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

8th
May 2017
David
@davidkeddydb
May 08 2017 00:16
ok, so I thought I put microbolus in the right spot to flag it as yes, and I went into preferences.json and put smb to try but I am not seeing a enact/smb-suggest.json. What did I not do?
katie disimone
@Kdisimone
May 08 2017 00:20
Are you saying you saw SMB options in the last three lines of preferences.json? Or they were not there?
David
@davidkeddydb
May 08 2017 00:21
sorry, typing skills. I saw all three in preferences.json and I set them to True, not try :)
this is from my Oref0-runagain: --enable=' autosens microbolus meal '
I am seeing it, I think I have something wacky somewhere as predicted always thinks my BG is going to 39:
Starting pump-loop at Sun May 7 20:19:14 EDT 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":56,"boluses":6.2,"mealCOB":41,"currentDeviation":15.06,"maxDeviation":16.05,"minDeviationSlope":-0.28}
{"iob":4.118,"activity":0.052,"bolussnooze":0.99,"basaliob":-0.218,"netbasalinsulin":-0.75,"hightempinsulin":0.25,"microBolusInsulin":0,"microBolusIOB":0,"time":"2017-05-08T00:19:22.000Z","lastBolusTime":1494199087000}
{"delta":1,"glucose":160,"short_avgdelta":-0.06,"long_avgdelta":1.49}
{"duration":4,"rate":0.9,"temp":"absolute"}
Autosens adjusting basal from 1 to 0.8; target_bg from 92.5 to 101; sens from 65 to 81.3
Carb Impact: 21.1 mg/dL per 5m; CI Duration: 2.6 hours; remaining 4h+ CI: 0 mg/dL per 5m
Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: 5.6 hours
UAM Impact: 21.1 mg/dL per 5m; UAM Duration: 3 hours
minPredBG: 93 minIOBPredBG: 39 minCOBPredBG: 93 minUAMPredBG: 999 COB: 41 carbs: 56
determine-basal://text/shell/enact/suggested.json
reporting enact/suggested.json
David
@davidkeddydb
May 08 2017 00:28
It says am I going to 39 but I am 179 diagonal up and nothing more being offered beyond an additional .1 in temp basal
I guess I have it, it is just not doing it yet
David
@davidkeddydb
May 08 2017 00:33
Maybe I have to use Autotune and see what happens
katie disimone
@Kdisimone
May 08 2017 00:55

@davidkeddydb read those tags a little slower…your minPredBG: 93

the other value of 39 is based on other values…iob only.
then there’s COB min predBG of 93
and then there’s min of the UAM predicted BG….999

But, it would be worth also checking for the reach of your pump loop…what’s the line say about the comparison of rig and pump times? What are the times on each (the loop-log says a message about checking for a 1 min difference)
katie disimone
@Kdisimone
May 08 2017 01:01
I’m not sure I understand the logic of SMB…here’s my example. I had thought that smb was designed to, in essence, front-load a correction rather than use the 30 min long temp basal to deliver it. The size of the front load depended on how much it could safely suspend to recover if it isn’t needed. I’m not seeing that behavior. Instead, 120 min basal suspensions seem to be the normal. And when SMBs have picked back up, they still continue to have a temp basal suspension, even when the SMBs are being delivered. Would seem to me that if we are correcting for a rising BG that is going to be above target…at least ending the suspension would be appropriate in light of an SMB being given.
IMG_2743 2.PNG
IMG_2744.PNG
IMG_2745.PNG
that suspension seems to be locked in for 2 hours. It ended two hours after it started…it didn’t end when the SMBs started or when the BGs started to rise.
David
@davidkeddydb
May 08 2017 01:06
Sorry, I misread that. But given what the predicted was, I don't get what SMB didn't kick in. Normally when I am 207 mg/dl I would just manually bolus. I was hoping that SMB was going to help with that rise
katie disimone
@Kdisimone
May 08 2017 01:29
again, read my question about the pump vs rig time. you need to check that
there’s a particular line in the pump-loop.log about looking for a 1 min difference. copy and paste that line and the resulting times from that check
David
@davidkeddydb
May 08 2017 01:30
let me look at the log
katie disimone
@Kdisimone
May 08 2017 01:30
:+1:
David
@davidkeddydb
May 08 2017 01:43
I am missing this time check. I am only seeing the last 2 times with timestamp and deliveredAT
Starting pump-loop at Sun May 7 21:30:12 EDT 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: Refreshed pumphistory
{"carbs":56,"boluses":6.2,"mealCOB":0,"currentDeviation":12.27,"maxDeviation":26.18,"minDeviationSlope":-1.545}
{"iob":0.71,"activity":0.03,"bolussnooze":0,"basaliob":0.047,"netbasalinsulin":-0.05,"hightempinsulin":0.15,"microBolusInsulin":0,"microBolusIOB":0,"time":"2017-05-08T01:30:36.000Z","lastBolusTime":1494199087000}
{"delta":0,"glucose":195,"short_avgdelta":-1.39,"long_avgdelta":-2.52}
{"duration":30,"rate":2.35,"temp":"absolute"}
Autosens adjusting basal from 1 to 0.85; target_bg from 92.5 to 99; sens from 90 to 107.1
Carb Impact: 14.7 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: 13.5 mg/dL per 5m; UAM Duration: 0.8 hours
minPredBG: 184 minIOBPredBG: 184 minCOBPredBG: 999 minUAMPredBG: 999 avgPredBG: 184 COB: 0 carbs: 56
determine-basal://text/shell/enact/suggested.json
reporting enact/suggested.json
{"temp":"absolute","bg":195,"tick":"+0","eventualBG":207,"snoozeBG":207,"insulinReq":0.74,"reservoir":null,"deliverAt":"2017-05-08T01:31:30.945Z","minPredBG":184,"predBGs":{"IOB":[195,192,190,189,187,187,186,187,187,188,190,191,191,190,190,189,189,188,188,187,187,187,186,186,186,185,185,185,184]},"COB":0,"IOB":0.71,"reason":"COB: 0, Dev: 88, BGI: -16.06, ISF: 107, Target: 99, minPredBG 184, IOBpredBG 184; Eventual BG 207 >= 102, temp 2.35 >~ req 2.35U/hr"}
{"insulinReq":0.76,"bg":195,"reservoir":null,"temp":"absolute","snoozeBG":211,"rate":2.35,"predBGs":{"IOB":[195,192,190,188,187,186,185,185,186,186,188,189,189,189,189,189,188,188,188,188,188,187,187,187,187,187,187,186]},"minPredBG":186,"IOB":0.714,"reason":"COB: 0, Dev: 92, BGI: -16.71, ISF: 107, Target: 99, minPredBG 186, IOBpredBG 186; Eventual BG 211 >= 102, temp 1.425<2.35U/hr","COB":0,"eventualBG":211,"recieved":true,"duration":30,"tick":"+0","timestamp":"2017-05-07T21:29:56.451277","deliverAt":"2017-05-08T01:29:49.591Z"}
Settings less than 10m old
Pumphistory-24 less than 20m old
Completed pump-loop at Sun May 7 21:31:33 EDT 2017
katie disimone
@Kdisimone
May 08 2017 01:45
Keep looking.
Scott Leibrand
@scottleibrand
May 08 2017 01:45
search for Microbolusing to find a time when it's trying to do one.
David
@davidkeddydb
May 08 2017 01:45
:+1:
katie disimone
@Kdisimone
May 08 2017 01:46

Should look like this

May 07 16:42:02 edisonSMB pump-loop.log: Checking deliverAt: 2017-05-07T23:42:00.676Z is within 1m of current time: Sun May 7 16:42:02 PDT 2017

