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

29th
Sep 2017
jaylagorio
@jaylagorio
Sep 29 2017 00:03
Zero temp'd all day even if bgs are fine makes you feel bad? I don't think being zero temp'd for a really long time has happened to me but I can't say I would have predicted that
Scott Leibrand
@scottleibrand
Sep 29 2017 00:04
lack of insulin causes ketones even if you're using up all the extra glucose on activity
Dana Lewis
@danamlewis
Sep 29 2017 00:04
^ that.
Jacob H
@jdhigh
Sep 29 2017 00:04
^
jaylagorio
@jaylagorio
Sep 29 2017 00:05
Bodies are stupid.
garykidd
@garykidd
Sep 29 2017 00:05
Not sure what I've done incorrectly but I sporadically am getting p/w related errors in my dexcom g5 mobile app...it states that someone has changed my password and asks me to re-enter new password.....so being that I didn't change my password--I enter my last known password...and it works...WTF? I called dexcom tech support and they have no idea.....does anyone know here?
Scott Leibrand
@scottleibrand
Sep 29 2017 00:05
might be the same mechanism that can cause euglycemic DKA in extreme cases (with certain glucose-lowering medications etc.)
Phil
@drnoname82_twitter
Sep 29 2017 00:07
Ok, so next time we will set an even higher temp target before activity. In very strenuous hikes our management is similar to @danamlewis - we also proactively eat in order to have some insulin on board.
Dana Lewis
@danamlewis
Sep 29 2017 00:07
@jaylagorio pre-looping I would zero temp manually, or low temp, for many hours. Which is when I figured this out. Only this year started figuring out good food to eat proactively => a little more insulin going in the hike => good BG outcomes but better energy during and not feeling icky halfway through afterward.
the other thing I try to do is in addition to activity mode prior, is eat right at start of hike. Ie if we drive Saturday morning, eat breakfast sandwich driving up to trailhead. Do a tiny bolus (or maybe no bolus now? Tbd) and let temps and loop take care of modulating rest and also the food being taken care of by first hour or so of hike.
Not perfect, but also helpful when I can vs eating at home hour or two prior and having iob on. Causes less tradeoff of choosing much higher and unpredictable starting bg.
/me has a feeling @scottleibrand will want to go hiking this weekend now ;)
Scott Leibrand
@scottleibrand
Sep 29 2017 00:10
hehe
jaylagorio
@jaylagorio
Sep 29 2017 00:16
My statement stands!
Phil
@drnoname82_twitter
Sep 29 2017 00:25
😁
Dana Lewis
@danamlewis
Sep 29 2017 00:29
@jaylagorio the same one from the other day? ;p
WeedsDIY
@WeedsDIY
Sep 29 2017 00:35
Having a problem with comms between Dexcom Clarity and Nightscout. It appears the bridge is broken and we don't know how to fix it. Prior Dexcom to verify transmitter was not malfunctioning and it is good. They advised uninstalling the G5 app, restarting the phone, then re-installing. This was done, but did not resolve issue.
Prior to the broken bridge problem, there was a problem with transmitter, which we call dexcom about. They advised the uninstall, reboot, reinstall to solve that problem. Fixed. But then the bridge seemed to be broken.
Is there a repository of experienced problems where a looper can search for their current problem and sort it out from the experiences of others?
Jacob H
@jdhigh
Sep 29 2017 00:41
@WeedsDIY https://openaps.readthedocs.io/en/latest/docs/Troubleshooting/oref0-setup-troubleshooting.html The troubleshooting section has some, but it's by not comprehensive.
WeedsDIY
@WeedsDIY
Sep 29 2017 00:44
Yea been there and as you mention, not very comprehensive. Didn't contain above problem, nor others we have had. Keep coming back here.
