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

6th
Jun 2017
Lin
@denglfddd
Jun 06 2017 00:53
Do I use G4share and need to connect to the rig by cable?
Scott Leibrand
@scottleibrand
Jun 06 2017 01:24
@denglfddd can you rephrase / clarify your question?
Lin
@denglfddd
Jun 06 2017 01:47
@scottleibrand My ns site can not receive cgm readings, I am using G4 share,
Kendra H.S.
@Kendra_HS_twitter
Jun 06 2017 01:52
Hi all - any advice for this problem I'm running into while running Autotune on my Mac. I've deleted all and re-done, and still getting the same error - so not really sure whats going on!
TypeError: Cannot set property 'i' of undefined
at tuneAllTheThings (/usr/local/lib/node_modules/oref0/lib/autotune/index.js:49:32)
at Object.<anonymous> (/usr/local/lib/node_modules/oref0/bin/oref0-autotune-core.js:59:27)
at Module._compile (module.js:569:30)
at Object.Module._extensions..js (module.js:580:10)
at Module.load (module.js:503:32)
at tryModuleLoad (module.js:466:12)
at Function.Module._load (module.js:458:3)
at Function.Module.runMain (module.js:605:10)
at startup (bootstrap_node.js:158:16)
at bootstrap_node.js:575:3
Could not run oref0-autotune-core autotune.1.2017-05-28.json profile.json profile.pump.json
Scott Leibrand
@scottleibrand
Jun 06 2017 01:59
@denglfddd which CGM option did you run oref0-setup with?
Lin
@denglfddd
Jun 06 2017 02:01
@scottleibrand G4
I entered: G4
Scott Leibrand
@scottleibrand
Jun 06 2017 02:13
if you want the rig to upload to Nightscout, you have to select g4-raw (which has been renamed to g4-upload in dev)
Lin
@denglfddd
Jun 06 2017 02:25
@scottleibrand Do I need an otg line for connecting rigs?
Scott Leibrand
@scottleibrand
Jun 06 2017 02:39
yes, if you want to plug in your G4 receiver to your rig and upload data from it, you'll need a mircoUSB OTG male-male cable, or a USB OTG adapter and regular microUSB cable
Lin
@denglfddd
Jun 06 2017 02:51
屏幕快照 2017-06-06 10.51.30.png
Lin
@denglfddd
Jun 06 2017 03:48
@scottleibrand i got this error
glenrob1940
@glenrob1940
Jun 06 2017 04:20
I printed/tried a different enclosure for my edison/explorer rig today which is smaller than my previous one (which was just a rectangular project box). I find that the new enclosure gets pretty hot while charging. The top of the edison chip is exposed in this enclosure. Would a small heatsink (such as the ones we commonly use on Raspberry Pi's help, or should I just drill some holes for airflow, or both, or,...
sounds like the entire inside of the enclosure is getting hot, so without knowing any details on actual temps (from sensors) I would think ventilation would help more than a heatsink
@denglfddd which branch are you running? how up to date is it? maybe run git branch and git show | head -6 ?
glenrob1940
@glenrob1940
Jun 06 2017 04:28
Okay. The bottom (where the lipo is) does not seem as hot as the top, and perhaps with the top of the edison exposed it is radiating enough heat there. To be on the safe side, I'll probably add holes under the Lipo which s hould let some air flow from there and out through the micro usb access slot.
Scott Leibrand
@scottleibrand
Jun 06 2017 04:35
sensors will tell you what the core and skin temps are, which might also give you some clues
glenrob1940
@glenrob1940
Jun 06 2017 04:56
How do I query sensors?
sdneufer
@sdneufer
Jun 06 2017 04:57
type sensors at edison prompt
Scott Leibrand
@scottleibrand
Jun 06 2017 04:58
and apt-get install lm-sensors if you don't have it. I think we added it to the install scripts though.
glenrob1940
@glenrob1940
Jun 06 2017 04:58
I get a bash command not found error
Okay I'll give that a try.
sudo apt-get install lm-sensors
sdneufer
@sdneufer
Jun 06 2017 04:59
I just re-installed minutes ago
glenrob1940
@glenrob1940
Jun 06 2017 04:59
sorry, I clicked on terminal but focus was still in this window
Scott Leibrand
@scottleibrand
Jun 06 2017 04:59
heh. you don't need sudo if you're root, but yeah
glenrob1940
@glenrob1940
Jun 06 2017 05:01
''

