newer than pumphistory: Refreshed pumphistory {"carbs":0,"mealCOB":0,"reason":"carb_ratio 2.5199496010079803 out of bounds"} {"ratio":1} {"iob":0.05,"activity":0,"bolussnooze":0,"basaliob":0.05,"netbasalinsulin":0.05,"hightempinsulin":0.05,"time":"2017-06-02T20:53:36.000Z"} {"delta":-3,"glucose":218,"short_avgdelta":-3.22,"long_avgdelta":-4.05} {"duration":30,"rate":2.375,"temp":"absolute"} target_bg unchanged: 100 Adjusting sens from 30.56 to 30.6 Carb Impact: -4 mg/dL per 5m; CI Duration: 0 hours Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: 0 hours determine-basal://text/shell/enact/suggested.json reporting enact/suggested.json {"temp":"absolute","bg":218,"tick":-3,"eventualBG":192,"snoozeBG":192,"predBGs":{"IOB":[218,214,211,208,205,203,201,199,197,196,195,194,193,192,192,191,190,189,188,187,186,185,184,183,182,182,181,180,180,179,178,178,178,177,177,177,176]},"COB":0,"IOB":0.05,"reason":"COB: 0, Dev: -24, BGI: 0, ISF: 31, Target: 100; Eventual BG 192 >= 100, temp 2.375 >~ req 2.3U/hr"}
{"bg":218,"temp":"absolute","snoozeBG":194,"recieved":true,"predBGs":{"IOB":[218,214,211,208,205,203,201,199,198,197,196,196,196]},"rate":2.375,"reason":"COB: 0, Dev: -24, BGI: 0, ISF: 31, Target: 100; Eventual BG 194 >= 100, temp 0.9<2.4U/hr","COB":0,"eventualBG":194,"timestamp":"2017-06-02T22:54:31.908425","duration":30,"tick":-3,"IOB":0}
Settings refreshed
pumphistory-24h refreshError, retrying
Listening: .No pump comms detected from other rigs
bash: oref0_init_pump_comms.py: command not found
mmtune: pump://JSON/mmtune/monitor/mmtune.json
reporting monitor/mmtune.json
"868.486", 5, -67 Starting pump-loop at Fri Jun 2 22:56:31 CEST 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Profile less than 60m old
glucose.json newer than temp_basal.json
Temp refreshed
Temp duration 27m or more
glucose.json newer than pumphistory: Refresh
Error, retrying
Starting pump-loop at Fri Jun 2 22:
Dana Lewis
@danamlewis
Jun 02 2017 21:02
:+1:
bryce
@bryce2071
Jun 02 2017 21:21
but i didnt get it working to exit the log with windows on putty
maybe some other windows useres are here, which have the problem that ctrl-c (i click on Strg-c) doesnt quit the logs
bryce
@bryce2071
Jun 02 2017 21:37
is it a problem that the logs is running all the time?
cause i cant quit logs with ctrl-c (or strg-c)
John Seaber
@jseaber
Jun 02 2017 21:39
Can you use a virtual keyboard on your Surface?
bryce
@bryce2071
Jun 02 2017 21:41
yes
elodaille01
@elodaille01
Jun 02 2017 21:49
Hi everybody...I reran oref0 script when Dana told me to do so (yesterday) to fix the autotune issue..(CSF = 0), all seems to be ok..expect for the carb ratio...(current = 12 instead of 10)..
Am I alone to have this trouble ?
thank you ;)

Parameter | Current | Autotune

ISF | 150.000 | 156.780
CSF | n/a | 12.887
Carb Ratio | 12.000 | 12.166
Basal Profile | |
00:00 | 0.100 | 0.100
00:30 | |
01:00 | | 0.100
01:30 | |
02:00 | 0.400 | 0.400
02:30 | 0.35 |
03:00 | | 0.348
03:30 | |
04:00 | 0.300 | 0.313
04:30 | |
05:00 | | 0.313
05:30 | |
06:00 | | 0.315
06:30 | 0.3 |
07:00 | | 0.300
07:30 | 0.3 |
08:00 | | 0.300
08:30 | 0.3 |
09:00 | | 0.290
09:30 | |
10:00 | 0.100 | 0.097
10:30 | |
11:00 | | 0.097
11:30 | |
12:00 | 0.500 | 0.500
12:30 | 1.5 |
13:00 | 1.200 | 1.200
13:30 | 0.1 |
14:00 | | 0.100
14:30 | |
15:00 | | 0.100
15:30 | 0.35 |
16:00 | | 0.350
16:30 | 0.5 |
17:00 | | 0.500
17:30 | 1 |
18:00 | | 1.000
18:30 | |
19:00 | | 1.000
19:30 | 0.5 |
20:00 | | 0.500
20:30 | |
21:00 | | 0.500
21:30 | 1 |
22:00 | 0.850 | 0.850
22:30 | |
23:00 | 0.800 | 0.800
23:30 | 0.5 |
Scott Leibrand
@scottleibrand
Jun 02 2017 21:56
what are your carb ratios in your pump? do you have multiple?
elodaille01
@elodaille01
Jun 02 2017 21:57
Yes Scott but the one at midnight is = 10
francesc0-cgm
@francesc0-cgm
Jun 02 2017 22:26
Q