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

29th
Jun 2018
Jon Cluck
@cluckj
Jun 29 2018 00:14
@richardshane the pi0w is the part that does that, and yes the setup is the same
Martin Haeberli
@mhaeberli
Jun 29 2018 00:37
I need to write up my experiments again, but, at the lower power consumption, 4400 mAH battery lasted maybe 21 hours; higher consumption proportionately shorter. Lines of inquiry: - can we pull consumption down further with dynamic clocking somehow - run fast for loop result, then slow until next loop time. Another line - battery percentage indication through NightScout seems unreliable to say the least. Lastly, batteries here were charged with an external charger; perhaps different (better, I hope) results would be found using internal charger?
tepidjuice
@tepidjuice
Jun 29 2018 02:15
This is probably the wrong place to ask but I need some help with dexcom cgm sites. I can't use my lower abdomen as I have to wear a harness for climbing and I think my arms are too exposed when climbing. I've been using the inside of my thighs for about a year but recently i.e. last few months the readings have been unreliable and the sites are now painful. Any ideas for reliable alternative sites? I can't use my loop anymore as the cgm readings are so bad.
Tim Gunn
@Tornado-Tim
Jun 29 2018 02:16
I use the bottom of my back... Above the waistline, works well for me....
tepidjuice
@tepidjuice
Jun 29 2018 02:18
@Tornado-Tim I'd try there but that's exactly where my harness sits.
Dana Lewis
@danamlewis
Jun 29 2018 02:18
@tepidjuice fronts of arm still an issue?
Tim Gunn
@Tornado-Tim
Jun 29 2018 02:19
If you do the arm I found the best place is close to the arm pit as it doesn't get in the way....
Less likely to get clipped etc....
tepidjuice
@tepidjuice
Jun 29 2018 02:19
@danamlewis I've never tried the fronts of my arms. I assume my thigh sites don't work anymore as the sensor is going into muscle. I guess my arms would be worse?
@Tornado-Tim like near the shoulder?
Dana Lewis
@danamlewis
Jun 29 2018 02:21
I get great experience with them at the tops of my arms, front and sides
Jon Cluck
@cluckj
Jun 29 2018 02:21
back of the calf?
Tim Gunn
@Tornado-Tim
Jun 29 2018 02:21
Yes, I did xfit with a sensor on doing muscle ups and found it good....
Dana Lewis
@danamlewis
Jun 29 2018 02:21
Worth trying. Even with mostly muscle.
tepidjuice
@tepidjuice
Jun 29 2018 02:22
So any ideas why the thighs stopped working + causing pain? If going into active muscle isn't a problem with the arms?
Dana Lewis
@danamlewis
Jun 29 2018 02:22
(Insert upside down, e.g. Where you are inserting down rather than inserting up into your arm, and it's easy to do one handed)
Maybe lack of rotation and overuse, etc. Bodies are weird. Diabetes sucks. Etc.
tepidjuice
@tepidjuice
Jun 29 2018 02:24
@danamlewis so the dexcom logo on the transmitter is pointing down i.e. D is closest to my head and M is closest to the floor?
@cluckj I've not tried the back of the calf. Slightly to one side high or low?
Dana Lewis
@danamlewis
Jun 29 2018 02:25
/me twists awkwardly to stare at her transmitter
Yep!
Jon Cluck
@cluckj
Jun 29 2018 02:27
@tepidjuice wherever there's enough skin to pull up & it won't restrict your movement
but...outside, high?
tepidjuice
@tepidjuice
Jun 29 2018 02:28
@cluckj For me the inside of my calf high up is much more pinchable.
Jon Cluck
@cluckj
Jun 29 2018 02:30
@tepidjuice YCMV :laughing:
tepidjuice
@tepidjuice
Jun 29 2018 02:30
:)
So back of calves and upper arms. How close to the shoulder do people usually go? I don't want to waste anymore sensors as I self fund.
Dana Lewis
@danamlewis
Jun 29 2018 02:37
Don't insert into the shoulder bone
Whatever feels right for you. Some people do them closer to the elbow; some people midway; some closer up to the shoulder
You could probably look at google images and YouTube videos for some example locations
tepidjuice
@tepidjuice
Jun 29 2018 03:01
@danamlewis just been doing that. Thanks for your advice.
Dana Lewis
@danamlewis
Jun 29 2018 03:09
:+1: hope things improve!
Patcalfe
@Patcalfe
Jun 29 2018 04:45
OMG! IT's WORKING!
AHHHHHH!!!!!! IT'S AMAZING!