The rig keeps trying to pull in new bg data, then displays "BG data too old". I can view current bgs on the G5 mobile app,but it's not pulling into NS. This is why we suspect a problem with the Dexcom bridge. Don't know how to fix it. Any advice much appreciated.
WeedsDIY
@WeedsDIY
Sep 29 2017 01:03
Just found solution here http://www.nightscout.info/wiki/welcome/nightscout-with-ios-and-dexcom-share/nightscout-bridge-for-dexcom-g5 Thanks @jdhigh for the input. Apparently after reinstalling G5 mobile app, I overlooked turning the Share feature ON. Gah! I'm sure that won't be my last dumb question, lol. Looping again - yay! : )
Jacob H
@jdhigh
Sep 29 2017 01:06
@WeedsDIY I'm not familiar enough with that problem, but do you think the docs could be updated with PR?
WeedsDIY
@WeedsDIY
Sep 29 2017 01:10
Good idea. Will submit. Simple fix, but no looping til you find it. Might as well help the next newbie like me.
jaylagorio
@jaylagorio
Sep 29 2017 01:14
@danamlewis Whatever day it was when I said bodies are stupid!
garykidd
@garykidd
Sep 29 2017 01:14
@WeedsDIY I have that problem periodically...mine may be a bit different...I get he p/w reset error inthe G5 mobile app on my phone....same end result as you though BGs dont hit NS and it kills the loop...anyone else? Let us know
Jacob H
@jdhigh
Sep 29 2017 01:25
I have a general question... would redundant rigs help avoid these sorts of hangups? Like if one rig flakes out, at least the other is still looping? (Unless it's a pump/low bat problem, etc.)
PopcornPam
@PopcornPam_twitter
Sep 29 2017 01:55
Hi all, I keep running up against maxsafebasalrate on a pretty routine basis, I have already increased to 4x max, 4x current, and still running into this issue. The docs suggest you shouldn't adjust either of these, so wanted to double check before bumping them again. my basal rates are pretty low compared to my carb ratio.. autotune thinks my rates are all ok, and this issue is only when I eat carbs
Dana Lewis
@danamlewis
Sep 29 2017 01:57
A lot of experienced people have cranked it up. So you can nudge it up further.
@jdhigh yep. Redundancy is awesome. Also so is installing nogit moving forward ;)
PopcornPam
@PopcornPam_twitter
Sep 29 2017 01:57
ok.. also, so glad you guys mentioned the thing about loop causing double carb entries... zero lows since that night!
Dana Lewis
@danamlewis
Sep 29 2017 01:58
:clap:
And you survived the weekend ;) also good! ;D
PopcornPam
@PopcornPam_twitter
Sep 29 2017 01:58
Such a fun weekend! And super nice to see you guys
And outstanding recommendation with Monorail coffee.. Their Tyrell latte (rose latte) was pretty much heaven
Dana Lewis
@danamlewis
Sep 29 2017 01:59
:+1: good to know it's actually good!
PopcornPam
@PopcornPam_twitter
Sep 29 2017 02:19
uh oh... all of a sudden problems!
Old settings refresh at Module.load (module.js:356:32)
ed. pump_loop_completed more than 10m ago: setting temp before refreshing pumphistory.
Temp refresh at Module.load (module.js:356:32)
ed
Could not parse input data: [SyntaxError: /root/myopenaps/monitor/iob.json: Unexpected end of input]
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
SMB pump-loop failed. pump_loop_completed more than 15m old; waiting for 40s silence before mmtuning
Radio ok. Listening: .
Jacob H
@jdhigh
Sep 29 2017 02:20
Another idea I thought of... since many of my rig's problems seemed to be solved by a reboot, has anyone thought about inserting into the software a reboot command if no looping after xx minutes?
@PopcornPam_twitter try reboot on that error.
PopcornPam
@PopcornPam_twitter
Sep 29 2017 02:22
k
Scott Leibrand
@scottleibrand
Sep 29 2017 03:46
@jdhigh the problems fixed by reboots generally do trigger automatic reboots. however, we don't want to reboot just because a rig is not looping, as that would result in plugged-in rigs rebooting repeatedly while the PWD is out of the house, increasing the chances they would lose connectivity or otherwise be unable to loop when the pump got home
