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

18th
Feb 2018
Dana Lewis
@danamlewis
Feb 18 2018 00:37
@brendanpuck should pull units from the pump, and ditto NS. Won't be in oref0 setup.
@drich4608 great! Glad you got it sorted.
Dave Rich
@drich4608
Feb 18 2018 00:38
loop stopped 5 hrs ago tried oref0 setup again this is what comes up..Couldn't pull latest decocare 0.1.0-dev
Brendan Puckeridge
@brendanpuck
Feb 18 2018 00:44
Thanks @danamlewis Any tips on where to look? ns config-vars has DISPLAY_UNITS=mmol. Pump is a 722 NA set to mmol/l. Rig is edison+explorer. pumpprofile.json has "out_units": "mmol/L"
Dana Lewis
@danamlewis
Feb 18 2018 01:13
@drich4608 usually means rig is not online. Ping google and check connectivity then retry
@brendanpuck if everything else is showing as mmol, I wouldn't (necessarily) worry about upload records on the graph. That's likely not an issue and probably not related to the time zone funk
Can't tell from previous convo; but can you describe what happens if you delete all the "future" treatments, do more then get uploaded in the same way?
Deweyoxberg
@Deweyoxberg
Feb 18 2018 01:21
Whenever someone has a minute please, a new error cropped up I've not seen before:
Feb 17 17:20:06 DiabeticinJapan pump-loop.log: /usr/local/bin/oref0-pump-loop: line 124: bc: command not found Feb 17 17:20:06 DiabeticinJapan pump-loop.log: /usr/local/bin/oref0-pump-loop: line 124: ((: < -55 : syntax error: operand expected (error token is "< -55 ") Feb 17 17:20:06 DiabeticinJapan pump-loop.log: /usr/local/bin/oref0-pump-loop: line 124: bc: command not found Feb 17 17:20:06 DiabeticinJapan pump-loop.log: /usr/local/bin/oref0-pump-loop: line 124: ((: > 55 : syntax error: operand expected (error token is "> 55 ") Feb 17 17:20:06 DiabeticinJapan pump-loop.log: /usr/local/bin/oref0-pump-loop: line 128: bc: command not found Feb 17 17:20:06 DiabeticinJapan pump-loop.log: /usr/local/bin/oref0-pump-loop: line 128: ((: > -60 : syntax error: operand expected (error token is "> -60 ")
no idea what this all about :(
Had to redo the oref0 setup today to change over to Heroku
Dave Rich
@drich4608
Feb 18 2018 03:53
TA @danamlewis
Sebastien Lussier
@SebastienLussier
Feb 18 2018 04:29
@brendanpuck Please take a look at openaps/decocare#29
This fix in decocare dev branch, which is not the one installed by the openaps setup script
Brendan Puckeridge
@brendanpuck
Feb 18 2018 06:20
@danamlewis I can confirm a previous delete of “future” entries meant that they deleted and from then, new “future” readings continued to appear. I will do it again though. Interestingly, a manual bolus on the pump i.e. no BG recorded does get plotted correctly and does NOT appear in the ‘future’ list.
@SebastienLussier Thanks for that PR link. That was the one I remember reading and has answered my initial question - that fix is not yet in master. I can see it was pushed to dev. I’ll have a closer look at that and comment on the ticket if it does make some difference here (I suspect it may).
LilDucky
@LilDucky
Feb 18 2018 07:42

Getting some strange things happening. Have restarted rig etc and loop failing consistently. Only thing obvious in the logs following lines

No deliverAt found.
{"reason":"CGM is calibrating, in ??? state, or noise is high. Temp 0 <= current basal 2.35U/hr; doing nothing. "}
Couldn't smb_verify_suggested

CGM is xdrip and rig has internet access. Any assistance welcome.

sooperlooper
@sooperlooper
Feb 18 2018 07:50

This is something i've been pondering since I started using the Spike App (https://spike-app.com). Would it be possible for Spike to send BG to an OpenAPS rig via Bluetooth?