David
@davidkeddydb
May 08 2017 01:47
so that is within SMB pump-loop.log? Is that a separate log from pump-loop?
Scott Leibrand
@scottleibrand
May 08 2017 01:49
@Kdisimone your sounds like it's working correctly. The way SMB works in that situation, is it calculates how large a microbolus is required based on how the various predBGs will play out if carb absorption continues as predicted. at the same time, it wants to cover you in case that doesn't happen, so it sets a long zero temp based on the naive_eventualBG - enough to bring BG back up to target eventually even if carb absorption suddenly stops (capped at 120 minutes)
that basically allows us to have it both ways: we can microbolus fairly aggressively, and as long as the microboluses are keeping up with the carb absorption, we can run a long zero temp as a backstop just in case
at that point, the SMBs are basically taking over your regular basal, plus whatever extra is needed for the carb absorption. if you compare the sizes of the SMBs to how much insulin would normally be delivered by your basal, you'll see that as long as carb absorption isn't slowing down, they're generally at or above what you'd get otherwise. for reference, a 0.1U microbolus every 5m is the equivalent of a 1.2U/hr basal.
David
@davidkeddydb
May 08 2017 01:51
@scottleibrand Thanks. When I look for microbolusing time on the log, I have ":0.3,"hightempinsulin":0.5,"microBolusInsulin":0,"microBolusIOB":0,"time":"2017-05-08T01:45:39.000Z","lastBolusTime":1494199087000}
Scott Leibrand
@scottleibrand
May 08 2017 01:52
@davidkeddydb that doesn't say Microbolusing
katie disimone
@Kdisimone
May 08 2017 01:52
Still not the right line
Jinx
Scott Leibrand
@scottleibrand
May 08 2017 01:52
look for Microbolusing, and then look at the lines immediately following that to see it attempting to check that it's safe to deliver the bolus. If no other checks are failing, then one of those will be the time check Katie mentioned
you could also search for is within 1m of current time to find that directly
katie disimone
@Kdisimone
May 08 2017 01:54
@scottleibrand ok...I'll redouble my math mental efforts on these. But for sure it sounds like the 120min temp basal suspend was coincidentally 2 hours today? It's not hard-coded (so to speak) to last 2 hours once enacted? So if a rise happens that suspension can be reverse sometime earlier than 120 min, correct?
David
@davidkeddydb
May 08 2017 01:54
I don't think I have had it Microbolusing yet. I will keep looking for it in the logs and I will search for that
Scott Leibrand
@scottleibrand
May 08 2017 01:54
and when I say "search", I don't mean "scan through the logs looking for it" ;-) I mean use something like less -R /var/log/openaps/pump-loop.log and then use the / key to search it
David
@davidkeddydb
May 08 2017 01:55
@scottleibrand I am glad you said that :)
Scott Leibrand
@scottleibrand
May 08 2017 01:55
@Kdisimone yes, if BG starts rising after SMB mode ends, it will set a high temp. if it's still in SMB mode, it'll usually just SMB more and let the zero temp run.
120m / 2 hours is not a coincidence: you see that a lot because that's the maximum it's capped out at, so as long as you have significant COB and deviations, it'll get to the point of needing a full 2h zero temp pretty quickly
katie disimone
@Kdisimone
May 08 2017 01:56
Ahhhhh. Now that is in-line with what I'm seeing. So seeing a zero temp will technically be more common with smb running
Scott Leibrand
@scottleibrand
May 08 2017 01:57
yes, with SMB running you should normally see really long zero temps, and all insulin delivered as microboluses, until COB drops near zero or something like that
David
@davidkeddydb
May 08 2017 01:57
@scottleibrand Hasn't shown up yet. I get pattern not found when I search for it thus far
Scott Leibrand
@scottleibrand
May 08 2017 01:58
the other thing you'll see is: immediately after entering lots of carbs and a big bolus for them, SMB will set a zero temp, but not start microbolusing yet until it observes enough deviations that it predicts all of the carbs will be absorbed within DIA. once that happens, it SMBs enough to catch up, and then just keeps going.
katie disimone
@Kdisimone
May 08 2017 01:59
Thank you so much. Final question then...(I think😜) it's still a mix of BGI and deviations and predBG that play into insulin needed calcs, right? (As I open GitHub to look more closely)
That makes so much more sense!
rsilvers129
@rsilvers129
May 08 2017 02:00
@diabeticgonewild - this is an example of what my friend got on eBay for $450. The pump is a brand-new 530G (not compatible with OpenAPS).
IMG_5575.JPG
katie disimone
@Kdisimone
May 08 2017 02:01
My bolus snooze has almost disappeared and zero temps start almost immediately. (Unsettling at first few uses obviously by my freak out brain fart here). I'm gonna have to write some crib sheets up for the new logic
Scott Leibrand
@scottleibrand
May 08 2017 02:03
@davidkeddydb another thing you can search for is insulinReq":1 to find any instances where you needed between 1 and 1.9 units of insulin and it should've been trying to SMB
@Kdisimone thanks for doing that. it's definitely a new paradigm that will require some explanation for people to get their heads around why it's doing what it's doing.
and yes, BGI and deviations and predBG are the main things involved in the insulinReq calcs. but definitely review the code and ask questions about the stuff that doesn't make sense. :)
David
@davidkeddydb
May 08 2017 02:05
@scottleibrand So, I really should be attempting to low bolus on something and see if SMB kicks in
If I just want to test if it is working or not working
Scott Leibrand
@scottleibrand
May 08 2017 02:06
@davidkeddydb and if that doesn't find anything, we can look at your NS for a time when it should've been SMB'ing. you can PM me your NS URL and I'll help you look
David
@davidkeddydb
May 08 2017 02:11
I have 2 instances of the time reference: 1 is Checking pump clock: "2017-05-07T20:47:28-04:00" is within 1m of current time: Sun May 7 20:48:19 EDT 2017
and that pumphistory is less than 1m old. Temp refreshed
{"carbs":56,"boluses":6.2,"mealCOB":13,"currentDeviation":26.18,"maxDeviation":30.47,"minDeviationSlope":-1.746}
Robin Kirkman
@esialb
May 08 2017 02:11
@scottleibrand I'm working on getting the power button on the explorer board to trigger restart or shutdown but it occurs to me that this might be a mal-feature
David
@davidkeddydb
May 08 2017 02:11
and the other is Checking deliverAt: 2017-05-08T00:48:31.449Z is within 1m of current time: Sun May 7 20:48:32 EDT 2017
and that smb-suggested.json is less than 1m old
enact/smb-suggested.json: {"insulinReq":0.125,"bg":212,"reservoir":"152.3","temp":"absolute","snoozeBG":179,"rate":1.05,"predBGs":{"COB":[212,213,211,207,201,19:
Robin Kirkman
@esialb
May 08 2017 02:11
@scottleibrand Is it desirable that the power button not do anything?
Scott Leibrand
@scottleibrand
May 08 2017 02:12
@esialb that used to work if you had acpid installed
Robin Kirkman
@esialb
May 08 2017 02:12
@scottleibrand yeah i'm recompiling a kernel with acpi support
Scott Leibrand
@scottleibrand
May 08 2017 02:12
I would prefer that a power button press issue a halt if that package is installed, but I haven't noticed it working lately
Robin Kirkman
@esialb
May 08 2017 02:12
alright i think that can work
Scott Leibrand
@scottleibrand
May 08 2017 02:12
it used to: not sure what the difference was
Robin Kirkman
@esialb
May 08 2017 02:12
different kernel probably
Scott Leibrand
@scottleibrand
May 08 2017 02:13
k
Robin Kirkman
@esialb
May 08 2017 02:13
anyway good to know that that is desired behavior
wasn't sure if edisons would be being powered off by accident
Scott Leibrand
@scottleibrand
May 08 2017 02:14
some people have designed their tic-tac cases to prevent pressure on the case from pressing down the power button
Eric
@ecc1
May 08 2017 02:15
I don't think vanilla acpid supports it, but it might be better to require something like three button presses in a row for a real shutdown, to avoid accidental poweroff.
Scott Leibrand
@scottleibrand
May 08 2017 02:15
but otherwise it wouldn't be an issue. and as long as those folks can remove the acpid package if they want, or otherwise disable the behavior, I'd prefer if a single press was all it took
Robin Kirkman
@esialb
May 08 2017 02:15
alright, acpid isn't installed by default anyway
Scott Leibrand
@scottleibrand
May 08 2017 02:16
:+1:
Eric
@ecc1
May 08 2017 02:16
Fair enough. Those who want something fancier can jump through more hoops :-)
Scott Leibrand
@scottleibrand
May 08 2017 02:16
yep. :)
@davidkeddydb from your NS, it looks like you don't have SMB enabled at all. what does your preferences.json look like?
David
@davidkeddydb
May 08 2017 02:20
well, the bottom half looks like: "min_5m_carbimpact": 5,
"carbratio_adjustmentratio": 1,
"autotune_isf_adjustmentFraction": 0.5,
"remainingCarbsCap": 60,
"enableUAM": false,
"enableSMB_with_bolus": true,
"enableSMB_with_COB": true,
"enableSMB_with_temptarget": true
Scott Leibrand
@scottleibrand
May 08 2017 02:20
cool, that looks correct
oh, I missed the key in the logs you pasted earlier. you have Starting pump-loop - if you had enabled SMB in oref0-setup, you'd have Starting supermicrobolus pump-loop
you'll need to re-run oref0-setup with the appropriate keyword in the command-line flag for enabling advanced featues
David
@davidkeddydb
May 08 2017 02:24
@scottleibrand Shit. I put microbolus in between autosens and meal, I didn't rerun Orefo set up. Smack forehead
Scott Leibrand
@scottleibrand
May 08 2017 02:26
there you go. :-)
David
@davidkeddydb
May 08 2017 02:31
@scottleibrand I think I am now on the right track glucose.json newer than pumphistory: Refresh
Starting supermicrobolus pump-loop at Sun May 7 22:30:03 EDT 2017 with 27 second wait_for_silence:
thanks as always
Dana Lewis
@danamlewis
May 08 2017 02:37
(Remember to reset maxiob since you reran, too)
Martin Haeberli
@mhaeberli
May 08 2017 02:45
@esialb running 1.1 kernel. I halt it. it shuts down. Then I plug it in again. I’m told that this should cause it to boot. But in my case it doesn’t . any advice?
David
@davidkeddydb
May 08 2017 02:47
@danamlewis and to go back and set the SMB to true after it unsets it!
Dana Lewis
@danamlewis
May 08 2017 02:48
:+1:
Robin Kirkman
@esialb
May 08 2017 02:49
@mhaeberli on explorer board with battery? i don't think it actually loses power with the battery
Martin Haeberli
@mhaeberli
May 08 2017 02:57
on explorer board, with battery
when battery goes to zero, (or small) then it shuts down
also, for testing, I was advised to try to shut it down, claim was then made that it should auto-power up
on another note, what are people finding is reasonable usage for papertrail per month? (my first device, which was on chatty 1.0 kernel, overflowed my papertrail free account in a few days)
Robin Kirkman
@esialb
May 08 2017 03:01
@mhaeberli i've never seen them auto-power-up on shutdown
@mhaeberli the only way i know of to restart a shut-down edison is to cycle the power
holding the reset button might do it
Martin Haeberli
@mhaeberli
May 08 2017 03:02
holding reset does the trick for me
but cycling external power with battery present doesn’t for some reason
Robin Kirkman
@esialb
May 08 2017 03:03
it shouldn't
with the battery present you aren't cycling the power
unless the battery is drained enough that it can't provide any power
Martin Haeberli
@mhaeberli
May 08 2017 03:04
@esialb test case is as follows - run edison / explorer until battery is so low it shuts down. plug in edison/explorer to charge battery. hope and pray edison starts up. doesn’t start up. unless button is pressed and held. but this is a bit of a problem because we’d like to keep the case closed, but can’t press button with case closed
Robin Kirkman
@esialb
May 08 2017 03:05
@mhaeberli the battery has to be drained enough that the edison is receiving no power at all
@mhaeberli this for example would work https://www.adafruit.com/product/3064
Martin Haeberli
@mhaeberli
May 08 2017 03:06
@esialb I assume it was as both red LEDs were off … and it wasn’t responding on network
Robin Kirkman
@esialb
May 08 2017 03:07
needs to be low enough that the edison is not receiving power at all, even to maintain a halt state
Martin Haeberli
@mhaeberli
May 08 2017 03:07
do red LEDs go off BEFORE that point
Robin Kirkman
@esialb
May 08 2017 03:07
i am not sure what the explorer board leds are indicating though
Martin Haeberli
@mhaeberli
May 08 2017 03:07
?
@esialb ok … thx
Robin Kirkman
@esialb
May 08 2017 03:07
they might
anyway if you're using a case where you can't get to the power button then a switch inline with the battery cable may be needed
Martin Haeberli
@mhaeberli
May 08 2017 03:15
@esialb don’t want to disturb @scottleibrand etc but few days ago he expressed surprise that the edison/explorer wasn’t auto-booting when power came back up far enough
Scott Leibrand
@scottleibrand
May 08 2017 03:24
My rigs always power on when I plug them in, if their batteries are charged and I had halted them or powered them off with the button.
Martin Haeberli
@mhaeberli
May 08 2017 03:26
@scottleibrand @esialb so I'm trying to figure out what to do differently
katie disimone
@Kdisimone
May 08 2017 03:32
mine behave as @scottleibrand described too. If I use power button to turn them off, plugging into wall socket wakes them up
RedCyclist1
@RedCyclist1
May 08 2017 03:36
I added the cannula age pill to my NS. How do you track it? How is the field populated?
rsilvers129
@rsilvers129
May 08 2017 03:37
I think the first version of the Edison didn't auto boot.
I mean Explorer.
Martin Haeberli
@mhaeberli
May 08 2017 03:38
@RedCyclist1 to track it, use the + button at the top right of a logged in NightScout instance to add a set change treatment
Scott Leibrand
@scottleibrand
May 08 2017 03:38
@mhaeberli what model of Explorer board do you have? It's probably printed on it, opposite the antenna.
Martin Haeberli
@mhaeberli
May 08 2017 03:39
One original, one 1.2, same behavior I think.
I have to audit that statement
As in double check the devices
Could be the non auto boots are both originals
RedCyclist1
@RedCyclist1
May 08 2017 03:43
@mhaeberli Got it, thanks!
Martin Haeberli
@mhaeberli
May 08 2017 04:03
confirmed that both boards are original Explorers - I’ll try my 1.2; probably means that for this particular packaging I’ll need 1.2 s
or later
Martin Haeberli
@mhaeberli
May 08 2017 04:23
besides commenting out the crontab line for pump-loop, is there any easy way to turn off ‘enact’ - ideally, so recommendations / basals are for now still shown in Nightscout but we pretend as if the pump is not loop able?
(I want to leave that crontab row in because I still want loop computation to run…)
Scott Leibrand
@scottleibrand
May 08 2017 04:54
If you're running master, deleting the enact alias should do it. With dev, you'd need to comment it out in oref0-pump-loop.sh
Martin Haeberli
@mhaeberli
May 08 2017 05:10
k thx
Martin Haeberli
@mhaeberli
May 08 2017 05:23
so, for example, on master, openaps alias show enact ; save what it says there; then openaps alias remove enact … then, later, openaps alias add enact "rm enact/suggested.json; openaps invoke enact/suggested.json && if (cat enact/suggested.json && grep -q duration enact/suggested.json); then ( rm enact/enacted.json; openaps invoke enact/enacted.json ; grep -q duration enact/enacted.json || openaps invoke enact/enacted.json ) 2>&1 | egrep -v \"^ |subg_rfspy|handler\"; fi; grep incorrectly enact/suggested.json && oref0-set-system-clock 2>/dev/null; cat enact/enacted.json | json -0" or something like that
Scott Leibrand
@scottleibrand
May 08 2017 05:27
Something like that. I think that add syntax is a little off.
Or you could re-run oref0-setup later.
Martin Haeberli
@mhaeberli
May 08 2017 05:53
:+1:
Scott Leibrand
@scottleibrand
May 08 2017 06:12
regarding pump clock mismatches, I just pushed https://github.com/openaps/oref0/tree/set-pump-clock if anyone wants to test it (by setting your pump clock to >10s off, and verifying it re-sets it). haven't tested it yet myself, as it's bed time.
PieterGit
@PieterGit
May 08 2017 06:32
@scottleibrand if the rig clock is off (e.g by not having rtc on rpi3) one should check if year>2016 and if it's not synchronized before changing the time of the pump I think
Scott Leibrand
@scottleibrand
May 08 2017 06:33
oref0-set-device-clocks makes sure ntp is sync'd first.
Martin Haeberli
@mhaeberli
May 08 2017 06:45
-my ntp sync is failing… and, on another front, bg values don’t seem to be updating even though NS API shows new values. cgm/glucose.json doesn’t get update unless I run ns-loop by hand. but ns-loop log shows it’s running regularly
Martin Haeberli
@mhaeberli
May 08 2017 06:52
although: openaps get-bg AttributeError: 'NoneType' object has no attribute 'iter_records' grep: cgm/cgm-glucose.json: No such file or directory
when run standalone...
(sorta like last week; even tried rebooting and didn’t settle it this time)...
also, what does ‘not enacted’ mean?
Martin Haeberli
@mhaeberli
May 08 2017 07:18
ok, figured out ‘not enacted’ meant for example no pump comms
David
@davidkeddydb
May 08 2017 12:54
Screenshot 2017-05-08 08.51.23.png
I wish that the zero temp of basal started later as for me, the post prandial rise was to great. Trying SMB for the first time. I also have autotune and autosens enacted as well. I know the exact carbs of the meal and typically things work but today they didn't.
tepidjuice
@tepidjuice
May 08 2017 12:58
I agree with @davidkeddydb the zero temp starts too quickly.
David
@davidkeddydb
May 08 2017 13:02
@tepidjuice In this case, I would have been better off and I haven't calculated whether less insulin or not with just the regular basal running in the background and skipped any SMB.
David
@davidkeddydb
May 08 2017 13:10
Normally I would have bolused going over 160 mg/dl post prandial 2 hours but I am letting this cycle through to see what it does. 227 mg/dl at present and seems to be flattening out
David
@davidkeddydb
May 08 2017 13:31
Screenshot 2017-05-08 09.30.53.png
tepidjuice
@tepidjuice
May 08 2017 13:35
that pretty much reflects what I see. I give a prebolus and remain relatively flat after eating and get 0 temp'ed until the inevitable rise comes which is made worse by the 0 temping. Then the smb's are just enough to keep me flat but not bring me in range.
David
@davidkeddydb
May 08 2017 13:37
@tepidjuice If that is the case, I will likely turn it off and just stick with AMA. How many meals have you used it over? Have you tried UAM yet?
tepidjuice
@tepidjuice
May 08 2017 13:38
@davidkeddydb I've a remaining carbs cap of 20 with UAM enabled.
and with enable_SMB_with_bolus set to true
now I've added enableSMB_with_COB: true to see if I get better responses
I've tried it on about 5 meals so far.
fidoman
@efidoman
May 08 2017 13:40
Anyone seen this error before .... Waiting up to 4 minutes for new BG: date: invalid date ‘@’
David
@davidkeddydb
May 08 2017 13:41
I wasn't sure what the right number was to put in the carb cap. I know it has a max in there preset of 90 and you can adjust. I think it was what carbs can be absorbed over 4 hours and I believe I used 50. I have enableSMB_with_COB: true
tepidjuice
@tepidjuice
May 08 2017 13:41
Every single one has had pretty bad responses to the point where I wondered if it was the insulin but when I went back to manual control I stayed in range the entire time,
I wanted to try out a small number at first to make sure it didn't over bolus but I think it has the inverse problem.
i.e. too little insulin too late.
David
@davidkeddydb
May 08 2017 13:44
@tepidjuice I think you are right. I only posted as I wanted to give @scottleibrand an instance to look at. It may be adjustments of some type have to be made with the PredBG somewhere so that what works for @danamlewis works for us. :)
tepidjuice
@tepidjuice
May 08 2017 13:45
@davidkeddydb Yes. Amazing how much we all vary :)
fidoman
@efidoman
May 08 2017 14:20
I figured out why SMB wasn't running ... added the appropriate flag to --enable in runagain. Now, I continue to get the following error and can't figure out why: Starting supermicrobolus pump-loop at Mon May 8 10:18:48 EDT 2017 with 13 second wait_for_silence:
Waiting up to 4 minutes for new BG: date: invalid date ‘@’
The "date: invalid date '@' " part seems to be what is causing problems
Dana Lewis
@danamlewis
May 08 2017 14:24
@ericfeibelman what's the date on your CGM? Does it match pump?
fidoman
@efidoman
May 08 2017 14:26
Hi Dana - I'm using the 722 with the enlite sensor and the date on the pump matches the date on the rig....
Here's what I see in the logpumphistory.json: "Battery 2017-05-08T10:15:11 head[2], body[0] op[0x1a]"
Checking pump clock: "2017-05-08T10:22:38-04:00" is within 1m of current time: Mon May 8 10:22:53 EDT 2017
Also see this No pump comms detected from other rigs
Preflight OK. Profile less than 60m old. date: invalid date ‘@’