root@edison1:~/myopenaps# sensors
coretemp-isa-0000
Adapter: ISA adapter
Core 0: +77.0°C (high = +90.0°C, crit = +90.0°C)
Core 1: +77.0°C (high = +90.0°C, crit = +90.0°C)

skin0-virtual-0
Adapter: Virtual device
temp1: N/A

skin1-virtual-0
Adapter: Virtual device
temp1: +11.9°C

msicdie-virtual-0
Adapter: Virtual device
temp1: N/A

SoC_DTS0-virtual-0
Adapter: Virtual device
temp1: +79.0°C

SoC_DTS1-virtual-0
Adapter: Virtual device
temp1: +77.0°C

root@edison1:~/myopenaps# ''

Apparently okay
Scott Leibrand
@scottleibrand
Jun 06 2017 05:01
77C is a bit warm, but still well below the 90C cutoff
katie disimone
@Kdisimone
Jun 06 2017 05:01
quick question about CGM type
if a person uses G4 and has android with xdrip app for uploading to NS….but they live in rural Montana. Frequently in no-cell area. They would like to plug receiver into rig in those situations and use that for offline looping. What CGM type do they select?
glenrob1940
@glenrob1940
Jun 06 2017 05:03
Thanks, I'll probably still drill the holes, plus I am going to print the vented case from git hub and try it
Scott Leibrand
@scottleibrand
Jun 06 2017 05:03
they'll have to choose between offline with xdrip vs. offline with OTG
apparently android can do permanent BT tethering, so they're both options I think
katie disimone
@Kdisimone
Jun 06 2017 05:04
would they still choose xdrip in the setup script?
if they wanted to still use xdrip app while in wifi area…but have the otg plugged in full time just as a backup when he goes into no cell area
Scott Leibrand
@scottleibrand
Jun 06 2017 05:05
the only way to do that is to have xdrip upload to NS when in wifi area, and just configure the rig for g4-upload or g4-local-only
katie disimone
@Kdisimone
Jun 06 2017 05:05
I’m scratching my head about what the best configuration would be.
Scott Leibrand
@scottleibrand
Jun 06 2017 05:05
you can't get BG data directly from xdrip and directly from the G4 via OTG with the same config
(with the options we currently have in oref0-setup)
Aaron Michelson
@Aaybob
Jun 06 2017 05:06
What is causing this to keep repeating? raise CommsException("Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?")
mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?
Usage: grep [OPTION]... PATTERN [FILE]...
Try 'grep --help' for more information.
Listening: ................
@Kdisimone I think the OTG option requires one to issue the lsusb command to see if that plug in worked and if not to unplug and replug and lsusb again, which if offline you can't do.
Scott Leibrand
@scottleibrand
Jun 06 2017 05:09
@Aaybob if that is your experience, it means some of your hardware is flaky
with a version 1.2 explorer board and a good microUSB OTG cable, you shouldn't have to fiddle with it.
katie disimone
@Kdisimone
Jun 06 2017 05:09
@Aaybob yeah, our OTG hasn’t required any lsusb commands…it’s just worked with plugging in.
Aaron Michelson
@Aaybob
Jun 06 2017 05:09
okay I have new board on order
Tim Street
@tim2000s
Jun 06 2017 05:13
@Kdisimone Via xDrip+? I know there was something via the receiver but wasn't aware that you could pull it direct from transmitter. As I mentioned, no settings for it!
katie disimone
@Kdisimone
Jun 06 2017 05:13
@scottleibrand so if she uses xdrip as the cgm type…that will continue to flow data to NS when he’s in wifi or cell area. and if the OTG receiver is setup, in offline areas, the rig will pull from the OTG port?
glenrob1940
@glenrob1940
Jun 06 2017 05:13
@scottleibrand and @Aaybob I see that every now and then myself and often saw it at initial boot, however it tends to cure itself in about 15 minutes. For the initial boot, I put in a cron job to run at boot resetting the Ti chip, sleeping for a minute and then running mmtune - which cured that problem. This with a brand new explorer board received a couple of weeks ago. My guess is that with autotune and most everything runinng in cron there is some kind of a race condition that puts the Ti chip out to lunch
Scott Leibrand
@scottleibrand
Jun 06 2017 05:14
@Kdisimone the config I'm suggesting is: xdrip uploads to NS. rig is configured to pull from receiver via OTG (and uses NS if available).
katie disimone
@Kdisimone
Jun 06 2017 05:15
@scottleibrand I guess i have my head spun over the CGM type in the setup script. That is specifically for offline situations, isn’t it? I don’t think I ever thought about this so much.
glenrob1940
@glenrob1940
Jun 06 2017 05:15
When is 'out to lunch' it runs mmtune and returns -99 for strength until it eventually gets well.
Scott Leibrand
@scottleibrand
Jun 06 2017 05:15
in the config I'm suggesting, CGM type could be g4-upload or g4-local-only
katie disimone
@Kdisimone
Jun 06 2017 05:16
yup…i think i’m following now.
slowly, i’m joining the flow
Aaron Michelson
@Aaybob
Jun 06 2017 05:49
@glenrob1940 would you write out for me the cron code you run at boot resetting Ti then running mmtune
DaisyHK
@laurathiessen
Jun 06 2017 06:07
says its been 4 hs since openaps has worked on my site here is the papertrail if anyone can help me out
image.png
image.png
however... i am not sure but I think my battery may have died as there was no red light I think when I hooked my my rig to my computer to start putty... but now even after over 2 hours of it attached to a power source the red light on the left hand side not the middle does not stay on when i disconnect it from the computer....
Scott Leibrand
@scottleibrand
Jun 06 2017 06:12
the red light in the corner indicates the battery is charging
DaisyHK
@laurathiessen
Jun 06 2017 06:14
oh lol ok nevermind then
about the light
Scott Leibrand
@scottleibrand
Jun 06 2017 06:15
it looks like there's a problem getting BG data, but I'm not sure exactly
the 'Cannot read property "date" of undefined' thing is probably your best clue in those logs
but you may need to log in and run some aliases and reports manually, per the troubleshooting section of the docs, to figure it out
glenrob1940
@glenrob1940
Jun 06 2017 06:16
Actually my memory was suspect. I didn't include mmtune in the current shell script, because it will happen normally from the pump loop (other line in cron) Here's the crontab lines and the shell script