If you have an Android phone you can use xDrip+ to loop offline but currently there is no good solution for iPhone users to loop offline. Lookout or xdrip-js-logger require the rig be BT tethered to the iPhone to show BGs when offline (through the local webpage). This is problematic because BT tethering is a huge battery drain and it prevents the iPhone from connecting to wifi networks.

I'm not too familiar with the Loop app but I believe it can send BG info to the RileyLink via Bluetooth and it doesn't uses the iPhone's Personal Hotspot. Could Spike be programmed to do the same thing?

Scott Leibrand
@scottleibrand
Feb 18 2018 08:16
Dana and I (and possibly others) have some concerns around whether it would be safe to use Spike as a BG source for looping yet. I think we’ll need to have some folks review the code and look at how it deals with CGM noise to see what changes would be needed before it’d be safe to consider adding an integration.
francesc0-cgm
@francesc0-cgm
Feb 18 2018 08:27
@scottleibrand
Scott Leibrand
@scottleibrand
Feb 18 2018 08:38
fidoman
@efidoman
Feb 18 2018 10:34
Lookout and Logger do not require bt tethering to the phone for offline looping. Lookout has a web page built-in and Openaps has a web page. I haven't tried it, but the way I understand a relatively new parameter option in openaps that provides additional offline support by allowing the rig itself to be a wifi-hotspot so you can connect to its wifi from any phone, tablet, etc. and bring up the offline web page ("offline-hotspot" is the parameter).
azizmarafie
@azizmarafie
Feb 18 2018 10:36
what do i need to do to know what build i am using ?
fidoman
@efidoman
Feb 18 2018 10:38
@azizmarafie - I know there is probably a better way, but this should work
cd /root/src/oref0 && grep version package.json
azizmarafie
@azizmarafie
Feb 18 2018 10:41
i am using
0.5.2
is it the latest?
root@unilinux:~/src/oref0# tail -F /var/log/openaps/pump-loop.log4 tail: cannot open ‘/var/log/openaps/pump-loop.log4’ for reading: No such file or directory
how can i solve it ?
azizmarafie
@azizmarafie
Feb 18 2018 10:47
i did run again
azizmarafie
@azizmarafie
Feb 18 2018 10:55
root@unilinux:~/src/oref0# tail -F /var/log/openaps/pump-loop.log4 tail: cannot open ‘/var/log/openaps/pump-loop.log4’ for reading: No such file or directory
i am still getting this
azizmarafie
@azizmarafie
Feb 18 2018 11:10
how can i delete the old logs ?
fidoman
@efidoman
Feb 18 2018 11:17
@azizmarafie - are you using master or the dev branch?
Are you trying to look at the loop log? If so, here's the command I use for that.
tail -F /var/log/openaps/pump-loop.log
fidoman
@efidoman
Feb 18 2018 11:27
If you want to delete the old logs then you can use the following command
cd /var/log/openaps && rm ./*.1
You could also remove the .gz files but that is probably not necessary given they don't take up that much space. I'm not sure space is your issue though.
azizmarafie
@azizmarafie
Feb 18 2018 11:33
thanks seems to me i had a space issue
sudo apt-get install ncdu used this
fidoman
@efidoman
Feb 18 2018 11:53
@azizmarafie - Nice work. I really like ncdu and use it on any unix machine where I want to find what is gobbling up the space. :smile:
th122
@th122
Feb 18 2018 12:07
What is the preferred way to sneak an addition into the openaps documentation? PR or edit on Github?
fidoman
@efidoman
Feb 18 2018 12:26