Here's some more, but I think it is related to the "date: invalid date '@' issue ........ Starting supermicrobolus pump-loop at Mon May 8 10:27:51 EDT 2017 with 24 second wait_for_silence:
Waiting up to 4 minutes for new BG: date: invalid date ‘@’
ls: cannot access monitor/pump_loop_completed: No such file or directory
Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK. Profile less than 60m old. date: invalid date ‘@’
find: `monitor/pump_loop_completed': No such file or directory
pump_loop_completed less than 5m ago. Refresh at node.js:906:3
ed pumphistory
pumphistory.json: "Battery 2017-05-08T10:15:11 head[2], body[0] op[0x1a]"
Checking pump clock: "2017-05-08T10:28:27-04:00" is within 1m of current time: Mon May 8 10:28:42 EDT 2017
and that pumphistory is less than 1m old. Temp refreshed
Could not parse input data: [SyntaxError: /root/myopenaps/monitor/glucose.json: Unexpected end of input]
Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 82, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 51, in call
self.run(self.args)
File "/usr/local/bin/openaps-report", line 75, in run
output = app(args, self)
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/subcommand.py", line 52, in call
return self.method.main(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps/reports/invoke.py", line 48, in main
reporters.Reporter(report, device, task)(output)
File "/usr/local/lib/python2.7/dist-packages/openaps/reports/reporters/init.py", line 33, in call
self.blob = self.serialize(data)
File "/usr/local/lib/python2.7/dist-packages/openaps/reports/reporters/init.py", line 31, in serialize
return self.method.serialize(render(data), self)
File "/usr/local/lib/python2.7/dist-packages/openaps/vendors/process.py", line 56, in prerender_json
return json.loads(data)
File "/usr/lib/python2.7/json/init.py", line 338, in loads
return _default_decoder.decode(s)
File "/usr/lib/python2.7/json/decoder.py", line 366, in decode
obj, end = self.raw_decode(s, idx=_w(s, 0).end())
File "/usr/lib/python2.7/json/decoder.py", line 384, in raw_decode
raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded
Error, retrying
Sleeping 2;