I've been hacking away at this for weeks in my spare time, getting parts organised and things shipped to me.....i actually started about 8 months back but none of the CGM's available were covered in my province, and now I have a freestyle libre and a smartwatch as a cgm and I got the edison and all the crap and it's freaking working!
WOOOOOOO!
Dana Lewis
@danamlewis
Jun 29 2018 04:48
@Patcalfe πŸŽ‰πŸŽŠπŸ™ŒπŸΌ
Patcalfe
@Patcalfe
Jun 29 2018 04:49
There are still many things to get sorted, but i'm so close now and It's totally going to work.
I'm so stoked.
You guys have done an amazing thing with this. Thank you.
Scott Leibrand
@scottleibrand
Jun 29 2018 04:57
:smile: your enthusiasm is infectious. Thanks for sharing it. πŸ˜ƒ
lyntonr
@lyntonr
Jun 29 2018 07:35

still getting shutwon / reboot failures on pibakery install with 0.6.2-dev This is driving me nuts. What Ive found is that the watchdog service fails on exit from daemon.log [1]+ Stopped tail -f /var/log/daemon.log root@lynton_rig:~# bg [1]+ tail -f /var/log/daemon.log & root@lynton_rig:~# root@lynton_rig:~# root@lynton_rig:~# root@lynton_rig:~# reboot Jun 29 17:24:21 localhost systemd[1]: Stopped target Timers. Jun 29 17:24:21 localhost systemd[1]: Stopping Save/Restore Sound Card State... Jun 29 17:24:21 localhost systemd[1]: Stopped Daily Cleanup of Temporary Directories. Jun 29 17:24:21 localhost systemd[1]: Stopped target Graphical Interface. Jun 29 17:24:21 localhost watchdog[555]: stopping daemon (5.15) Jun 29 17:24:21 localhost systemd[1]: Stopping watchdog daemon... Jun 29 17:24:21 localhost systemd[1]: Stopped target Bluetooth. Jun 29 17:24:21 localhost acpid: exiting Jun 29 17:24:21 localhost systemd[1]: Closed Load/Save RF Kill Switch Status /dev/rfkill Watch. Jun 29 17:24:21 localhost systemd[1]: Stopping ACPI event daemon... Jun 29 17:24:21 localhost systemd[1]: Stopped target Sound Card. Jun 29 17:24:21 localhost systemd[1]: Unmounting RPC Pipe File System... Jun 29 17:24:21 localhost bluetoothd[354]: Terminating Jun 29 17:24:21 localhost systemd[1]: Stopped Daily apt upgrade and clean activities. Jun 29 17:24:21 localhost bluetoothd[354]: Stopping SDP server Jun 29 17:24:21 localhost systemd[1]: Stopped Daily apt download activities. Jun 29 17:24:21 localhost bluetoothd[354]: Exit Jun 29 17:24:21 localhost systemd[1]: Stopping Bluetooth service... Jun 29 17:24:21 localhost systemd[1]: Stopped target System Time Synchronized. Jun 29 17:24:22 localhost systemd[1]: Stopped ACPI event daemon. Jun 29 17:24:22 localhost systemd[1]: Stopped Bluetooth service. Jun 29 17:24:22 localhost systemd[1]: Stopped Save/Restore Sound Card State. Jun 29 17:24:22 localhost systemd[1]: Unmounted RPC Pipe File System. Jun 29 17:24:22 localhost systemd[1]: watchdog.service: Control process exited, code=exited status=1 Jun 29 17:24:22 localhost systemd[1]: Stopped watchdog daemon. Jun 29 17:24:22 localhost systemd[1]: watchdog.service: Unit entered failed state. Jun 29 17:24:22 localhost systemd[1]: watchdog.service: Triggering OnFailure= dependencies. Jun 29 17:24:22 localhost systemd[1]: Requested transaction contradicts existing jobs: Resource deadlock avoided Jun 29 17:24:22 localhost systemd[1]: watchdog.service: Failed to enqueue OnFailure= job: Resource deadlock avoided Jun 29 17:24:22 localhost systemd[1]: watchdog.service: Failed with result 'exit-code'. Jun 29 17:24:22 localhost systemd[1]: Stopped target Multi-User System.

I tried removing the OnFailure=wd_keepalive (or similar) line in the watchdog.service after that watchdog will shutdown without failing and it will continue stopping services until it gets up to openaps-menu service where it then hangs, by then most services are gone and the shell is lost.

Can anyone that has good systemd knowledge please advise a way to debug futher ? I already tried validateing the watchdog.service file.

lyntonr
@lyntonr
Jun 29 2018 07:49
@peterfulvi can you tell me how you get the nice formatted text markdown in the black box into gitter ? I tried var x = 1; above but not what I expected.