@th122 - Both, you do an edit on Github and then PR the edit.
th122
@th122
Feb 18 2018 12:28
Ah, thanks - I hadn’t noticed that the edit would use a fork (didn’t want to mess with the main branch unauthorized)
fidoman
@efidoman
Feb 18 2018 12:31
Yes, I've had those same worries / troubles trying to use github. It does not have the most user-friendly workflow / UI. I actually use unix command line git 99% of the time now. It is even less user friendly but no bugs and I know what it is doing. :smile:
Dana Lewis
@danamlewis
Feb 18 2018 12:34
@th122 you can target PRs of the docs to master. http://openaps.readthedocs.io/en/latest/docs/Resources/my-first-pr.html
th122
@th122
Feb 18 2018 12:36
@danamlewis thanks, will do - Noticed that the disk space problem on edison is a recurring one, and getting a pre-owned one yesterday gave me a good grip why.
Dana Lewis
@danamlewis
Feb 18 2018 12:41
:+1:
parenthetic
@parenthetic
Feb 18 2018 14:36
Has anyone successfully set up npm test on a Windows machine?
Deweyoxberg
@Deweyoxberg
Feb 18 2018 15:09
Good morning everyone :)
Curious: Any reason why the APS pill would report unknown along with the pump pill showing "undefined"? Only change I've made last night is migrated from Azure to Heroku, and enabled token authentication. Yes, I checked to make sure the token was entered correctly :P
Dana Lewis
@danamlewis
Feb 18 2018 15:15
Did you update your URL in your runagain, and run it again?
Deweyoxberg
@Deweyoxberg
Feb 18 2018 15:16
I redid the oref0 setup in full, yes.
Dana Lewis
@danamlewis
Feb 18 2018 15:17
Cat your runagain and double check everything looks right?
Deweyoxberg
@Deweyoxberg
Feb 18 2018 15:17
Did, several times. :)
Dana Lewis
@danamlewis
Feb 18 2018 15:18
Are you looping?
Deweyoxberg
@Deweyoxberg
Feb 18 2018 15:19
Apparently not; I looked for the big blob of calculations to show the loop was happening but no go.
Found new errors I'd never seen before as well.
things like: (: < -55 : syntax error: operand expected (error token is "< -55 ")
Dana Lewis
@danamlewis
Feb 18 2018 15:20
Haven't seen those. (Also the ultimate check is to look on pump and see if recent temp was set)
Did you add token auth during oref0 setup?
Deweyoxberg
@Deweyoxberg
Feb 18 2018 15:21
Yep; setup asked if I wanted to enable token authentication and asked for the token at that time. It's in ns.ini along with the new url.
Dana Lewis
@danamlewis
Feb 18 2018 15:23
Ok. I'd probably rerun setup again, and see if it still ends up with the same errors. And if you get them again, i would test switching back to API secret and see if that clears the errors
Deweyoxberg
@Deweyoxberg
Feb 18 2018 15:24
That was my thinking as well :) Will give that a go. At the moment rig seems to be stuck in a loop establishing network connectivity so will have to wait until I can get it powercycled.
Hooray for remote support :D
Dana Lewis
@danamlewis
Feb 18 2018 15:24
:+1: 😊
Deweyoxberg
@Deweyoxberg
Feb 18 2018 15:36
Managed to get it going again; setup re-run (I messed up the CGM setting). How long before I know if this works? 20ish minutes?
Deweyoxberg
@Deweyoxberg
Feb 18 2018 15:41
Looking at the log, seems like I'm still having trouble establishing a reliable connection between rig and pump. :(
was working fine before the migration. bleh.
just keeps doing this in the log
Preflight OK. Done waiting for rigs with better signal.
Unsuccessful oref0-pump-loop at Mon Feb 19 00:45:20 JST 2018