Starting supermicrobolus pump-loop at Mon May 8 10:28:56 EDT 2017 with 10 second wait_for_silence:
Waiting up to 4 minutes for new BG: date: invalid date ‘@’
ls: cannot access monitor/pump_loop_completed: No such file or directory
Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK. Profile less than 60m old. date: invalid date ‘@’

Scott Leibrand
@scottleibrand
May 08 2017 14:42
The invalid date '@' and lack of pump_loop_completed are symptoms, not the cause. Looks like there is a problem with your glucose.json.
@tepidjuice @davidkeddydb if SMB is starting too late for you, I would raise remainingCarbsCap to 90 for starters.
Do you have any examples that don't have a CGM gap exactly where the SMBs should've been starting? 🙂
David
@davidkeddydb
May 08 2017 14:45
@scottleibrand Done. See what happens with lunch in a few hours
Scott Leibrand
@scottleibrand
May 08 2017 14:46
Would also be useful to see the purple line predictions from when you think it should be dosing insulin and it's holding off.
tepidjuice
@tepidjuice
May 08 2017 14:46
@scottleibrand I'll try it out when I next eat.
All of them are predicting a low.
David
@davidkeddydb
May 08 2017 14:47
@scottleibrand I have those on my phone but not on my mac, but same as @tepidjuice with the pred lines
Scott Leibrand
@scottleibrand
May 08 2017 14:47
That's typical at first: they should tick up as soon as carb absorption actually starts.
Also, try to do Eating soon and/or pre-enter your carbs to get some insulin going early: you started high and bolused right with the meal it looks like.
David
@davidkeddydb
May 08 2017 14:48
:+1:
fidoman
@efidoman
May 08 2017 14:53
Thanks David - you may be on to something - The date on the glucose.json file is right but the time is 00:00
74608 May 8 00:00 glucose.json
David
@davidkeddydb
May 08 2017 14:53
Autotune question, do most people after running autotune go in and change their pump settings to what autotune has?
tepidjuice
@tepidjuice
May 08 2017 14:56
@scottleibrand my problem seems to be that the prebolus is making openaps predict a low which won't happen as the carb absorption is about to start. Also I never eat more than about 30g CHO in a single meal so would a cap of 90g not be overkill?
Dana Lewis
@danamlewis
May 08 2017 14:59
@davidkeddydb two options: run it as a one-off, or run it automatically as part of rig. The latter will do tweaks to the basal profile, capped by difference in underlying pump profile. Worth keeping an eye on over time to see if you're constantly hitting the cap; if so probably worth changing the underlying settings.
for one-off usage, seems to vary person to person on whether they want to do that or not.
Scott Leibrand
@scottleibrand
May 08 2017 15:01
@tepidjuice ok, yeah. Just over your max meal size is fine. Another thing to try is using eating soon to make SMB more aggressive for the first hour or so after the meal.
We do the immediate zero temping to help avoid postprandial lows from larger boluses.
David
@davidkeddydb
May 08 2017 15:02
@danamlewis Gotcha. I am running it as part of rig and will keep an eye on it.
Scott Leibrand
@scottleibrand
May 08 2017 15:02
With the idea that we can counteract them quickly if absorption actually does pick up.
But if you don't have postprandial lows, being more aggressive with eating soon and pre-entering carbs will preemptively counteract the zero temps and help a lot.
fidoman
@efidoman
May 08 2017 15:12
tried doing a force bg first upload and that didn't change things ... getting this error also ... Checking deliverAt: null is within 1m of current time: Mon May 8 11:11:35 EDT 2017
date: invalid date ‘null’
fidoman
@efidoman
May 08 2017 15:39
Problem solved - just reinstalled everything from scratch again and the date: problem went away.
ejoe132
@ejoe132
May 08 2017 15:42
Had the battery wire break off my batter on my edison, anyone know where I can order another jst connector battery qucik? Or if there is one that is larger capacity then the one I had gotten from the explorer board page?
katie disimone
@Kdisimone
May 08 2017 15:55
@ejoe132 there's a link in here to buy the 2500 adafruit, but through amazon since adafruit direct is backordered. Also, look at bottom of post for hint to keep battery wires safe with tape
Eric
@ecc1
May 08 2017 16:10
There are also a bunch of hits on how to repair a disconnected LiPo cell wire yourself ...
Martin Haeberli
@mhaeberli
May 08 2017 16:41
re-asking - what are typical amounts of monthly data for papertrail for one edison running smoothly?
katie disimone
@Kdisimone
May 08 2017 17:02
@mhaeberli under 1gb. I set up log filters to prevent uploading of CRON libmraa and sudo to save on papertrail data limits (those add little value for troubleshooting rig performance). I have three rigs and stay under 1gb easily
Martin Haeberli
@mhaeberli
May 08 2017 17:03
@Kdisimone :+1:
katie disimone
@Kdisimone
May 08 2017 17:06
@mhaeberli I just looked it up. Four weeks was 452mb. Halfway through that four weeks was when I setup the filters for the cron, libmraa, sudo. Data use dropped after that for sure
ejoe132
@ejoe132
May 08 2017 18:03
@Kdisimone Thank you
bjarecki
@bjarecki
May 08 2017 18:08
where should I be able to confirm my loop is using my autotune settings. Autotune is running each night and autotune.json in my settings folder is updated, and differs from my orginal pump settings. However from the looks of everything ISF, carb ratio, neutral basil , etc all seem to remain set to my original pump settings.
Dana Lewis
@danamlewis
May 08 2017 18:10
@bjarecki that means it's using it.
It can't change your pump settings on the pump; but it sets a new profile on the rig that it uses. So if it differs, that's what it's using if you enabled it to run automatically
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 18:13
I have been having NTP issues so I finally resorted to re-flashing my edison to see it that helps. I am on this page http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-0/setup-edison.html
root@in10aps:~# dpkg-reconfigure tzdata

Current default time zone: 'US/Central'
Local time is now:      Fri Dec 31 19:17:56 CST 1999.
Universal Time is now:  Sat Jan  1 01:17:56 UTC 2000.

root@in10aps:~#
is it ok that the time is WAY off???
(this has been my problem)
I am connected to wifi, I just completed all the steps to get here.
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 18:19
ran again to confirm
root@in10aps:~# dpkg-reconfigure tzdata