sadafakahmad
@sadafakahmad
Sep 29 2017 03:48
Just another request for advice. Two problems preventing me from looping outside the house with G5:
1) rig won't take information from Dexcom receiver when connected to it directly by USB
2)when rig is tethered to iPhone hotspot, Dexcom Share says internet connection not working and so BG values not taken in by NS.
0.6.0 is unbelievable, and I'm not even using SMB. Unfortunately I can't use openaps outside the house like everyone seems to be able to.
Scott Leibrand
@scottleibrand
Sep 29 2017 04:04
what kind of microUSB OTG cable are you using? what port is it plugged into? does the charging icon turn on when you plug it in (with an external 5V USB battery on the other port)? does lsusb show a new device (the Dexcom, but without a name) when you plug it in?
sadafakahmad
@sadafakahmad
Sep 29 2017 04:14
@scottleibrand: Using the BestGame Setups microUSB cable that was linked to in the docs.
Plugged into OTG port.
The charging icon turns on in the Dexcom receiver. Do I have to type in "lsusb" in Terminal?
If so, how do I do that when both ports in rig are being used up, and therefore can't connect to computer?
Scott Leibrand
@scottleibrand
Sep 29 2017 04:26
your computer can serve as the 5V USB power source
or you can do it over SSH
sadafakahmad
@sadafakahmad
Sep 29 2017 04:28
@scottleibrand : Is this correct?: root@sadafakahmad:~# lsusb
unable to initialize libusb: -99
Scott Leibrand
@scottleibrand
Sep 29 2017 04:31
that's the correct command, and a -99 there means your OTG cable/port is not working properly
try reversing the direction?
sadafakahmad
@sadafakahmad
Sep 29 2017 04:33
@scottleibrand: switched the two ends of the OTG cable and reconnected and got same -99 message.
Scott Leibrand
@scottleibrand
Sep 29 2017 04:34
which model of explorer board?
sadafakahmad
@sadafakahmad
Sep 29 2017 04:34
@scottleibrand: v1.3
sadafakahmad
@sadafakahmad
Sep 29 2017 04:41
BTW, this is another problem I have when I try to log into rig via SSH: Saad-Ahmads-MBP:~ saadahmad$ ssh root@sadfakahmad.local
ssh: Could not resolve hostname sadfakahmad.local: nodename nor servname provided, or not known
model of explorer board is v 1.3, GMD-2017
Scott Leibrand
@scottleibrand
Sep 29 2017 04:45
you may want to try getting the IP address off the rig over console and then ssh to that instead of the name
that EB should do OTG fine. do you have any other hardware to swap out and see if you can isolate which piece is having the problem?
sadafakahmad
@sadafakahmad
Sep 29 2017 04:48
figured out the ssh issue: had to type: ssh root@sadafakahmad and not ssh root@sadafakahmad.local
only one rig and only one receiver and only one otg cable, unfortunately.
Maybe a Jubilinux issue? running 0.2.0. Been thinking about flashing for a few days especially.
Scott Leibrand
@scottleibrand
Sep 29 2017 04:53
I don't know of any issues with 0.2.0 and OTG, but a reflash wouldn't hurt if you have the time to get everything set back up again
or you can start ordering spare parts to test things out (and eventually build another rig)
Tim Street
@tim2000s
Sep 29 2017 12:56
@popcornpam that issue is usually caused by a missing comma in preferences when setting up the exponential curves.
cameronrenwick
@cameronrenwick
Sep 29 2017 13:12
I'm updating one of my rigs and it has stalled right at the end. The last line of code in terminal is first-upload ! bash -c "cat monitor/glucose.json | json 1 > nightscout/recent-missing-entries.json && openaps report invoke nightscout/uploaded-entries.json" does this generally take a while or??
Marco
@CaptainBalou
Sep 29 2017 13:19
grafik.png