Jens Heuschkel
@juehv
Jun 29 2018 07:57
@lyntonr user ``` at the beginning and at the end
for more information klick on this M button on the right of the text field :)
lyntonr
@lyntonr
Jun 29 2018 07:58
thats what i did do
testing   ``` (shift+enter for line break)
var code = "formatted";
```
oh it did work maybe doesnt work in compose mode
Jens Heuschkel
@juehv
Jun 29 2018 07:59
it does ;)
lyntonr
@lyntonr
Jun 29 2018 08:02
line 1 line 2 line 3
dont know what happening here :( lokks like my triple backticks are getting morphed into single ones somehow. sorry for the noise pollution
wilson99
@wilson99
Jun 29 2018 08:58
Does anyone can tell me how to setup peb-urchin-status.sh in cron? I have installed pancreabble and paired with pebble, successfully get the hello testing notification. Also I have manually created ~/myopenaps/pancreoptions.json, when I run peb-urchin-status ~/myopenaps/pancreoptions.json, I get below info:
root@wilson-edison1:~/myopenaps# peb-urchin-status ~/myopenaps/pancreoptions.json
Get info failed: No such device
[Errno 2] No such file or directory
openaps: 'urchin-loop' is not an openaps command. See 'openaps --help'.
Cylon75
@Cylon75
Jun 29 2018 11:45

How precise does the remaining insulin units check work? I got an alert at 11:52 that there were 0.0 units remaining from my 754 Veo pump. At 12:03 there was a 0.4 units bolus delivered by SMB which should not have been possible because I did not change the reservoir yet. The pump does not seem to block it, but OpenAps still issued the command to bolus the 0.4 units. Nightscout also recorded this.

The pump-loop.log:

Listening for 5 s silence: Radio ok. Listening: .No interfering pump comms detected from other rigs (this is a good thing!)
Checking reservoir: reservoir level before: 0.0, suggested: 0.0 and after: 0.0
Checking pump status (suspended/bolusing): {"status":"normal","bolusing":false,"suspended":false}
Sending ESC ESC ESC to exit any open menus before SMBing: true
enact/bolused.json: {"insulinReq":0.88,"status":"normal","_type":"BolusRequest","bg":2
09,"bolusing":true,"reservoir":"0.0","requested":{"units":0.4,"params"
:[0,16]},"timestamp":"2018-06-29T12:03:47.062844","raw":"000000000

How is this possible?
What is the difference between before: and suggested:?

I am on 0.6.2-dev

peterfulvi
@peterfulvi
Jun 29 2018 12:15
I turn the rig on but the screen remains blank, even when its plugged in. Its looping fine with Lookout, but the screen remains blank. Is there a manuall way to initiate it?
alimhassam
@alimhassam
Jun 29 2018 13:13
@Cylon75 when the pump reports an empty reservoir it's not really fully empty there is a hidden small amount left
ArvoNiilo
@ArvoNiilo
Jun 29 2018 14:01
trying to set-up a reserve rig (Edison board) for the holidays. I'm stuck with wifi-settings. What does the following error message mean and how can it be fixed. The networks names and password are as in the functioning rig: root@keinohaima1:~# ifup wlan0
wpa_supplicant: /sbin/wpa_supplicant daemon failed to start
run-parts: /etc/network/if-pre-up.d/wpasupplicant exited with return code 1
Failed to bring up wlan0.
Scott Leibrand
@scottleibrand
Jun 29 2018 15:21
@Cylon75 there are probably 5U or so available before the rubber of the plunger hits the end of the reservoir. After that the amount delivered starts to be less than the pump thinks, until you finally get a No Delivery at 8-10U beyond empty. By then you are about 2U short, so don’t let it get that far. Also, being below zero disables the reservoir based safety check on SMB, so it’s possible for two rigs to deliver boluses back to back.
@lyntonr make sure you have a newline after/before your triple backticks.
Scott Leibrand
@scottleibrand
Jun 29 2018 15:30
@richardshane your messages are coming through as gibberish
please stop posting, and come back when you've got Gitter working properly
lyntonr
@lyntonr
Jun 29 2018 15:36

thanks @scottleibrand testing now


var test ="with blank lines";

thanks that was the final straw in a string of frustrating disasters today :)

Dana Lewis
@danamlewis
Jun 29 2018 15:39
@lyntonr (and everyone else), there's tips for formatting for gitter in the docs: http://openaps.readthedocs.io/en/latest/docs/Understanding%20OpenAPS-Overview/communication-support-channels.html#gitter
Jon Cluck
@cluckj
Jun 29 2018 15:45
@lyntonr does directly calling reboot/shutdown with systemctl work? like systemctl reboot or systemctl halt?