Current default time zone: 'America/North_Dakota/Center'
Local time is now:      Fri Dec 31 19:27:30 CST 1999.
Universal Time is now:  Sat Jan  1 01:27:30 UTC 2000.

root@in10aps:~#
I can only guess it is my NTP server at work causing issues???
Martin Haeberli
@mhaeberli
May 08 2017 18:21
@diabetic_dude_twitter a few hypotheses: - a) your work firewall blocks requests out to public ntp servers
not sure I have many others
I’m about to look up what ntp servers are configured in my edison.
you could try pinging whatever that turns out to be.
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 18:21
That could be the case I am thinking the same
Martin Haeberli
@mhaeberli
May 08 2017 18:21
one place to look would be /etc/ntp.conf
bjarecki
@bjarecki
May 08 2017 18:25
@danamlewis So all the numbers I am shown in my log file and on hover in nightscout actually do not accurately reflect the number the loop is using? Cause when i hover my APS badge in NS for instance I see my pumps ISF not the one in autotune.json.
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 18:26
@mhaeberli what would I use to view ntp.conf ??? type ntp.conf or nano ntp.conf ???
Martin Haeberli
@mhaeberli
May 08 2017 18:27
nano /etc/ntp.conf
to edit it
or less /etc/ntp.conf - to type it out and you can page forward with “space” if it’s more than a screen
dan lightner
@danimaniac
May 08 2017 18:27
Anyone successfully bt tether / offline loop with a Virgin Mobile phone? I am considering a change for other reasons but am unwilling to switch if my OpenAPS rig will not function how I want it to function. I paired my rig and established a bt-pan with my son's Virgin Mobile device...It seemed to work fine, but I just want to verify with the collective...and the search function yields nothing...yes, I searched.
Martin Haeberli
@mhaeberli
May 08 2017 18:27
or vi /etc/ntp.conf (if you prefer vi as an editor)
Dana Lewis
@danamlewis
May 08 2017 18:28
@bjarecki what do your pump loop logs and OpenAPS pill say?
aabramowicz
@aabramowicz
May 08 2017 18:29
in building my second OpenAPS rig, when trying to install the dependencies for Oref0 using the setup script, it doesn't finish with Openaps 0.1.5 but instead "sudo pip command not found"
Martin Haeberli
@mhaeberli
May 08 2017 18:31
@diabetic_dude_twitter ok, looked at mine, not clear to me whether the 0.debian.pool.ntp.org is what should be doing the trick
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 18:31
yup, I see what you are looking at
bjarecki
@bjarecki
May 08 2017 18:32
blob
Eric
@ecc1
May 08 2017 18:32
I'm using systemd's built-in NTP replacement, systemd-timesyncd. I wonder whether it is more robust. It does at least one thing better: it saves its last time update to disk, so at least the time will be in the right ballpark upon reboot on systems like Edison with no RTC.
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 18:33
@mhaeberli I did get replies
Martin Haeberli
@mhaeberli
May 08 2017 18:33
@diabetic_dude_twitter see diagnostic suggestions at http://www.tldp.org/LDP/sag/html/ntp-toolkit.html
@ecc1 thx - can you send a link to setup instructions?
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 18:33
@ecc1 I tried that too but got drug down the same rabbit hole
(but different errors)
Eric
@ecc1
May 08 2017 18:36
systemctl enable systemd-timesyncd
apt-get purge ntp
systemctl start systemd-timesyncd
(Above commands all as root.) Wait a minute, then do timedatectl to see see if time is synchronized correctly.
Dana Lewis
@danamlewis
May 08 2017 18:36
@bjarecki what's your pump isf vs Autotune isf? And do you have autosens also turned on?
bjarecki
@bjarecki
May 08 2017 18:37
@danamlewis So in my autotune.json my sensitivity is "sens":10.151
but my pill and logs say 12. That is the same value in my pump
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 18:38
Thanks
I am going to save the info from @ecc1 and have a look at the NTP toolkit...
also going to try leaving work and see what I get from my phone's wifi
Dana Lewis
@danamlewis
May 08 2017 18:40
@bjarecki do you have autosens turned on?
bjarecki
@bjarecki
May 08 2017 18:41
@danamlewis Hmm I did not enable autosense when I ran oref
kenstack
@kenstack
May 08 2017 18:42
@diabetic_dude_twitter my kid runs into a couple of networks where ntp is blocked - so now I just set the Edison time off the pump time whenever the Edison reboots - it's all relative time anyway I think
bjarecki
@bjarecki
May 08 2017 18:42
@danamlewis Though now that you mention it there is stuff about it making adjustments in the log
Dana Lewis
@danamlewis
May 08 2017 18:44
@bjarecki cat oref0-runagain.sh for me and paste it here, but xx out your NS URL and API
Martin Haeberli
@mhaeberli
May 08 2017 18:45
@ecc1 systemctl start systemd-timesyncd Job for systemd-timesyncd.service failed. See 'systemctl status systemd-timesyncd.service' and 'journalctl -xn' for details.
i’ll try rebooting as an initial matter
bjarecki
@bjarecki
May 08 2017 18:46
@danamlewis oref0-setup --dir=/root/myopenaps --serial=x --cgm=xdrip --ns-host=x --api-secret=x --tty=/dev/spidev5.1 --enable=' autotune ' --radio_locale='US' --btmac=x
Dana Lewis
@danamlewis
May 08 2017 18:47
@bjarecki k. Ok, paste some of your pump-loop.log?
aabramowicz
@aabramowicz
May 08 2017 18:51
@scottleibrand I'm reflashing
bjarecki
@bjarecki
May 08 2017 18:51
@danamlewis sure
Starting pump-loop at Mon May 8 13:53:14 EDT 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
{"iob":6.766,"activity":0.0815,"bolussnooze":5.529,"basaliob":0.029,"netbasalinsulin":0.5,"hightempinsulin":0.5,"microBolusInsulin":0,"microBolusIOB":0,"time":"2017-05-08T17:53:42.000Z","lastBolusTime":1494265419000}
{"delta":-6,"glucose":177,"short_avgdelta":-3.5,"long_avgdelta":-0.45}
{"duration":0,"rate":1.9,"temp":"absolute"}
Autosens adjusting basal from 1.92 to 1.9; target_bg unchanged: 90; sens unchanged: 12
Carb Impact: -1.1 mg/dL per 5m; CI Duration: NaN hours; remaining 4h+ CI: NaN mg/dL per 5m
Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: NaN hours
UAM Impact: 1.4 mg/dL per 5m; UAM Duration: 0 hours
minPredBG: 86 minIOBPredBG: 86 minCOBPredBG: 999 minUAMPredBG: 999 avgPredBG: 86 COB: undefined carbs: undefined
determine-basal://text/shell/enact/suggested.json
reporting enact/suggested.json
{"temp":"absolute","bg":177,"tick":-6,"eventualBG":104,"snoozeBG":204,"insulinReq":0,"reservoir":null,"deliverAt":"2017-05-08T17:54:17.642Z","minPredBG":999,"predBGs":{"IOB":[177,171,165,159,154,148,142,137,131,125,119,114,109,105,101,98,95,92,90,88,87,86]},"IOB":6.766,"reason":"COB: undefined, Dev: 8, BGI: -4.89, ISF: 12, Target: 90, minPredBG 86, IOBpredBG 86; 104-86 in range: no temp required; setting current basal of 1.9 as temp","duration":30,"rate":1.9}
rm: cannot remove ‘enact/enacted.json’: No such file or directory
pump://JSON/set_temp_basal/enact/enacted.json
reporting enact/enacted.json
{"insulinReq":0,"bg":177,"reservoir":null,"temp":"absolute","snoozeBG":204,"rate":1.9,"predBGs":{"IOB":[177,171,165,159,154,148,142,137,131,125,119,114,109,105,101,98,95,92,90,88,87,86]},"minPredBG":999,"IOB":6.766,"reason":"COB: undefined, Dev: 8, BGI: -4.89, ISF: 12, Target: 90, minPredBG 86, IOBpredBG 86; 104-86 in range: no temp required; setting current basal of 1.9 as temp","timestamp":"2017-05-08T13:54:22.372660","eventualBG":104,"recieved":true,"duration":30,"tick":-6,"deliverAt":"2017-05-08T17:54:17.642Z"}
glucose.json newer than pumphistory: Refreshed pumphistory
{"iob":6.732,"activity":0.0816,"bolussnooze":5.492,"basaliob":0.028,"netbasalinsulin":0.5,"hightempinsulin":0.5,"microBolusInsulin":0,"microBolusIOB":0,"time":"2017-05-08T17:54:07.000Z","lastBolusTime":1494265419000}
{"delta":-6,"glucose":177,"short_avgdelta":-3.5,"long_avgdelta":-0.45}
{"duration":30,"rate":1.9,"temp":"absolute"}//cm