Lines from crontab
‘’

* oref0-radio-reboot

@reboot /usr/local/bin/glenfoo.sh
‘’

root@edison1:~# cat /usr/local/bin/glenfoo.sh

!/bin/bash

/bin/sleep 60
/usr/bin/killall -g openaps >& /dev/null
/usr/local/bin/reset_spi_serial.py
root@edison1:~#

That's what is currently working for me.
Scott Leibrand
@scottleibrand
Jun 06 2017 06:17
you want three backticks (`) before and after, not single quotes (')
backticks are on the ~ key
glenrob1940
@glenrob1940
Jun 06 2017 06:18
Right, sorry it's getting late.
Scott Leibrand
@scottleibrand
Jun 06 2017 06:18
if you're on a desktop browser you can edit your old posts by mousing over the invisible ... to the right of the timestamp
DaisyHK
@laurathiessen
Jun 06 2017 06:26
@scottleibrand ok that looks tricky to try that... I will try this out tomorrow then and come here for some hel as i have to go to bed thanks!
Scott Leibrand
@scottleibrand
Jun 06 2017 06:28
sounds good. or maybe someone else has seen that issue recently and will have some tips to try
Tim Street
@tim2000s
Jun 06 2017 06:29
@Kdisimone I've put a direct question on the fb group. @scottleibrand pulled down new dev last night for the Autotune fix. Is the bolus snooze change a preference or automatic?
Scott Leibrand
@scottleibrand
Jun 06 2017 06:30
SMB bolus snooze is a change in the default behavior. it will honor the regular bolus snooze preferences, though, and be more likely to activate if remainingCarbsCap and remainingCarbsFraction are higher
Tim Street
@tim2000s
Jun 06 2017 06:31
Okay thanks. Makes sense.
glenrob1940
@glenrob1940
Jun 06 2017 06:34
Tried editing the post but the edit button was grayed out
ijustlovemath
@ijustlovemath
Jun 06 2017 06:35
@thebookins Thanks! I'll again have to port it to LabVIEW, but it's nice to have it all in one place
Paul Dickens
@thebookins
Jun 06 2017 06:36
:+1:
PieterGit
@PieterGit
Jun 06 2017 07:28
@Aaybob @glenrob1940 @scottleibrand : I also experience mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?
with my newer explorer boards. Sometimes the problem fixes itselves because a succesfull mmtune, but sometimes that takes quite long. I noticed if I stop my pump loop and run a mmtune.py from the command line it's fixed.
I think the rootcause is in mmeowlink, see oskarpearson/mmeowlink#60
I have a WW-pump. Can it be that it occurs more with WW-pumps?
I will try @glenrob1940 script. I added a mmtune.py --radio_type subg_rfspy --radio_locale WW --port /dev/spidev5.1 --serial 123456 to force a mmtune during bootup.
PieterGit
@PieterGit
Jun 06 2017 08:02
@glenrob1940 : two minor issues with your script. if you wait 60 seconds after boot, all the cronjobs have already been started. your fix script does not work for SMB loops. I now test with:
#!/bin/bash
date >> /var/log/openaps/mmtune.log
/usr/bin/killall -g openaps >& /dev/null
/usr/bin/killall -g oref0-pump-loop >& /dev/null
/usr/bin/killall -g oref0 >& /dev/null
sleep 3
/usr/local/bin/reset_spi_serial.py
mmtune.py --radio_type subg_rfspy --radio_locale WW --port /dev/spidev5.1 --serial 123456 >> /var/log/openaps/mmtune.log
Tim Street
@tim2000s
Jun 06 2017 08:18