Starting oref0-pump-loop at Mon Feb 19 00:45:32 JST 2018 with 2 second wait_for_silence:
Waiting up to 4 minutes for new BG: ls: cannot access /tmp/pump_loop_completed: No such file or directory
Radio ok. Listening: .
Deweyoxberg
@Deweyoxberg
Feb 18 2018 15:55
Well, loop completed. Finally. :P Still no pill updates. Going to let it run for a bit longer before retrying API_SECRET instead of token.
Do I need to re-run setup all over again for API_SECRET or can I just edit a file somewhere?
Deweyoxberg
@Deweyoxberg
Feb 18 2018 16:10
WOOO
Success. token based :D hooray.
jeez, docs are not kidding when it says "be patient". That was more like 35 minutes heh
Dana Lewis
@danamlewis
Feb 18 2018 16:31
🤣 yea depending on orientation of rig etc it can be longer. Totally depends. Glad it's working! 👍🏻
Deweyoxberg
@Deweyoxberg
Feb 18 2018 16:35
I think rig was just scared you might get involved. :P
"Oh no! Digital mom is going to monkey with my code! Quick, work!" :P
Dana Lewis
@danamlewis
Feb 18 2018 16:37
😂
Doug
@dougboss
Feb 18 2018 16:58
Anyone tried using a charger like this for second or backup battery ? https://www.horizonhobby.com/1s-usb-li-po-charger--500mah-high-current-umx-p-eflc1013
th122
@th122
Feb 18 2018 18:38
Beginner problem? on dev branch, openaps init bails (called within myopenaps directory, with and without the pathname to the directory as argument).
root@antlia:~/myopenaps# openaps init
Traceback (most recent call last):
  File "/usr/local/bin/openaps", line 170, 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", line 155, in run
    builtins.dispatch(args, self)
  File "/usr/local/lib/python2.7/dist-packages/openaps/builtins.py", line 44, in dispatch
    command(args)
  File "/usr/local/lib/python2.7/dist-packages/openaps/init.py", line 6, in init
    if args.args[1] == '--nogit':
IndexError: list index out of range
Eric
@ecc1
Feb 18 2018 18:42
@dougboss I don't know what kind of connector that is, or how you'd connect it to your batteries.
Dana Lewis
@danamlewis
Feb 18 2018 18:48
@th122 why do you need to init? What are you trying to do?
th122
@th122
Feb 18 2018 18:49
Trying to set up a openaps/oref0, I hadn’t before - busy with documenting and parsing cgm data :)
Doug
@dougboss
Feb 18 2018 19:11
@ecc1 the ada fruit battery has a JST PH 2.0mm connector and this charger is designed for that size connector. It only charges at .5 ah so its relatively slow and safe. I fly RC planes and had the charger in my kit. It would be used for charging a second or backup battery
Carl
@c-robertson
Feb 18 2018 19:28
My bluetooth tethering stopped working last week. Troubleshooting now I notice that the rig doesn't get an IP addr when tethered
Carl
@c-robertson
Feb 18 2018 19:34
after running 'sudo bt-pan client <my phone BT MAC addr>' and then 'sudo dhclient bnep0', I check status via 'ifconfig bnep0' and get the following
'''
root@cr-apsrig:~# ifconfig bnep0
bnep0 Link encap:Ethernet HWaddr 58:a8:39:01:14:8a
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:3 errors:0 dropped:0 overruns:0 frame:0
TX packets:3 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:391 (391.0 B) TX bytes:96 (96.0 B)
'''
...and that markdown formatting never works for me for some reason
anyway, no IP even though bnep0 is connected to Ethernet... any suggestions?
root@cr-apsrig:~# ifconfig bnep0
bnep0     Link encap:Ethernet  HWaddr 58:a8:39:01:14:8a
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:3 errors:0 dropped:0 overruns:0 frame:0
          TX packets:3 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:391 (391.0 B)  TX bytes:96 (96.0 B)
(got it!)
Jon Cluck
@cluckj
Feb 18 2018 19:37
it usually takes a little while for my rig to get an IP
did dhclient return any errors?
Carl
@c-robertson
Feb 18 2018 19:38
@cluckj no errors returned from dhclient
I did notice the BT tether connection 'flapping', testing again to grab the network log...
Jon Cluck
@cluckj
Feb 18 2018 19:41
if cron is running, it'll be calling oref0-online every minute and probably resetting the connection
Carl
@c-robertson
Feb 18 2018 19:44
so, after running 'iwconfig wlan txpower off' on the rig, I see the following in my networklog...
At Sun Feb 18 14:42:03 EST 2018 my wifi is not connected