Autosens adjusting basal from 1.92 to 1.9; target_bg unchanged: 90; sens unchanged: 12
Carb Impact: -1.1 mg/dL per 5m; CI Duration: NaN hours; remaining 4h+ CI: NaN mg/dL per 5m
Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: NaN hours
UAM Impact: 1.4 mg/dL per 5m; UAM Duration: 0 hours
minPredBG: 86 minIOBPredBG: 86 minCOBPredBG: 999 minUAMPredBG: 999 avgPredBG: 86 COB: undefined carbs: undefined
determine-basal://text/shell/enact/suggested.json
reporting enact/suggested.json
{"temp":"absolute","bg":177,"tick":-6,"eventualBG":104,"snoozeBG":203,"insulinReq":0,"reservoir":null,"deliverAt":"2017-05-08T17:54:54.674Z","minPredBG":999,"predBGs":{"IOB":[177,171,165,159,154,148,142,137,131,125,119,114,109,105,101,98,95,92,90,89,87,86]},"IOB":6.732,"reason":"COB: undefined, Dev: 8, BGI: -4.9, ISF: 12, Target: 90, minPredBG 86, IOBpredBG 86; 104-86 in range: no temp required, temp 1.9 ~ req 1.9U/hr"}
{"insulinReq":0,"bg":177,"reservoir":null,"temp":"absolute","snoozeBG":204,"rate":1.9,"predBGs":{"IOB":[177,171,165,159,154,148,142,137,131,125,119,114,109,105,101,98,95,92,90,88,87,86]},"minPredBG":999,"IOB":6.766,"reason":"COB: undefined, Dev: 8, BGI: -4.89, ISF: 12, Target: 90, minPredBG 86, IOBpredBG 86; 104-86 in range: no temp required; setting current basal of 1.9 as temp","timestamp":"2017-05-08T13:54:22.372660","eventualBG":104,"recieved":true,"duration":30,"tick":-6,"deliverAt":"2017-05-08T17:54:17.642Z"}
Settings less than 10m old
Pumphistory-24 less than 20m old
Completed pump-loop at Mon May 8 13:54:57 EDT 2017
Martin Haeberli
@mhaeberli
May 08 2017 18:51
@ecc1
# systemctl status systemd-timesyncd.service
● systemd-timesyncd.service - Network Time Synchronization
   Loaded: loaded (/lib/systemd/system/systemd-timesyncd.service; enabled)
   Active: failed (Result: start-limit) since Mon 2017-05-08 11:49:08 PDT; 56s ago
     Docs: man:systemd-timesyncd.service(8)
  Process: 1469 ExecStart=/lib/systemd/systemd-timesyncd (code=exited, status=226/NAMESPACE)
 Main PID: 1469 (code=exited, status=226/NAMESPACE)

May 08 11:49:06 openaps_n systemd[1]: Failed to start Network Time Synchronization.
May 08 11:49:06 openaps_n systemd[1]: Unit systemd-timesyncd.service entered failed state.
May 08 11:49:06 openaps_n systemd[1]: systemd-timesyncd.service has no holdoff time, scheduling restart.
May 08 11:49:06 openaps_n systemd[1]: Stopping Network Time Synchronization...
May 08 11:49:06 openaps_n systemd[1]: Starting Network Time Synchronization...
May 08 11:49:07 openaps_n systemd[1]: systemd-timesyncd.service: main process exited, code=exited, status=226/NAMESPACE
May 08 11:49:07 openaps_n systemd[1]: Failed to start Network Time Synchronization.
May 08 11:49:08 openaps_n systemd[1]: systemd-timesyncd.service: main process exited, code=exited, status=226/NAMESPACE
May 08 11:49:08 openaps_n systemd[1]: Failed to start Network Time Synchronization.
May 08 11:49:08 openaps_n systemd[1]: systemd-timesyncd.service has no holdoff time, scheduling restart.
May 08 11:49:08 openaps_n systemd[1]: Stopping Network Time Synchronization...
May 08 11:49:08 openaps_n systemd[1]: Starting Network Time Synchronization...
May 08 11:49:08 openaps_n systemd[1]: systemd-timesyncd.service start request repeated too quickly, refusing to start.
May 08 11:49:08 openaps_n systemd[1]: Failed to start Network Time Synchronization.
May 08 11:49:08 openaps_n systemd[1]: Unit systemd-timesyncd.service entered failed state.
Chuck Vann
@djtechman
May 08 2017 18:55
So I tried the dev build for a while, but it would not stay reliably connected to the pump (posted some of the errors on Saturday morning.... But now I'm trying to start clean with the master branch.
I put everything from my /root/ dir into an "oldstuff" dir (including the "src") dir.
But now when I run the automated install script from Phase 2 in the doc, I get this error:
oref0@0.4.2 /usr/local/lib/node_modules/oref0
├── crypto@0.0.3
├── moment-timezone@0.5.11
├── moment@2.18.1
├── lodash@4.17.4
├── share2nightscout-bridge@0.1.5 (request@2.53.0)
├── request@2.81.0 (is-typedarray@1.0.0, aws-sign2@0.6.0, oauth-sign@0.8.2, forever-agent@0.6.1, tunnel-agent@0.6.0, stringstream@0.0.5, caseless@0.12.0, isstream@0.1.2, json-stringify-safe@5.0.1, performance-now@0.2.0, safe-buffer@5.0.1, extend@3.0.1, aws4@1.6.0, uuid@3.0.1, qs@6.4.0, combined-stream@1.0.5, tough-cookie@2.3.2, form-data@2.1.4, mime-types@2.1.15, hawk@3.1.3, http-signature@1.1.1, har-validator@4.2.1)
└── yargs@4.3.2 (decamelize@1.2.0, camelcase@2.1.1, y18n@3.2.1, window-size@0.2.0, require-main-filename@1.0.1, lodash.assign@4.2.0, yargs-parser@2.4.1, cliui@3.2.0, os-locale@1.4.0, pkg-conf@1.1.3, string-width@1.0.2, read-pkg-up@1.0.1)
openaps installed
Traceback (most recent call last):
File "/usr/local/bin/openaps", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
IOError: [Errno 2] No such file or directory: '/root/src/openaps/bin/openaps'
any thoughts?
Dana Lewis
@danamlewis
May 08 2017 18:56
@bjarecki k next cat preferences.json for me?
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 18:56
@kenstack how do you do that? set the Edison time off the pump time whenever the Edison reboots
Chuck Vann
@djtechman
May 08 2017 18:56
I basically want to start all over with my Edison... but I can't figure out what's getting in my way
bjarecki
@bjarecki
May 08 2017 18:58
@danamlewis
        "max_iob": 30,
        "max_daily_safety_multiplier": 30,
        "current_basal_safety_multiplier": 40,
        "autosens_max": 4.0,
        "autosens_min": 0.7,
        "rewind_resets_autosens": false,
        "autosens_adjust_targets": true,
        "adv_target_adjustments": false,
        "maxCOB": 120,
        "override_high_target_with_low": false,
        "skip_neutral_temps": false,
        "bolussnooze_dia_divisor": 2,
        "min_5m_carbimpact": 5,
        "carbratio_adjustmentratio": 20,
        "autotune_isf_adjustmentFraction": 10,
        "remainingCarbsCap": 0,
        "enableUAM": true,
        "enableSMB_with_bolus": true,
        "enableSMB_with_COB": true,
        "enableSMB_with_temptarget": true
some are crazy at the moment as I thought I might be hitting a limit
Eric
@ecc1
May 08 2017 19:09
@mhaeberli Did you purge the ntp package? And do you have network connectivity when this happens? Any more information in syslog?
David
@davidkeddydb
May 08 2017 19:14
Screenshot 2017-05-08 15.12.42.png
So latest run of SMB, definitely better. Hopefully I am not really going to 21 mg/dl and if so, my autotune is way off with ISF :)
Dana Lewis
@danamlewis
May 08 2017 19:15
@davidkeddydb you should have more purple lines - did you grab floating carbs branch of NS?
David
@davidkeddydb
May 08 2017 19:16
@danamlewis No, I am on NS version 0.9.1-dev-20161112. I gather I need to update?
Dana Lewis
@danamlewis
May 08 2017 19:16
yep
David
@davidkeddydb
May 08 2017 19:16
:+1:
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 19:17
@ecc1 I got the same errors that @mhaeberli got when I tried that install some time ago. note I did purge the NTP package
Eric
@ecc1
May 08 2017 19:20
What does /etc/systemd/timesyncd.conf contain? Also, you can try running it from the command line with SYSTEMD_LOG_LEVEL=debug set in the environment.
Sorry this isn't working out-of-the-box like it did for me. Feel free to stick with NTP if debugging this is too much trouble.
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 19:21
I reflashed so I have nothing good at this pont
@ecc1 I appreciate your help very much but am going to see if I can find out how to get the edison to look for the time off of my pump like what @kenstack was saying.
fidoman
@efidoman
May 08 2017 19:23
Still getting the date issue ... The completely redone system worked fine until I realized that I didn't have --enable microbolus in the runagain script. Once I did that, I went back tot he date: invalid date issue ... this is where the log is stuck over and over. Any ideas?
Starting supermicrobolus pump-loop at Mon May 8 15:13:05 EDT 2017 with 26 second wait_for_silence:
Waiting up to 4 minutes for new BG: ........................Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK. Profile less than 60m old. pump_loop_completed less than 5m ago. RefreshRemoved duplicate bolus/carb entries:1
ed pumphistory
pumphistory.json: ^[[32m"Bolus 2017-05-08T14:42:01 head[4], body[0] op[0x01]"^[[0m
Checking pump clock: "2017-05-08T15:18:14-04:00" is within 1m of current time: Mon May 8 15:18:32 EDT 2017
and that pumphistory is less than 1m old. Temp refreshed
{"carbs":68,"boluses":3.8,"mealCOB":37,"currentDeviation":null,"maxDeviation":0,"minDeviationSlope":0}
BG data is too old (it's probably this), or clock set incorrectly. The last BG data was read at Mon May 08 2017 14:17:00 GMT-0400 (EDT) but your system time currently is Mon May 08 2017 15:18:45 GMT-0400 (EDT)
Checking deliverAt: null is within 1m of current time: Mon May 8 15:18:47 EDT 2017
date: invalid date ‘null’
Error, retrying
Waiting for 30s silence before mmtuning
Eric
@ecc1
May 08 2017 19:25
@diabetic_dude_twitter Sure. There's an "openaps use pump" command to get the time from the pump (I forget what, exactly), and then you can use that output with the "date" command
katie disimone
@Kdisimone
May 08 2017 19:45
@aabramowicz try rerunning this line. Sounds like it got missed during your edison setup
apt-get install -y sudo strace tcpdump screen acpid vim python-pip locate
Martin Haeberli
@mhaeberli
May 08 2017 19:47
@ecc1
#  under the terms of the GNU Lesser General Public License as published by
#  the Free Software Foundation; either version 2.1 of the License, or
#  (at your option) any later version.
#
# See timesyncd.conf(5) for details