@Kdisimone we've some answers on the xDrip G5 group. The standard xDrip G5 display is raw. However experience says that you don't get any raw data from xDrip that can be used while switching over a sensor until you have done the double calibration, so I think it's a bit of a moot point.
Lin
@denglfddd
Jun 06 2017 09:51
@scottleibrand root@edisonhost:~/src/oref0# git branch
  • master
fidoman
@efidoman
Jun 06 2017 11:45
Another great experience with the latest SMB. Dinner last night was 4 large tacos and only a single 0.7 manual bolus about an hour after the meal and SMB took care of the rest while keeping me in range. Thank you Scott, Dana, and others who pitched in to make the upcoming release (current dev) version so stable with innovative new advanced features.
image.png
Matthias
@ceben80
Jun 06 2017 12:58
:+1:
rsilvers129
@rsilvers129
Jun 06 2017 13:02
So, this is the third time my rig was off for no known reason in the middle of the night.
I woke up to the Dexcom alarm and I was at 45 BG. My pump was at the default basal of around 0.850 units/hr. The rig failed to suspend my basal due to being off. It was plugged in and charging (charging LED was on), but the power light was off.
I unplugged it from USB power, plugged it back in, and it powered up on its own. I looked at the loop log, and there was some text typical of a radio error. Not sure if I needed to, but I rebooted it after seeing that (but it may have worked had I not rebooted it). It started up and worked, and the battery was nearly full of charge (at 78%).
Is there a specific log I can look at to see what may have happened to make it shut off?
IMG_6498.PNG
IMG_6499.PNG
@ericfeibelman How many grams of carb and fat? Last night I had 124 grams of carb in pizza with 64 grams of protein and 72 gramds of fat. Highest I went was 170 with oref0. I am going to re-try it today with a second manual bolus rather than let the rig do it, and see if I can get lower. (This was 1/2 of a large Papa Ginos Sausage Pizza, by the way).
fidoman
@efidoman
Jun 06 2017 13:18
@rsilvers129 - I put in 60 total grams of carbs but I actually ate about 90 grams of total carbs and the meat and cheese I would guestimate around 30 grams of protein and 35 grams of fat (just a guess though). I never really account for protein and fat except to remember that it might cause the carb absorption to take much longer, thus no significant upfront bolus.
@rsilvers129 - I have a cron task to reboot my rig every night at around 1am. Not sure if this would help or not in your case.
I also have a script that will reboot my rig if it recognizes that openaps hasn't enacted anything in greater than x minutes. In my case it is 30 minutes. Then, it will not reboot if uptime is less than x hours ... in my case x is 1 hour. I've only seen this reboot happen a couple of times but it seems to have helped me "unstick" my rig a couple of times.
Lin
@denglfddd
Jun 06 2017 13:24
hi,all,How to remove all the data from the rig, I received a prompt "no enough space"
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 13:28
@rsilvers129 my rig use to lock up during the night because of heat issues while charging, I have even melted two cases because of the same issue. Now I remove it from my case when charging.
Annette Valle
@annettemvalle
Jun 06 2017 13:32