Getting new wlan0 IP

No Internet access detected, attempting to connect BT to 70:28:8B:97:CE:E9
Connected to Bluetooth with IP: unassigned

Offline hotspot not enabled in preferences.json
At Sun Feb 18 14:42:07 EST 2018 my local hotspot is not running (and not enabled in preferences.json)
Cycling wlan0
Dana Lewis
@danamlewis
Feb 18 2018 19:45
@th122 you shouldn't have to init for setup. That's usually an error message when you run an OpenAPS command outside of your myopenaps directory. So, now that you're in there- rerun the previous command that had failed before
I have an improved error message sitting in the openaps toolkit that we need to get merged to help cut down on that confusion
Carl
@c-robertson
Feb 18 2018 19:47
and now I see the flapping (on my phone the BT tether disappears every minute, then reappears...)
th122
@th122
Feb 18 2018 19:47
I found the hints on that (searched for others with the same problem) Running from within the myopenaps directory makes no difference. (called within myopenaps directory, with and without the pathname to the directory as argument).
@danamlewis tried again with removing the myopenaps directory, and running src/oref0/bin/oref0-setup.sh interactively (had been running it with arguments the first time) but that doesn’t seem to help either.
Carl
@c-robertson
Feb 18 2018 19:50
Starting oref0-online at Sun Feb 18 14:49:02 EST 2018.
At Sun Feb 18 14:49:02 EST 2018 my Bluetooth is connected to Carl's Samsung GS7
At Sun Feb 18 14:49:02 EST 2018 my local Bluetooth IP is: unassigned
At Sun Feb 18 14:49:02 EST 2018 my public IP is: not found
Jon Cluck
@cluckj
Feb 18 2018 19:51
Do you have a program to auto-tether?
Carl
@c-robertson
Feb 18 2018 19:53
So, I had the "BTautotether" app installed on my phone, but every time the rig would initiate a connection I would get a dialog on my phone (in other words, it wasn't as automatic as I would like
so I tried another app, same behavior
when I uninstalled that app this week, the erroneous behavior started...
maybe I need the app, even though it is not really 'autotethering' (I have to acknowledge it via the dialog on my phone every time...)?
Jon Cluck
@cluckj
Feb 18 2018 19:55
BT autotether shows me a message that something is connected, but I don't have to do anything to allow the connection
Carl
@c-robertson
Feb 18 2018 19:57
yeah, I had asked before and the theory was that the dialog is a Samsung thing...
Jon Cluck
@cluckj
Feb 18 2018 19:58
Ah
Dana Lewis
@danamlewis
Feb 18 2018 20:03
@th122 usually means your installation didn't finish correctly - I'd back up and try to reinstall then
th122
@th122
Feb 18 2018 20:04
starting with recloning oref0? I’ll feed my family and go back to that. Thanks @danamlewis!
Dana Lewis
@danamlewis
Feb 18 2018 20:08
Yup
Carl
@c-robertson
Feb 18 2018 20:13
so, I think I will unpair my rig from my phone and start over with the docs
peterfulvi
@peterfulvi
Feb 18 2018 21:17
What causes the offline_hotspot to broadcast its ssid? It does not show up on any of my devices?
Doug
@dougboss
Feb 18 2018 21:22
looking for help with papertrail - have everything else set up and working, bluetooth tether, loop is closed, all looks good except that after running the papertrail setup 3x I get no logs .... except what looks like default entries from papertrail itself Feb 18 15:01:01 PMEDIAAPP-01 run-parts(/etc/cron.hourly): finished 0anacron
Feb 18 15:16:26 PMEDIAAPP-01 dhclient: DHCPREQUEST on eth0 to 172.31.0.1 port 67 (xid=0x69ec4bdf)
Feb 18 15:16:26 PMEDIAAPP-01 dhclient: DHCPACK from 172.31.0.1 (xid=0x69ec4bdf)
Feb 18 15:16:27 PMEDIAAPP-01 dhclient: bound to 172.31.1.71 -- renewal in 1790 seconds.
Any ideas ? I can see that I have logs in var/logs/openaps but they dont make it to papertrail
Dana Lewis
@danamlewis
Feb 18 2018 21:31
@peterfulvi if the rig doesn't have connectivity.
Doug
@dougboss
Feb 18 2018 21:50
set remote _syslog to debug level and saw this in logs but still havent figured out whats going on