[Time]
#Servers=0.debian.pool.ntp.org 1.debian.pool.ntp.org 2.debian.pool.ntp.org 3.debian.pool.ntp.org
so i should at least uncomment the server list!
no space left on device !
but 500 m free
David
@davidkeddydb
May 08 2017 19:51
@danamlewis Do I have to be on floating carbs or can I just use dev version of NS?
Dana Lewis
@danamlewis
May 08 2017 19:52
@davidkeddydb floating carbs
fidoman
@efidoman
May 08 2017 19:56
@danamlewis - I may try my other rig. Maybe I missed a step in updating to dev and configuring for supermicrobolus? Is there a documented list of steps to do that? Basically, I followed what the docs showed to update to dev, which seemed to work. Then I edited the runagain script to add --enable=' ... ... microbolus ...' and then editted preferences.json.
Dana Lewis
@danamlewis
May 08 2017 19:57
@ericfeibelman but did you run again with your flag? or just edit the script?
you have to run it again, not just edit the script.
Martin Haeberli
@mhaeberli
May 08 2017 20:00
@ecc1 i kinda gave up on timesyncd but if good advice comes in i’ll try it again
fidoman
@efidoman
May 08 2017 20:01
yes, I ran it again and it shows supermicrobolus in the log so it worked, but still the invalid date issue.
Dana Lewis
@danamlewis
May 08 2017 20:01
@bjarecki sorry - got pulled into work. Am wondering how you ended up with autosens multipliers if you’re not using? (Also trying to remember if anyone’s done thorough testing on autotune withOUT autosens running…) But, next steps will be to start poking around in files to see where the pump isf is; where if any autotune ISF is, and see if we can figure out where the snarl/misconnect is
live4sw
@live4sw
May 08 2017 20:06
I'm in this weird situation with my mobile rig in which after about an hour or two of working fine, it will start giving me messages about a time mismatch. I'll then turn it off and on and then it works fine again for an hour or two.
blob
Dana Lewis
@danamlewis
May 08 2017 20:08
@ericfeibelman does it keep working, and just show that in the logs? or worked and SMBed once and then stopped due to error?
live4sw
@live4sw
May 08 2017 20:08
Can't SSH to it while at the office, so sort of flying blind. But my other rigs (all set up using the same setup script) are working fine. Just curious if this is a known issue.
Dana Lewis
@danamlewis
May 08 2017 20:08
@live4sw may want to try the new clock reset stuff too?
Martin Haeberli
@mhaeberli
May 08 2017 20:08
@danamlewis @live4sw separate from SMB configuration, I’m having exactly the same problem
and fwiw pump and system clock are fine ...
fidoman
@efidoman
May 08 2017 20:09
Same for me
Martin Haeberli
@mhaeberli
May 08 2017 20:09
@live4sw are you running oref1 w/SMB? or oref0
live4sw
@live4sw
May 08 2017 20:09
oref0
Dana Lewis
@danamlewis
May 08 2017 20:09
Everyone with a nail: here’s my hammer suggestion: openaps/oref0#460
fidoman
@efidoman
May 08 2017 20:09
oh - I'm running oref1 w/SMB
live4sw
@live4sw
May 08 2017 20:10
but for some reason when i use the curl command to update openaps, the message it gives me at the end is 0.2.0-dev (even though it's the one meant to install master)
Dana Lewis
@danamlewis
May 08 2017 20:10
(nail = rig time mismatch, possibly @live4sw & others)
@live4sw that’s normal
live4sw
@live4sw
May 08 2017 20:11
It's weird, because when I use the curl command on my other two rigs, it will say 0.1.5 at the end.
Thanks for the hammer - how do we implement that, just run the curl command again? Or do we need to install dev?
Dana Lewis
@danamlewis
May 08 2017 20:12
it’s not in dev - it’s set-system-clock branch
Martin Haeberli
@mhaeberli
May 08 2017 20:13
and when this happens for me I have confirmed that the Nightscout curl api is getting good recent data
dhermanns
@dhermanns
May 08 2017 20:15

Hi! I'm still having problems setting up openaps. My pump-loop looks fine to me, but no bg values seem to be transferred to my nightscout cgm-remote-monitor. In my log I see:

{"temp":"absolute","bg":74,"tick":"+4","eventualBG":79,"snoozeBG":79,"predBGs":{"IOB":[74,75,76,76,77,77,78,78,79,79,79,79,79]},"IOB":0,"reason":"COB: undefined, Dev: 5, BGI: 0, ISF: 200, Target: 130; BG 74<90, but 37m left and 0 ~ req 0U/hr: no action required"}

So bg: 74 is what I would expect to be transferred to the remote-monitor. Should it?

Martin Haeberli
@mhaeberli
May 08 2017 20:15
here, when it’s ‘behaving’, glucose.json and ns-glucose.json get updated nicely:
cgm# ls -l
total 88
-rw-r--r-- 1 root root 44313 May  8 13:14 glucose.json
-rw-r--r-- 1 root root 44313 May  8 13:14 ns-glucose.json
but when it’s failing, the timestamp on glucose.json doesn’t advance (unless I run ns-loop manually, which fixes things for an iteration or so), and ns-glucose is empty
Dana Lewis
@danamlewis
May 08 2017 20:16
@dhermanns what kind of CGM are you using?
dhermanns
@dhermanns
May 08 2017 20:16
Medtronic Enlight.
Dana Lewis
@danamlewis
May 08 2017 20:16
are you seeing anything OpenAPS related in NS and just missing BGs? Or are you missing everything?
k. did you run openaps first-upload yet?
dhermanns
@dhermanns
May 08 2017 20:17
Nothing shows up in NS.
and if nothing shows up - check your NS url and api
probably a typo/incorrect
Martin Haeberli
@mhaeberli
May 08 2017 20:17
@live4sw @ericfeibelman I’m running CGM data from Dexcom G4 via Dexcom Share to Nightscout to openaps . you?
bjarecki
@bjarecki
May 08 2017 20:18
@danamlewis anywhere I grep for sensitivity looks like my pump setting, other than autotune.json.
live4sw
@live4sw
May 08 2017 20:18
@mhaeberli I'm using xDrip to get from Dexcom G4 w/share directly via bluetooth.
Martin Haeberli
@mhaeberli
May 08 2017 20:19
@live4sw so your openaps reads directly from share receiver, NOT via cloud … interesting
bjarecki
@bjarecki
May 08 2017 20:19
@danamlewis what files should reflect the autotune values. I suppose I can turn autosensativity on, just in the past I had better results without it
Dana Lewis
@danamlewis
May 08 2017 20:20
@bjarecki you shouldn’t have to turn it on (but if you’re up for it, turn it on and see if that fixes it; then you can turn it off); but it may be you found a nice edge case
dhermanns
@dhermanns
May 08 2017 20:20
pi@pi3t:~/myopenaps $ openaps first-upload
[Errno 2] No such file or directory
openaps: 'first-upload' is not an openaps command. See 'openaps --help'.
Hm - there seems to be something wrong with my setup ;-)
Dana Lewis
@danamlewis
May 08 2017 20:20
might change your multipliers back down while you turn that on temporarily first
live4sw
@live4sw
May 08 2017 20:20
@mhaeberli oh, xDrip uploads first to Nightscout and then Openaps gets it from there
I just like xDrip a lot more than the Nightscout uploader because I don't have to use the cable and the battery usage is way better
Martin Haeberli
@mhaeberli
May 08 2017 20:22
@live4sw I’m with you on that aspect - so you’re running xdrip physical gadget and an android phone xdrip to get the data to Nightscout
sounds like our downstream (from Nightscout) BG data flow is the same… I was looking for common elements in the failure mode.
live4sw
@live4sw
May 08 2017 20:25
Yep, and I have two other rigs that are working totally fine, so it's not on the Nightscout end.
katie disimone
@Kdisimone
May 08 2017 20:26
Is there some sort of internet glitch today? Another user (not on dev...totally vanilla OpenAPS setup) has rig time mismatch error message since 7:51am this morning. Reboot doesn't solve it.
bjarecki
@bjarecki
May 08 2017 20:27
@danamlewis so if rerun oref and turn it on. I would need to wait till the next day for the autotune setting to be picked up?
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 20:27
I think loads of people are having that issue today
Martin Haeberli
@mhaeberli
May 08 2017 20:27
just for grins, what happens if you cd ~/src/oref0; git log | head ?
in my case, I get
commit 5e55ecec2eda2db5260d72c366f9b3c49d86c693
Author: Scott <scottleibrand@gmail.com>
Date:   Sat Mar 11 16:27:58 2017 -0800