I have a question concerning eSMB.

maxBolus = round( profile.current_basal * profile.maxSMBBasalMinutes / 60 ,1);

I understood that maxSMBBasalMinutes is a multiplier to calulate maxBolus. If I have a basal profile value of 0.9U and a maxSMBBasalMinutes: 300 I should end at maxBolus=4.5U. Is that correct?

Assuming I am correct - can someone explain me the two screenshots? What is limiting maxBolus?

grafik.png
Sorry for the screenshot order. So please notice that the second screenshot is before my question. ;-)
...above my question...
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 13:24
What I understood is that maxSMBBasalMinutes covers the max dosagte for SMB
the minutes is the max minutes it will take to recover for the correction made by suspending
live4sw
@live4sw
Sep 29 2017 13:26
Has anyone had trouble finding an external battery pack that will continue to charge the Edison's LiPo? One of my home rigs isn't connected to an outlet, so I periodically swap external battery packs, but for some reason the Edison stops charging from the external power pretty quickly and then just runs the LiPo down to zero over time. Only one of my external power banks seems to still stay on and charge it.
jaylagorio
@jaylagorio
Sep 29 2017 13:29
@live4sw I use this one if it helps: https://www.amazon.com/gp/product/B0194WDVHI/
Goes 24 hours before dying with a G5 receiver attached, I'm sure it'll go much longer if it's just the Edison
My only complaint is that I can't charge it and power the Edison at the same time, but the lipo takes care of that time period
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 13:36
ls
live4sw
@live4sw
Sep 29 2017 13:38
@jaylagorio Just to make sure, can you have the Edison on with the Lipo attached and that battery charging the Lipo?
jaylagorio
@jaylagorio
Sep 29 2017 13:40
Yes, when you're not charging the USB battery the USB battery both charges the lipo and powers the Edison. But when you're charging the USB battery it neither charges the lipo nor powers the Edison.
I'm still in search of a USB battery that I can charge while it passes power through to the Edison while it charges.
Marco
@CaptainBalou
Sep 29 2017 13:43

the minutes is the max minutes it will take to recover for the correction made by suspending

