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

6th
Feb 2019
renegadeandy
@renegadeandy
Feb 06 00:49
Should I use 3 or 4 hours in pump settings insulin time for fiasp?
Dana Lewis
@danamlewis
Feb 06 00:52
No
You should keep same 5-7hr dia, just change peak time in preferences to ultra rapid
Dylan Leonard
@dylanleonard
Feb 06 02:45
Thanks for the reassurance @danamlewis
I gave it another shot but got the same result so I went into oref0-log-shortcuts.sh and removed the -q flags from the grep commands to hopefully see what's happening in there and it seems to be echoing out the same lines in an infinite loop...

Sample output:
alias networklog="tail -n 100 -F /var/log/openaps/network.log"

alias xdrip-looplog="tail -n 100 -F /var/log/openaps/xdrip-loop.log"

alias cgm-looplog="tail -n 100 -F /var/log/openaps/cgm-loop.log"

alias autosens-looplog="tail -n 100 -F /var/log/openaps/autosens-loop.log"

alias autotunelog="tail -n 100 -F /var/log/openaps/autotune.log"

alias pump-looplog="tail -n 100 -F /var/log/openaps/pump-loop.log"

alias urchin-looplog="tail -n 100 -F /var/log/openaps/urchin-loop.log"

alias autosens-looplog="tail -n 100 -F /var/log/openaps/autosens-loop.log"

alias cat-pref="cd /root/myopenaps && cat preferences.json"

alias edit-pref="cd /root/myopenaps && nano preferences.json"

alias cat-wifi="cat /etc/wpa_supplicant/wpa_supplicant.conf"

alias edit-wifi="vi /etc/wpa_supplicant/wpa_supplicant.conf"

alias cat-runagain="cd /root/myopenaps && cat oref0-runagain.sh"

alias edit-runagain="cd /root/myopenaps && nano oref0-runagain.sh"

alias cat-autotune="cd /root/myopenaps/autotune && cat autotune_recommendations.log"

alias git-branch="cd $HOME/src/oref0 && git branch"

alias cat-runagain="cd /root/myopenaps && cat oref0-runagain.sh"

alias edit-runagain="cd /root/myopenaps && nano oref0-runagain.sh"

alias edison-battery="cd /root/myopenaps/monitor && cat edison-battery.json"

alias cat-reservoir="cd /root/myopenaps/monitor && cat reservoir.json"

alias stop-cron="cd /root/myopenaps && /etc/init.d/cron stop && killall -g oref0-pump-loop"

alias start-cron="/etc/init.d/cron start"

source /etc/skel/.profile

alias networklog="tail -n 100 -F /var/log/openaps/network.log"

alias xdrip-looplog="tail -n 100 -F /var/log/openaps/xdrip-loop.log"

alias cgm-looplog="tail -n 100 -F /var/log/openaps/cgm-loop.log"

alias autosens-looplog="tail -n 100 -F /var/log/openaps/autosens-loop.log"

alias autotunelog="tail -n 100 -F /var/log/openaps/autotune.log"

alias pump-looplog="tail -n 100 -F /var/log/openaps/pump-loop.log"

alias urchin-looplog="tail -n 100 -F /var/log/openaps/urchin-loop.log"

alias autosens-looplog="tail -n 100 -F /var/log/openaps/autosens-loop.log"

alias cat-pref="cd /root/myopenaps && cat preferences.json"

alias edit-pref="cd /root/myopenaps && nano preferences.json"

alias cat-wifi="cat /etc/wpa_supplicant/wpa_supplicant.conf"

alias edit-wifi="vi /etc/wpa_supplicant/wpa_supplicant.conf"

alias cat-runagain="cd /root/myopenaps && cat oref0-runagain.sh"

alias edit-runagain="cd /root/myopenaps && nano oref0-runagain.sh"

alias cat-autotune="cd /root/myopenaps/autotune && cat autotune_recommendations.log"

alias git-branch="cd $HOME/src/oref0 && git branch"

alias cat-runagain="cd /root/myopenaps && cat oref0-runagain.sh"

alias edit-runagain="cd /root/myopenaps && nano oref0-runagain.sh"

alias edison-battery="cd /root/myopenaps/monitor && cat edison-battery.json"

alias cat-reservoir="cd /root/myopenaps/monitor && cat reservoir.json"

alias stop-cron="cd /root/myopenaps && /etc/init.d/cron stop && killall -g oref0-pump-loop"

alias start-cron="/etc/init.d/cron start"