2018-02-18 21:35:06 INFO remote_syslog.go:55 Connecting to logs2.papertrailapp.com:30XX over tls
2018-02-18 21:35:07 DEBUG remote_syslog.go:167 Evaluating globs every 10s
2018-02-18 21:35:07 DEBUG remote_syslog.go:183 Evaluating file globs
2018-02-18 21:35:07 INFO remote_syslog.go:202 Forwarding file: /var/log/openaps/pump-loop.log
2018-02-18 21:35:07 INFO remote_syslog.go:202 Forwarding file: /var/log/openaps/autosens-loop.log
2018-02-18 21:35:07 INFO remote_syslog.go:202 Forwarding file: /var/log/openaps/ns-loop.log
2018-02-18 21:35:07 INFO remote_syslog.go:202 Forwarding file: /var/log/openaps/network.log
2018-02-18 21:35:07 INFO remote_syslog.go:202 Forwarding file: /var/log/openaps/autotune.log
2018-02-18 21:35:07 ERROR remote_syslog.go:192 Cannot forward /var/log/openaps/cgm-loop.log, it may not exist
2018-02-18 21:35:07 INFO remote_syslog.go:202 Forwarding file: /var/log/openaps/pushover.log
2018-02-18 21:35:17 DEBUG remote_syslog.go:183 Evaluating file globs
2018-02-18 21:35:17 DEBUG remote_syslog.go:198 Skipping /var/log/openaps/pump-loop.log because it is already running
2018-02-18 21:35:17 DEBUG remote_syslog.go:198 Skipping /var/log/openaps/autosens-loop.log because it is already running
2018-02-18 21:35:17 DEBUG remote_syslog.go:198 Skipping /var/log/openaps/ns-loop.log because it is already running
2018-02-18 21:35:17 DEBUG remote_syslog.go:198 Skipping /var/log/openaps/network.log because it is already running
2018-02-18 21:35:17 DEBUG remote_syslog.go:198 Skipping /var/log/openaps/autotune.log because it is already running
2018-02-18 21:35:17 DEBUG remote_syslog.go:198 Skipping /var/log/openaps/pushover.log because it is already running
bryanslaski
@bryanslaski
Feb 18 2018 21:50
Hello all; I'm brand new to the community and getting ready to start my build... very excited!!! Plan to take it slow over the next month to give Nightscout plenty of data to start off with. Speaking of, my first question is regarding the collect and prepare section of Openaps. It makes mention of "log your carbs and boluses" so I can get the most out of Autotune later on. What is the best way of capturing this data? I am currently using an Animas Vibe along with an independent Dexcom G5. Can I capture it on the Dexcom mobile app / is there a way to upload my pump to the site / is there a way to manually enter into nightscout? I have a MM 522 pump as well if I would have to switch over as well... Appreciate any help!!
Dana Lewis
@danamlewis
Feb 18 2018 22:00
Welcome @bryanslaski! You can input that into NS, either direct through care portal or setting up things like IFTTT to easily enter things from your watch or phone. The automatic upload option would be if you switched over to your 522 and had OpenAPS running on the rig to read only and update the data to NS, even if you weren't ready to loop with it
bryanslaski
@bryanslaski
Feb 18 2018 22:08
Thanks Dana (both for the quick response and ALL of the great info I've seen so far!) I had seen careportal referenced but for whatever reason, I am having trouble finding any info around installing / linking it to my nightscout page, but I'll keep digging. Seems like the easiest / most comprehensive route may just be to switch back to the Medtronic and use it for a few weeks ahead of looping
Dana Lewis
@danamlewis
Feb 18 2018 22:14
bryanslaski
@bryanslaski
Feb 18 2018 22:14
never mind... I got it!! (At least the careportal part) Thanks again for the help @danamlewis I'm sure I will have plenty more (and hopefully much better..) questions as I get into this!!
Dana Lewis
@danamlewis
Feb 18 2018 22:14
Great! :smile:
Doug
@dougboss
Feb 18 2018 22:30
Hello all .... I started in Edison rig about 5 hours ago and have been really impressed so far. Thanks to all the folks to have contributed to this ... Its very impressive.
Dana Lewis
@danamlewis
Feb 18 2018 22:31
@dougboss :wave: welcome!