I'm seeing this on my log and no temp basals are enacted. Is it a proximity issue? Just got up and going yesterday.

"Jun 06 2017 07:22:55 GMT-0600 (MDT)"}
{"reason":"BG data is too old (it's probably this), or clock set incorrectly. The last BG data was read at Tue Jun 06 2017 02:42:21 GMT-0600 (MDT) but your system time currently is Tue Jun 06 2017 07:22:55 GMT-0600 (MDT)"}
Waiting for ntpd to synchronize... No!
ntpd did not synchronize."

Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 13:53
@annettemvalle are you using the dev or the normal version of oref0?
Annette Valle
@annettemvalle
Jun 06 2017 13:53
Regular
(Normal)
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 13:54
that issue should fix itself when it is not able to get the info from an NTP server it pulls the time from the pump clock
Thank @scottleibrand from merging that fix into the master
Annette Valle
@annettemvalle
Jun 06 2017 13:55
Ok great. How long does it take? It's been about 12 hours...
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 13:55
ouch
that is not good
about 5 minutes
Annette Valle
@annettemvalle
Jun 06 2017 13:55
Yay I'm getting s little frustrated...
Anything else I can do to hurry it up?
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 13:56
are you using an edision?
Annette Valle
@annettemvalle
Jun 06 2017 13:56
Yes
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 13:56
did you confirm you are currently connected to the internet ?
Annette Valle
@annettemvalle
Jun 06 2017 13:57
Yes I must be because I can visualize everything on termius
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 13:58
you may want to try pinging google
in a terminal connected to your edison
Annette Valle
@annettemvalle
Jun 06 2017 13:58
Ok
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 13:59
it should resolve to ip and keep going
you will have to hold down ctrl + c to get it to stop
Annette Valle
@annettemvalle
Jun 06 2017 14:01
Yes it seems to be connected
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:02
then it is time to watch the logs
tail -F /var/log/openaps/pump-loop.log
wait
check your pump clock and the ns site also
make sure the time is the same on both of them
Annette Valle
@annettemvalle
Jun 06 2017 14:04
@Diabetic_Dude Ok I'll keep watching. It looks pretty similar to what I originally posted however
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:05
is the time the same on your NS site and your pump?
Annette Valle
@annettemvalle
Jun 06 2017 14:08
Yes the times are the same
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:09
@annettemvalle also how are you getting to the internet? your phone? (if it is your phone if so BT or wireless) or are you using home or work wifi
Annette Valle
@annettemvalle
Jun 06 2017 14:12
Home wifi
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:13
has this setup ever worked for you before?
Annette Valle
@annettemvalle
Jun 06 2017 14:14
I tried 6 months ago with a pi and had similar issues but didn't have the time to put in to really figure it out and had to quit (I never learned to watch the logs well).
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:16
if I was you, I would re run the setup script
and check back here if/when you have any more problems
or you can wait for someone else to help you
Annette Valle
@annettemvalle
Jun 06 2017 14:17
Sounds good, thank you so much for your help!
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:18
keep with it, you are almost there, I had problems like you are having now for like a month
then found out it was my work network causing the issue
Annette Valle
@annettemvalle
Jun 06 2017 14:19
Oh wow, so do you connect to your phone?
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:20
I connect using my home network, and when I leave my edison connecting to my phone using Bluetooth