Aha. So a complete other understanding of what it means and does. ;-)

Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 13:48
@CaptainBalou

Enhanced SMB (eSMB)

The default behaviour for SMB is that the max bolus that can be delivered is no greater than 30 mins of basal insulin. An additional preferences value, "maxSMBBasalMinutes", has been added to allow SMB to deliver a different amount of insulin as an SMB. This gives the ability to make SMB more aggressive if you choose. As with standard SMB, it is triggered by temp targets, carb entries and a pump bolus.

Increasing "maxSMBBasalMinutes" will allow the SMB functionality to deliver more insulin, earlier in the SMB process.

"maxSMBBasalMinutes" can be added into myopenaps/preferences.json. It is recommended that the value is set to start at 30, in line with the default, and if you choose to increase this value, do so in no more than 15 minute increments, keeping a close eye on the effects of the changes.

It is not recommended to set this value higher than 90 mins, as this may affect the ability for the algorithm to safely zero temp. It is also recommended that pushover is used when setting the value to be greater than default, so that alerts are generated for any predicted lows or highs.

Marco
@CaptainBalou
Sep 29 2017 14:00
@diabetic_dude_twitter This is what I read as well. ;-)
ejoe132
@ejoe132
Sep 29 2017 14:01
Having an issue with my edison, stopped working on me so did a fresh install ran great for two days, now it stopped operating again but I cant get into it either way with putty. any tips?
Marco
@CaptainBalou
Sep 29 2017 14:05
@diabetic_dude_twitter And so I would say I understood it right. But nevertheless I would expect another result for maxBolus then.
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 14:06
unless there is something for safety hard coded somewhere that is not mentioned
gregorschlump
@gregorschlump
Sep 29 2017 14:11
Hey folks. I'd like to hear, what you think about Medtronic 640G with SmartGuard vs. OpenAPS. What are the strengths? what are the weaknesses? Thx a lot...
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 14:12
@CaptainBalou This could be a factor openaps/oref0#639 "don't allow maxSMBBasalMinutes > 30 unless IOB (and COB) are > 0 #639"
ejoe132
@ejoe132
Sep 29 2017 14:14
@gregorschlump From what Ive read and talked to a nurse I review numbers with about the 670g its nowhere near as elaborate. One the medtronic sensors even the new ones are not as accurate as dexcom and two the 670 locks you into a target of 120 and doesnt offer: SMB, AUTOSENSE, AUTOTUNE, etc.
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 14:17
@gregorschlump I like faster updates and personal control of what my pump does. I like to ability to modify the software to specifically meet my personal diabetes needs. That is why I choose openaps. At this point I can get any pump I want and be covered by insurance but I choose what I have because it works so well.
gregorschlump
@gregorschlump
Sep 29 2017 14:20
@ejoe132 @diabetic_dude_twitter fair points. Thank you... Especially SMB and AUTOTUNE seem to be quite nice functionalities.
cameronrenwick
@cameronrenwick
Sep 29 2017 14:29
anyone got a recipe that the can share for carb entry on IFTTT? I've tried to make one and it doesn't seem to be working. Secret etc is all good...just not sure what I'm not doing.
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 14:59
@cameronrenwick how do you know it is not working? I made one for beer last week, and also did not think it was working until I checked my NS site. My notifications are not working but the entry is
I had 6 entries on my NS site that I missed all for beer.
(I cant drink that much beer )
This is what I used (remember notifications are not working for me)
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 15:05
This is what I put in the "BODY" section
{"enteredBy": "IFTTT-button", "reason": "Beer", "carbs": 20, "secret": "ILL NEVER TELL ITS A SECRET"}
cameronrenwick
@cameronrenwick
Sep 29 2017 15:30
thanks...I think I must simply have a type... I'll recheck!
dougtakeuti
@dougtakeuti
Sep 29 2017 16:21
Hi there, the loop stopped working this morning after a calibration error. There seems to be an issue with Old pumphistory-24h, as shown below. Any idea what the refreshKeyError indicates?
Starting supermicrobolus pump-loop at Fri Sep 29 13:14:02 BRT 2017 with 7 second wait_for_silence:
MDT CGM configured; not waiting
Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK.
Attempting to retrieve MDT CGM data from pump
MDT CGM data retrieved
MDT New cgm data reformatted
Old pumphistory-24h, waiting for 7 seconds of silence: Radio ok. Listening: .No pump comms detected from other rigs
Old pumphistory-24h refreshKeyError: 2
SMB pump-loop failed. grep: monitor/temp_basal.json: No such file or directory
Waiting for 30s silence before mmtuning
Radio ok. Listening: .No pump comms detected from other rigs
Listening for 30s silence before mmtuning: .
Dana Lewis
@danamlewis
Sep 29 2017 16:34
@gregorschlump note - you asked about 640 and people responded as if you asked about 670. The 640 only reduces basals. The 670 gives more and less insulin. So there is an even greater difference in the 640 (basal reduction only) and OpenAPS (gives more and less insulin as needed, plus everything else already stated)
@cameronrenwick paste here sans secret? The other day someone missed a {
David
@davidkeddydb
Sep 29 2017 16:40
If I am switching one of my rigs from an explorer board to using a TI stick, I just need to change the pump.ini file to reflect using /dev/ttyACM0 in my case as the port, or is there more to be done than that?
Dana Lewis
@danamlewis
Sep 29 2017 16:43
Make sure that's actually the port, too? And not just the example from the docs. You could try that, but likely you may need to run setup from scratch
There may be some other adjustments needed but not sure what they are.
Scott Leibrand
@scottleibrand
Sep 29 2017 16:44
@CaptainBalou you're abusing maxBolusSMBMinutes if you set it that high. I would change it back to 45m and observe it in action before raising it any further.
If you enter carbs and underbolus, you should see it kick in with a higher maxBolus than before.
cameronrenwick
@cameronrenwick
Sep 29 2017 16:46
thanks @danamlewis yes.. typos are show stoppers...new problem now tho... can't seem to figure out. Just updated to 0.6.0 using non git method. This rig previously had range of ~20 feet. Now it will only connect to my pump about every 5th or 6th loop. Is there something new with 0.6.0 that would do this? EB and edison are connected properly etc..
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 16:58
root
damnit!
David
@david-t1
Sep 29 2017 17:02
Just throwing this out there but before I give up on using my 515 pump for openaps, does anyone have any thoughts on somehow repairing the pump antenna? It is damaged/defective because I have never gotten good signal strength using my rpi2 or Edison rig. I used to connect mabye 50% of the time but now it is always -99.
Marco
@CaptainBalou
Sep 29 2017 17:05
@scottleibrand Abuse...hmm... I don't know why but my basal rates are super low and I pass every basal rate test without any issue. But my bolus needs to be very intense. So I normally not enough by SMB. That's why I hacked the code in 0.5.3. - we had a short chat some days ago together concerning that. So I don't think I'll use this way to get more units during SMB. 😂
Scott Leibrand
@scottleibrand
Sep 29 2017 17:26