Scott Leibrand
@scottleibrand
Jun 29 2018 15:47
/kick @richardshane
(or really @richardshane's cat, I suspect)
lyntonr
@lyntonr
Jun 29 2018 15:48
just had a look @danamlewis hadnt noticed that before (it was hiding in the drop down link) the coverage in the docs is indeed excellent.
@cluckj no same behaviour I sure its a misconfuration in one of the services I'll have another go soon its time consuming cos i dont know a lot about systemd. shutdown is working its just that it gets stuck shuting down watchdog.service appears to be the cultprit when its OnFailure method tries to call wd_keepalive ... If anyone has any debugging tips pleased chip in on this.
lyntonr
@lyntonr
Jun 29 2018 15:54
can someone do something about richards cat the beeping is driving me crazy ?
Scott Leibrand
@scottleibrand
Jun 29 2018 15:56
I don't seem to have admin perms at the moment. why do you have Gitter beeping?
richardshane
@richardshane
Jun 29 2018 15:58
I apologize.
Scott Leibrand
@scottleibrand
Jun 29 2018 16:00
ah, good, you're back. :)
and, I figured out how to fix my admin perms, so I'm going to go delete some of the noise
lyntonr
@lyntonr
Jun 29 2018 16:03
its just the notification alerts .. yes i could have turned them off but i like seeing new posts as they come in .
im in the process of changing pump / oref / nightscout / meters / libre ? to using mg/dl istead of mmol/l .. Aus for some reason adopted mmol/l but most docs etc dont. mg/dl makes more sense to me and doesnt need an extra "." in displays. I am assuming that this will work ok and that under the hood all these things are using mg/dl anyway
wilson99
@wilson99
Jun 29 2018 16:07
@scottleibrand Could you help tell why when I run oref0-setup.sh, there is no prompt for btpeb?
    if [[ ! -z $BT_PEB ]]; then
        read -p "For Pancreabble enter Pebble mac id (i.e. AA:BB:CC:DD:EE:FF) hit enter to skip " -r
        BT_PEB=$REPLY
        echocolor "Ok, $BT_PEB it is."
        echo
    fi
Can I just add --btpeb in oref0-runagain.sh then directly run it?
Scott Leibrand
@scottleibrand
Jun 29 2018 16:08
because that setup is not widely used enough, or polished enough, to present as an option in the interactive setup, so you have to specify it at the command-line or add it to your runagain, ya
wilson99
@wilson99
Jun 29 2018 16:10
Thanks! now I know why I got error as below:
root@wilson-edison1:~/myopenaps# peb-urchin-status ~/myopenaps/pancreoptions.json
Get info failed: No such device
[Errno 2] No such file or directory
openaps: 'urchin-loop' is not an openaps command. See 'openaps --help'.
wilson99
@wilson99
Jun 29 2018 16:33
@scottleibrand I got below error, is there any place to config this json file?
IOError: [Errno 2] No such file or directory: 'monitor/glucose-unzoned.json'
upload/urchin-data.json  raised  [Errno 2] No such file or directory: 'monitor/glucose-unzoned.json'
pbbl://JSON/format_urchin_data/upload/urchin-data.json
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 40, in main
    output = task.method(args, app)
  File "/usr/local/lib/python2.7/dist-packages/openaps/uses/use.py", line 45, in __call__
    output = self.main(args, app)
  File "/root/.local/lib/python2.7/site-packages/pancreabble/urchin.py", line 123, in main
    cgm_history = json.loads(open(args.glucose_history).read())
IOError: [Errno 2] No such file or directory: 'monitor/glucose-unzoned.json'
Seems monitor/glucose-unzoned.json should be changed to monitor/glucose.json
wilson99
@wilson99
Jun 29 2018 16:40
@scottleibrand Should I change it here in openeaps.ini ?
[report "upload/urchin-data.json"]
use = format_urchin_data
reporter = JSON
cgm_clock = monitor/clock.json
action = add
device = pbbl
glucose_history = monitor/glucose-unzoned.json
status_text =
status_json = upload/urchin-status.json
peterfulvi
@peterfulvi
Jun 29 2018 16:51
I have my PI-HAT looping with Lookout. Everything seems to work good, updated my preferences. I cant seem to get the screen to come up though(no unicorn). Is there a way to turn it on. I would hate to think its broken already, i hardly used it. This is the fires time im actually taking out for me
Jon Cluck
@cluckj
Jun 29 2018 16:54
@peterfulvi which branch are you using?