    0.4.2
Dana Lewis
@danamlewis
May 08 2017 20:27
@bjarecki nah. it should still have your autotune and use that, but I want to see if adding autosens gets it to properly pull in a non-12 ISF
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 20:27
@Kdisimone like 3 I have seen including me
Martin Haeberli
@mhaeberli
May 08 2017 20:27
and git branch -a shows I’m on master
Dana Lewis
@danamlewis
May 08 2017 20:27
so turn it on; watch the logs for half an hour or until you see if ISF is changed and used; then you can turn it off. this is mostly for troubleshooting. @bjarecki
Martin Haeberli
@mhaeberli
May 08 2017 20:28
@live4sw and when you get a chance, try the same on the others - just confirming which rev of oref0 is running
@danamlewis fwiw it looks like at least some of us are having these BG values out of date issue on oref0 thus not likely beneficiaries of the new improved @scottleibrand hammer
jmcrawford
@jmcrawford
May 08 2017 20:31
@diabetic_dude_twitter @Kdisimone I don't have access to a rig right now, what servers are configured in ntp.conf by default (what most people will be using)? Could start with checking there
dhermanns
@dhermanns
May 08 2017 20:31
pi@pi3t:~/myopenaps $ openaps first-upload
[Errno 2] No such file or directory
openaps: 'first-upload' is not an openaps command. See 'openaps --help'.
@all: Any idea what's missing in my setup?
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 20:32
@jmcrawford 0.debian.pool.ntp.org and it pings just fine from my rig
katie disimone
@Kdisimone
May 08 2017 20:33
@jmcrawford can't get to her rig right now (kid is at school)
Martin Haeberli
@mhaeberli
May 08 2017 20:34
@Kdisimone (a theoretical argument for reverse ssh tunnel into rigs, but SAFETY and SECURITY, so maybe not)
live4sw
@live4sw
May 08 2017 20:34
@mhaeberli unfortunately not on SSH until tonight but I will try cd ~/src/oref0; git log | head and let you know then
so weird that we are all having this time problem now
Martin Haeberli
@mhaeberli
May 08 2017 20:36
@live4sw agree - also not clear that this time problem is always exactly the same problem ...
live4sw
@live4sw
May 08 2017 20:36
I am sort of tempted to just add a cron command to reboot the rig every 30 minutes or so but i know that is just a bad bandaid
Martin Haeberli
@mhaeberli
May 08 2017 20:36
and for me reboot didn’t necessarily fix it ...
live4sw
@live4sw
May 08 2017 20:36
So I will not do that
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 20:36
@mhaeberli I agree
jmcrawford
@jmcrawford
May 08 2017 20:38
I can't see anything indicating an issue with debian's ntp servers.
dhermanns
@dhermanns
May 08 2017 20:38
Wow - cool. Found it. Openaps first-upload seemed to have fixed it. Works like a charme now :) . Thx!!
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 20:38
does anyone in the room know how to set the edison to get the time from your pump? I am looking for that as a solution for me at this point
live4sw
@live4sw
May 08 2017 20:40
I should also note that I am still using pi's
Martin Haeberli
@mhaeberli
May 08 2017 20:40
@live4sw in all cases or in 2 out of 3 rigs?
here I’m using Edisons
live4sw
@live4sw
May 08 2017 20:41
All 3 rigs
Martin Haeberli
@mhaeberli
May 08 2017 20:41
ok, so in at least one case, Edison vs Pi works (breaks) the same
live4sw
@live4sw
May 08 2017 20:42
I have my edison/explorer board stuff coming this week. I previously used a manually-flashed ubilinux edison with the TI stick but the reliability was really poor so I switched back to Pi's
Martin Haeberli
@mhaeberli
May 08 2017 20:42
explorer FTW!
Diabetic_Dude
@diabetic_dude_twitter
May 08 2017 20:45
lol
bjarecki
@bjarecki
May 08 2017 21:53
@danamlewis well I ran for an hour or so with --enable='autosens autotune ' and my ISF is still reflecting my pump value and not my autotune values.
Dana Lewis
@danamlewis
May 08 2017 21:54
@bjarecki one more try: change autotune_isf_adjustmentFraction back to 0.5
bjarecki
@bjarecki
May 08 2017 21:55
@danamlewis let me give it a shot
Dana Lewis
@danamlewis
May 08 2017 22:05
FYI all - I just created this issue (openaps/oref0#461) for SMB/UAM related problem reports and troubleshooting during the testing process to help keep the PR less cluttered. I’d ask here first; but if you’re not getting answers by the next day (i.e. your question/report gets buried), that issue (https://github.com/openaps/oref0/issues/461) is a good place to also post your question or issue so we can keep track of it and any related resolutions.
bjarecki
@bjarecki
May 08 2017 22:23
@danamlewis well looks like changing autotune_isf_adjustmentFraction back did not help.
Dana Lewis
@danamlewis
May 08 2017 22:24
ok. I’m out of thoughts; want to create an issue explaining the situation, your setup, and what you’ve tried? That way we can noodle on it some more, and others may be able to suggest other things to test next
bjarecki
@bjarecki
May 08 2017 22:26
@danamlewis sure. Thanks for the help
live4sw
@live4sw
May 08 2017 23:27
@mhaeberli just got home and checked, I'm on 0.4.2 as well
Martin Haeberli
@mhaeberli
May 08 2017 23:31
@live4sw ok, thx 4 update - lets keep on hunting for the bug, if it is one ...
jcorbett80
@jcorbett80
May 08 2017 23:35
Andrew M I been watching. For your time issue that is gone after reboot. I changed to Heroku from azure. From that point on I would get that same message after a certain amount of time. The time was the date I moved from azure to Heroku. I went into Heroku and redeployed. I have never seen it again in the past several days. That may not be your exact issue but if you have Heroku that may be your fix.
live4sw
@live4sw
May 08 2017 23:36
@jcorbett80 unfortunately i've been on heroku for months now!
but maybe i should check and perhaps update my nightscout version
Martin Haeberli
@mhaeberli
May 08 2017 23:37
@live4sw ok let me pile on here and ask - how do I easily update Heroku Nightscout version without dropping and rebuilding the Heroku instance?
live4sw
@live4sw
May 08 2017 23:39
@mhaeberli I believe if you have automatic deploys enabled, and then update your github repo by comparing and then accepting changes, it will then automatically deploy to your heroku site.
What version is your Nightscout site?
Mine is 0.9.3-dev-20161224, I'm probably overdue for an upgrade
Martin Haeberli
@mhaeberli
May 08 2017 23:40
0.10.0 dev on one, 0.9.2 on the other, but nota bene - i foolishly Heroku deployed directly from public github master rather than my own fork - long story - so I may be hosed? or maybe just a bunch of RTFM to figure out how to change the upstream github URL in Heroku (maybe from CLI)
katie disimone
@Kdisimone
May 08 2017 23:41
@mhaeberli not hosed. There's a screenshot on bottom of this page that shows where to connect your GitHub cgm-remote-monitor repo and select branch. One sec
Martin Haeberli
@mhaeberli
May 08 2017 23:42
@Kdisimone thx in advance
katie disimone
@Kdisimone
May 08 2017 23:43
Martin Haeberli
@mhaeberli
May 08 2017 23:44
yay thx -i’ll probably try to re-deploy later tonight from my own fork...
katie disimone
@Kdisimone
May 08 2017 23:45
Looks like you don't have a fork of the cgm-remote-monitor repo yet from Nightscout. Just fork that into your GitHub and then follow that image above to connect to your repo. You won't have to do anything else. It will update when you deploy from your repo because it's a fresh fork (and newer version than your existing deployment)
Martin Haeberli
@mhaeberli
May 08 2017 23:47
@Kdisimone thx - I have had forks in the past, but don’t now - one of my problems (more more RTFM) is that I would actually like to have multiple forks of public cgm-remote-monitor, say, with different names, so that I can try different branches (ok, different branches would just be cleaner, but you get my drift) but github won’t let me...
live4sw
@live4sw
May 08 2017 23:48
@mhaeberli were you having the problem with both nightscout sites, or only the older one? I just upgraded to NS 0.10-dev (the newest) and am sort of hoping that fixes things.
Martin Haeberli
@mhaeberli
May 08 2017 23:49
having problem with both - but I’ll give that a try nonetheless. problem has not re-appeared since sometime earlier today - even though dexcom share feed dropped a few times - i’ll keep an eye on it.
live4sw
@live4sw
May 08 2017 23:49
I had it happen a few hours ago but a restart fixed it, going to see how things go tomorrow.
katie disimone
@Kdisimone
May 08 2017 23:50
@mhaeberli I just create new branches myself in my own repo when I want to try something new. Like I prefer a different preset for the activity temp targets than the standard (120-140 for 2 hours). So I created a branch where I modded that and use that branch for my NS
Martin Haeberli
@mhaeberli
May 08 2017 23:50
@live4sw :+1: @Kdisimone thx