@CaptainBalou I was just referring to your "300 minutes" comment. I think you would do well with a higher than typical maxSMBBasalMinutes value, but need to get there incrementally while observing the behavior, not jump straight to a crazy-high value.

are you trying to replace manual basals, or augment them? eSMB is designed to give several relatively large basals in sequence (it still has "micro" in the name), not replace a single meal bolus with a single automated bolus.

ejoe132
@ejoe132
Sep 29 2017 17:54
Having an issue with my edison, stopped working on me so did a fresh install ran great for two days, now it stopped operating again but I cant get into it either way with putty. any tips?
Marco
@CaptainBalou
Sep 29 2017 18:08
@scottleibrand Sure Scott, I understood. I wasn’t going strait to 300 - I started at a lower level and increased it. But after don’t seeing anything happen I tried a freaky value just to have a drastic example. :-) So does that mean there is a mximum value after which I get set back to the default or so? I just need to get something like „maxBolus = 4,5 * basalofcurrenthour“. This is what I had in the version before. That worked really good for me. Otherwise I try to handle a Pizza and a drink which is huge with a eSMB of 0,something per injection. That takes too long. I would prefer to have something like 1,5U to get corrected faster if the algorythm says I need as much insulin which I normally do with such a meal. But I don’t know if I can get this of eSMB with my some kind of strange difference between basal rate and bolus. I don’t why it’s like that inside my body… :-(
cameronrenwick
@cameronrenwick
Sep 29 2017 18:20
anyone have any advice here? Just updated to 0.6.0 (twice - 1st time the no-git way; 2nd time the git way) can't seem to get looping

```ValueError: No JSON object could be decoded
SMB pump-loop failed. pump_loop_completed more than 15m old; waiting for 40s silence before mmtuning
Radio ok. Listening: .No pump comms detected from other rigs
Listening for 40s silence before mmtuning: .No pump comms detected from other rigs
mmtune: "916.636", 5, -78 waiting for 36 second silence before continuing
Radio ok. Listening: .No pump comms detected from other rigs
Done waiting for rigs with better signal.
Unsuccessful supermicrobolus pump-loop at Fri Sep 29 14:12:16 EDT 2017

Starting supermicrobolus pump-loop at Fri Sep 29 14:13:02 EDT 2017 with 14 second wait_for_silence:
Waiting up to 4 minutes for new BG: ........................Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK. Old settings refresh at Module.load (module.js:356:32)
ed. Temp duration >20m. Refresh at Module.load (module.js:356:32)
ed pumphistoryCould not parse input data: [SyntaxError: Unexpected end of input]
and meal.json
Checking pump clock: "2017-09-29T14:17:54-04:00" is within 1m of current time: Fri Sep 29 14:18:28 EDT 2017
and that pumphistory is less than 1m old. Temp refresh at Module.load (module.js:356:32)
ed
Could not parse input data: [SyntaxError: /root/myopenaps/monitor/iob.json: Unexpected end of input]
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
SMB pump-loop failed. Unsuccessful supermicrobolus pump-loop at Fri Sep 29 14:18:38 EDT 2017```

Scott Leibrand
@scottleibrand
Sep 29 2017 18:21
Time to drill down and figure out what is wrong with your iob.json (and probably profile.json)
Marco
@CaptainBalou
Sep 29 2017 18:22
I was facing a problem which could have been the same. job.json was empty and the error message the same. I searched a lot and some helped me out. After 2 hours they saw I had a „cut and paste“ error in preferences.json. Possibly you can investigate if you have the same before digging further into it. ;-)
Scott Leibrand
@scottleibrand
Sep 29 2017 18:24
@CaptainBalou if you go too high it will just cap it at the max, not reset to default. Are you trying to no-bolus, or get SMB to assist a manual bolus?
Marco
@CaptainBalou
Sep 29 2017 18:26
@scottleibrand Oh sorry, forgot to answer that: I would like to try a no-bolus scenario. So I bolus 1/5 at the beginning but then I try let it go.
Scott Leibrand
@scottleibrand
Sep 29 2017 18:26
With Fiasp or ?
Marco
@CaptainBalou
Sep 29 2017 18:26
Yes.
Scott Leibrand
@scottleibrand
Sep 29 2017 18:41
k. what is your hourly basal, and what is the meal amount and CR?
Marco
@CaptainBalou
Sep 29 2017 18:46
I wanted to answer with screenshots. But I can’t paste my clipboard. Do you know how I can paste a screenshot out of my clipboard in macOS with the gitter App?
Scott Leibrand
@scottleibrand
Sep 29 2017 18:47
I always drag and drop the actual file (which gets saved to my Screenshots folder any time I clip something)
Marco
@CaptainBalou
Sep 29 2017 18:48
Ok, other keys to hold I think. I try that...
Scott Leibrand
@scottleibrand
Sep 29 2017 18:49
I use command-shift-4 and then drag and drop a rectangle around what I want to clip
Marco
@CaptainBalou
Sep 29 2017 18:49