Scott Leibrand
@scottleibrand
Jun 29 2018 16:58
@wilson99 I'm not much help with pancreabble (which is partly why it's unsupported)
and it's quite likely that pancreabble no longer works in 0.7.0-dev because no one has been testing or maintaining it. so if you'd like to use it, you may need to volunteer to do that.
wilson99
@wilson99
Jun 29 2018 16:59
Ok, thanks.
I'm using 0.6.2 now.
Jon Cluck
@cluckj
Jun 29 2018 17:06
@peterfulvi chances are that openaps-menu was built with the old version of node, go into ~/src/openaps-menu/ and run sudo npm install
peterfulvi
@peterfulvi
Jun 29 2018 18:16
@cluckj i just upgradede to node-js v8.10.0 to run Lookoout. I will do that right now
Jon Cluck
@cluckj
Jun 29 2018 18:17
you may have to restart the service after it's done: systemctl restart openaps-menu
peterfulvi
@peterfulvi
Jun 29 2018 18:19
Ok thanks im doing it right now
Jon Cluck
@cluckj
Jun 29 2018 18:20
:thumbsup:
peterfulvi
@peterfulvi
Jun 29 2018 18:21
this was the output
root@pdn:~/src/openaps-menu# sudo npm install
npm WARN openaps-menu@1.0.1 No repository field.

up to date in 23.351s
Jon Cluck
@cluckj
Jun 29 2018 18:22
if the service restart doesn't fix it, run systemctl status openaps-menu and paste the output :)
peterfulvi
@peterfulvi
Jun 29 2018 18:22
ok
I did that and this is what it says
 openaps-menu.service - OpenAPS Menu service for Explorer HAT.
   Loaded: loaded (/etc/systemd/system/openaps-menu.service; enabled; vendor preset: enabled)
   Active: active (running) since Fri 2018-06-29 14:22:58 EDT; 928ms ago
 Main PID: 4264 (openaps-menu.sh)
   CGroup: /system.slice/openaps-menu.service
           β”œβ”€4264 /bin/bash /root/src/openaps-menu/openaps-menu.sh
           β”œβ”€4265 /bin/bash /root/src/openaps-menu/openaps-menu.sh
           β”œβ”€4266 node index.js
           β”œβ”€4267 /bin/sh ./setup.sh
           └─4268 sleep 3

Jun 29 14:22:58 pdn systemd[1]: Started OpenAPS Menu service for Explorer HAT..
Jun 29 14:22:58 pdn openaps-menu.sh[4264]: sh: echo: I/O error
Jun 29 14:22:58 pdn openaps-menu.sh[4264]: sh: echo: I/O error
Jon Cluck
@cluckj
Jun 29 2018 18:24
screen's not on?
peterfulvi
@peterfulvi
Jun 29 2018 18:25
@cluckj no nothing. It loops, Lookout reads my transmitter but no screen.
Jon Cluck
@cluckj
Jun 29 2018 18:26
do you have time to debug now?
peterfulvi
@peterfulvi
Jun 29 2018 18:27
No not really. Im at work. If i know what to do i can work on it tommorrow night.
Jon Cluck
@cluckj
Jun 29 2018 18:27
okay :)
run the service status again, let's see if it crashed with any nice errors?
peterfulvi
@peterfulvi
Jun 29 2018 18:29
It was different that time
● openaps-menu.service - OpenAPS Menu service for Explorer HAT.
   Loaded: loaded (/etc/systemd/system/openaps-menu.service; enabled; vendor preset: enabled)
   Active: activating (auto-restart) (Result: exit-code) since Fri 2018-06-29 14:28:35 EDT; 6s ago
  Process: 6542 ExecStart=/root/src/openaps-menu/openaps-menu.sh (code=exited, status=1/FAILURE)
 Main PID: 6542 (code=exited, status=1/FAILURE)

Jun 29 14:28:35 pdn systemd[1]: openaps-menu.service: Main process exited, code=exited, status=1/FAILURE
Jun 29 14:28:35 pdn systemd[1]: openaps-menu.service: Unit entered failed state.
Jun 29 14:28:35 pdn systemd[1]: openaps-menu.service: Failed with result 'exit-code'.
peterfulvi
@peterfulvi
Jun 29 2018 18:34
@cluckj I will have to dig later on it. If you have any advice i will chip away at it tomorrow night and let you know the status of things. Thanks
Jon Cluck
@cluckj
Jun 29 2018 18:34
thanks. it could be that the software didn't setup properly, or the screen is broken
ping me when you're free
(the systemd service just calls ~/src/openaps-menu/openaps-menu.sh, so you may want to run that manually for more detailed error messages)
peterfulvi
@peterfulvi
Jun 29 2018 19:08
Thanks