source /etc/skel/.profile

Abigail Cember
@acember
Feb 06 04:27
@ecc1 Would that essentially be the same as ssh-ing in over USB? I actually tried to set that up, but probably didn't know what I was doing... @scottleibrand @cluckj I'm actually starting to suspect I somehow disallowed ssh-ing into the rig at all. When I try to, I get the "hostname not resolved" error. According to the docs, this "likely means that your rig and computer are not on the same wifi network", but I only have one network and configured only that one on the SD card, and my rig and computer are definitely both online! So there's something else going on.
Scott Leibrand
@scottleibrand
Feb 06 04:59
How do you know the rig is online?
There is a page somewhere in the docs about how to figure out your rig’s IP address so you can ssh to it without relying on hostname resolution.
Stargazer32584
@Stargazer32584
Feb 06 05:21
Hi, @ecc1 and @cluckj thank you for your help. Today, i will configure the /dev/ - Part. But, is there a opportunity to scan the adress of the Adafruit HCW ? And is there a simple Python-Script to test the connection between the rig and the pump ? Thank you !
john36m
@john36m
Feb 06 14:01
Too many things are going wrong on my rig. Segmentation faults, almost impossible to connect wifi. So I decided to reflash. Well, that isn't working either . This is on a mac. When I run the ./flashall.sh command, it asks me to reboot my rig (an edison rig) And what happens is that it just reboots and nothing gets flashed. Any ideas?
Eric
@ecc1
Feb 06 16:33
@Stargazer32584 it's not pre-compiled, but you can cd to the rfm69/cmd/regtest directory (wherever that gets downloaded to) and do "go build", then run the resulting program "./regtest" to test that the radio is wired up correctly. It should print all the radio register values, without warnings about unexpected values
If that works OK, then the "mdt" command can do all the low-level pump operations like read the battery, reservoir level, etc. Just do "mdt" with no args to see the possible commands.
Abigail Cember
@acember
Feb 06 16:58
@scottleibrand I'm basing the "online" assumption on the fact that rig is uploading information to nightscout, and getting my bloodsugar data despite the fact that the receiver is not plugged in (i.e., my loop is operational with no physical connectivity). Is that stuff possible if the rig doesn't have wifi?
When I start up the rig, before it goes dark I get a feed from the operating system where at one point it says "my IP address is: **". So I've been using that (in addition to the hostname I gave it) to try to ssh in -- neither is working. I've also tried with and without the ".local" added.
Oh, also on the "wifi" front: before I did that reinstall where I lost HDMI, I successfully pinged Google and Github to check on the wifi connectivity.
Jon Cluck
@cluckj
Feb 06 17:39
is your hat screen readable enough to see what the IP is by clicking into wifi --> current ip address?
Abigail Cember
@acember
Feb 06 19:27
The screen is probably insufficiently legible to make out numbers, but more importantly, my rig doesn't respond to buttons being pressed :/
Scott Leibrand
@scottleibrand
Feb 06 19:56
Are you on work wifi, or a network you control? I would get your rig set up on a phone hotspot so you can enable that any time you need to log in at work. Many enterprise networks lock down local connectivity and only allow connections out to the Internet.
Abigail Cember
@acember
Feb 06 20:07
On a network I can control (home). At work is a problem I haven't even attempted to solve, because I also don't have cell service there, so phone hotspot is out.
Stargazer32584
@Stargazer32584
Feb 06 21:12
@ecc1 Thank you for your tip. I installed the hole rig again, because of the HCW-Shield. I installed the whole system BEFORE i installed the shield. I think, it will work fine with the new installation and i'am on the right side that the HCW will work. So, i installed the Raspbian Lite and OpenAPS as described in the documentaion. But now, i got some errors when i do the "npm run global-install" - command. It is the following error: npm WARN package.json crypto@0.0.3 crypto is also the name of a node core module. --- npm WARN package.json crypto@0.0.3 No license field. --- npm WARN package.json json@9.0.6 No license field. --- npm WARN package.json share2nightscout-bridge@0.1.6 license should be a valid SPDX license expression. The second error part is the following: npm WARN deprecated crypto@0.03: This package is no longer supported. It's now a built-in Node module. If you've dependet on crypto, you should switch to the one that's built-in. What can i do ? I have no idea at this moment. Can anyone help me ? Thanks a lot...
Scott Leibrand
@scottleibrand
Feb 06 21:20
WARNings are not errors: they can be ignored.
mdhawkins91
@mdhawkins91
Feb 06 22:12

My rig dropped out a dew days ago (battery ran out I think), then I got my pump a bit wet and got a button error that I've rectified.

Cannot get the rig to communicate with the pump again (have been trying for days).
I get this - Mmewolink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type? Listening for 40s silence before mmtuning:

I had this radio error when I was originally setting up openAPS, it did resolve after a couple of days of reboots etc. though I don't know what fixed it. openAPS was running successfully for 4 days or so which was amazing. I really want to get this working again and don't want to give up, but I've really got no other ideas.

I've tried:

  • Rebooting rig (with reboot command in putty, power button, disconnecting battery)
  • Reflashing radio
  • New battery in pump
  • Addind phantom 'devices' to pump (I saw a suggestion this may encourage the pump to communicate via rf)
  • Placing the pump on top of the rig
  • Rerunning openAPS script (both from scratch, and also with retained setting)

I've read troubleshooting docs, but reflashing the radio hasn't worked.
Any other ideas? Pump is getting cgm data from minilink so obviously rf is working fine.

Running edison and explorer, medtronic 754 with minilink cgm