Bildschirmfoto 2017-09-29 um 20.49.27.png
Bildschirmfoto 2017-09-29 um 20.50.17.png
So that’s what you asked for I hope. ;-)
Scott Leibrand
@scottleibrand
Sep 29 2017 18:51
those are some pretty drastic basal swings. have you started running the latest autotune from the dev branch (as of about a week ago)? it smooths out daytime basals on days when there's no data on which to move them up or down (to be in between the hours before/afterward that do have tuning data)
and how many g do you enter for the pizza?
Marco
@CaptainBalou
Sep 29 2017 18:53
No haven’t tried it yet. Is that a special branch or inside dev already?
Scott Leibrand
@scottleibrand
Sep 29 2017 18:53
it's been in dev for about a week
Marco
@CaptainBalou
Sep 29 2017 18:53
I had a wonderful pizzy these days. Was round about 70g of carbs plus a wonderful half a liter of beer with let’s say 30g. :-D Delicious. ;-)
cameronrenwick
@cameronrenwick
Sep 29 2017 18:56
should the curve chosen have quotes around it? as in "curve": "ultra-rapid",
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 18:57
yes
cameronrenwick
@cameronrenwick
Sep 29 2017 18:57
odd eh as all the other word values (true/false) DON'T have quotes
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 18:57
"curve": "ultra-rapid",
Scott Leibrand
@scottleibrand
Sep 29 2017 18:57
@cameronrenwick quotes go around strings. true/false are boolean
cameronrenwick
@cameronrenwick
Sep 29 2017 18:58
ya I was just dusting off 20yrs ago knowledge.. that's been my prob... sigh
Dana Lewis
@danamlewis
Sep 29 2017 18:59
also where is it in your list? last one on list = no comma, but if you've got it higher up, yes comma
cameronrenwick
@cameronrenwick
Sep 29 2017 18:59
yup... got that part of the syntax correct..
thx!
Diabetic_Dude
@diabetic_dude_twitter
Sep 29 2017 19:01
Unsuccessful supermicrobolus pump-loop at Fri Dec 31 18:36:57 CST 1999
:(
sdneufer
@sdneufer
Sep 29 2017 19:01
@danamlewis Do you still use ThisButton on the Pebble watch?
Scott Leibrand
@scottleibrand
Sep 29 2017 19:02
@CaptainBalou ok, so if you enter 100g with a 1U/h basal and a 15g CR, then you need 6.7U total, and with a maxSMBBasalMinutes it should start dosing 0.5U every 5m until it gets insulinReq down to 1U. If we assume that it wants to deliver about 4U up front, then that would take 8*5m = 40m to get there. if you bump maxSMBBasalMinutes to 60, it can get it done in 20m using 1U SMBs every 5m. that should be plenty fast for pizza with Fiasp, but if it's not, 75m would get you 1.25U SMBs, so it could get to 3.75U in 15m.
Dana Lewis
@danamlewis
Sep 29 2017 19:02
@sdneufer yep
Scott Leibrand
@scottleibrand
Sep 29 2017 19:03
@CaptainBalou how does that compare to what you're actually seeing? and if you're being limited by a maxBolus that's too low, what does the full pump-loop.log output for that run say? there's some debugging lines in there that indicate whether it sees enough COB to justify using maxSMBBasalMinutes, for example
sdneufer
@sdneufer
Sep 29 2017 19:09
@danamlewis Hmm... It is not working for me and yet it seems so simple to setup. For the Event side on the pebble app do you match case and spaces to your IFTTT applet names? or is an Event not the same as an applet?
Dana Lewis
@danamlewis
Sep 29 2017 19:11
event name needs to match. I use underscores, i.e. force_zero_temp
FYI I think my pebble thisbutton applets in IFTTT are actually maker/maker now
if maker event "force_zero_temp" (i.e. pebble thisbutton pushed), then make a web request to NS
the do_button ones are for my phone I think
so check to see - I have to set up different ones for phone IFTTT vs pebble
Marco
@CaptainBalou
Sep 29 2017 19:21
@scottleibrand I cannot look into the data because i am not at my client anymore. But later I will. Can I go back some hours in NightScout to see what OpenAPS did at a particular time?
Scott Leibrand
@scottleibrand
Sep 29 2017 19:31
yeah, you can click and drag on the bottom NS graph to scroll back and then look at the OpenAPS pill. you can also use something like less to view /var/log/openaps/pump-loop.log
Dana Lewis
@danamlewis
Sep 29 2017 19:41
also NS reports has a nice way (not sure which chart?) to click on a BG dot and see the pill info now, too
so that might be easier than scrollback on NS itself in the app/site if you have trouble getting to the right time
(if someone figures out which report that's in, would make a great PR to the docs!)
Dana Lewis
@danamlewis
Sep 29 2017 20:02
speaking of PRs and :clap: -worthy notes - props to @parenthetic for this great chart addition to help illustrate what's likely to cap basal recommendations & better understanding the safety multpliers vs other values: http://openaps.readthedocs.io/en/latest/docs/While%20You%20Wait%20For%20Gear/preferences-and-safety-settings.html#important-note-about-safety-multipliers
also http://openaps.readthedocs.io/en/latest/docs/While%20You%20Wait%20For%20Gear/understanding-insulin-on-board-calculations.html for more drill down details on netIOB and how it's calculated (will get updated when 0.6.0 is released with exponential curves )
it hasn't been updated since around 0.3/0.4, and there have definitely been some additions in recent versions that need adding to the example, plus more visuals. in case anyone is thirsting for a really great PR idea :D and wants to better document determine-basal.
PopcornPam
@PopcornPam_twitter
Sep 29 2017 20:53
Hi all, one of my rigs (edison explorer, dev branch) keeps coming up with this error as I try to re-run oref with a new phone's address. Re-install worked fine on my other rig...
so it is not finishing the setup script
Scott Leibrand
@scottleibrand
Sep 29 2017 20:54
I just deleted that, as it contained your NS URL and (hashed) API secret
you'll need to go change that now
PopcornPam
@PopcornPam_twitter
Sep 29 2017 20:55
crap.. ok
any idea of whats going on with the loose object?
Scott Leibrand
@scottleibrand
Sep 29 2017 20:55
not sure, but you'll want to re-install following the --nogit instructions in the dev PR
PopcornPam
@PopcornPam_twitter
Sep 29 2017 20:56
OK, will do
Also, I am noticing that the minguardbg is showing some wierd numbers. Is this a number you can set in openaps like in loop, or it changes by the algorithm?
PopcornPam
@PopcornPam_twitter
Sep 29 2017 21:04
Also, @scottleibrand , I don't really understand what git is, or the job it does with openaps. Google seems to indicate it is a type of code, but I can't figure out what it does
Savek-CC
@savek-cc
Sep 29 2017 21:04
@PopcornPam_twitter it's basically a versioning system, allowing multiple ppl to edit code in one project at the same time
so git is used for collaboration and to have a versioning for the data (source code)
Scott Leibrand
@scottleibrand
Sep 29 2017 21:22
minGuardBG is not a setting, it's the lowest predicted BG. I probably would've called in minPredBG, but I had already used that name for something else.
git was originally used by openaps for tracking history on data files. it was never a perfect fit for non-interactive use, and we finally decided it had outlived its usefulness, as the only time anyone interacted with it was when it broke. so we removed git in 0.6.0 for new installs.
git's intended use case is as a command-line tool for keeping source code up to date. we still use it for that (for example in ~/src/oref0), just not in ~/myopenaps/
David
@david-t1
Sep 29 2017 21:36
@gkreiner You can check your local library, some have 3d printers.
PopcornPam
@PopcornPam_twitter
Sep 29 2017 21:46
gotcha
garykidd
@garykidd
Sep 29 2017 22:29
@PopcornPam_twitter get your batteries back yet?
jaylagorio
@jaylagorio
Sep 29 2017 23:23
@scottleibrand I'm about to do yet another pull but while I'm here I should probably ask: why after running setup on my Edison with a directly connected USB G5 receiver do I not see BGs get pushed to Nightscout? The rig picks them up (they appear in the cgm-loop log), the loop definitely gets them and uses them, and even the predicted BGs get put on the Nightscout screen, but until I manually add a ns-upload entries command to my cron the current and historic BGs don't get pushed. Here's my setup command:
oref0-setup --dir=/root/openaps --serial=386xxx --cgm=G5 --ns-host=http://<urlhere> --api-secret=<secret> --tty=/dev/spidev5.1 --enable=' autosens meal autotune microbolus ' --radio_locale='US' --btmac=CC:F3:A5:7B:xx:xx
gregorschlump
@gregorschlump
Sep 29 2017 23:29
@danamlewis Thanks for emphasizing this point. I'm looking forward to build a whole new setup (new pump and new cgm). So I'm strating to compare different settings... ;)
Dana Lewis
@danamlewis
Sep 29 2017 23:36
@jaylagorio hmm. wonder if that's what another G5'er was experiencing, not sure if they had looping and just no uploads or if different issue.
@gregorschlump :+1:
jaylagorio
@jaylagorio
Sep 29 2017 23:38
Yeah, I don't know. Once I add the ns-upload entry to cron everything's fine but occasionally the CGM display number does jump ahead of the other numbers and makes things look off occasionally. What entry should I see and in what file that would do the upload of the CGM to Nightscout in a loop that was installed as expected?
Dana Lewis
@danamlewis
Sep 29 2017 23:38
Big thank you to everyone who's been testing dev, and previous versions of 0.6.0, and providing feedback. Sharing a little more of my testing experiences and outcome here: https://twitter.com/danamlewis/status/913904281182339073
jaylagorio
@jaylagorio
Sep 29 2017 23:38
I can go looking for that and maybe see why the setup script didn't make it happen
Scott Leibrand
@scottleibrand
Sep 29 2017 23:56
@jaylagorio there is no g5-upload config for oref0-setup. why aren't you just pulling BG over from Dexcom's servers?
the g5 config is the equivalent of g4-local-only, because there's no raw data to upload, and because everyone using the G5 has the ability to upload to NS via Dexcom's app and servers