if that fails I open my wifihotspot and allow my edison to connect to that
Annette Valle
@annettemvalle
Jun 06 2017 14:20
And thanks, I'm feeling ready to quit again but I' really want this to work!
How does that affect your data on your phone?
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:21
naw like I said I had issues for a month, this is a known issue
your job is to get help figuring out how to get around it
it really does not affect my data on my phone
never give up
Annette Valle
@annettemvalle
Jun 06 2017 14:21
Ok that's good to hear
Thanks, I'll get there. It's just finding the time as I'm sure you know
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:22
but remember I have a 15 gb a month limit
Lin
@denglfddd
Jun 06 2017 14:22
@diabetic_dude_twitter hi,How to delete all the data for the rig, I would like to reinstall openaps
Annette Valle
@annettemvalle
Jun 06 2017 14:22
I have 2 GB haha
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:23
you are right, I have two 5 year old children and old children and one 4 year old and still find the time for this
@annettemvalle WOAH who do you use for your phone ?
@denglfddd gimme a sec I will find the command but this is why I have not answered your question
I know that you should delete the myopenaps
but I am not sure if you should delete the src folder
Annette Valle
@annettemvalle
Jun 06 2017 14:25
@diabetic_dude_twitter I have a 2 year old and a 3 month old that don't give me much time...(I stayed up til 4am 2 nights ago to get this thing going). I have a family plan with AT&T and have never needed more than that
Lin
@denglfddd
Jun 06 2017 14:26
thank you very much for your help @diabetic_dude_twitter
Annette Valle
@annettemvalle
Jun 06 2017 14:27
Thanks for the help
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:28
@denglfddd try "rm -r myopenaps" (caution doing this will remove your myopenaps directory)
but we both need do confirm weather you need to run "rm -r src" from your home directory also
(I think you do)
@annettemvalle I switched from ATT to cricked and never looked back, they use the same towers as ATT but I get 15 GB per family member and am saving over 50% on my monthly bill
excuse me cricket is who I use.
I will be back ( I have to go do the work I get paid to do :) )
Lin
@denglfddd
Jun 06 2017 14:33
@diabetic_dude_twitter I would like to ask how can I enter what you said "home directory"
Annette Valle
@annettemvalle
Jun 06 2017 14:33
I'll have to look into that. Thanks again for all your help!
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:36
@denglfddd that is the directory where all your files are stored, if you type "ls" and you are in your root folder, you should see things like remote_syslog, myopenaps, src, crontab.txt
excuse me your home folder not your root folder
@denglfddd what do you see when you type "pwd"
(that should tell you what directory you are in)
Lin
@denglfddd
Jun 06 2017 14:39
@diabetic_dude_twitter root@localhost:~# pwd
/root
root@localhost:~#
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:39
ok you are logged in as root and your are in your home directory
Lin
@denglfddd
Jun 06 2017 14:40
I have landed
logged
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:40
so if I am not mistaken you can type "rm -r src" to remove the src and then "rm -r myopenaps" to remove the myopenaps folder
but then you have to start all over (IE the nuclear option)
Lin
@denglfddd
Jun 06 2017 14:42
root@localhost:~# rm -r src
root@localhost:~# rm -r myopenaps
rm: cannot remove ‘myopenaps’: No such file or directory
Does not seem to respond
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:42
type "ls" and tell me if you see myopenaps
Lin
@denglfddd
Jun 06 2017 14:43
root@localhost:~# ls
crontab.old.txt remote_syslog_linux_i386.tar.gz.2 save
crontab.txt remote_syslog_linux_i386.tar.gz.3 (save)
kernel.config remote_syslog_linux_i386.tar.gz.4 wget-log
remote_syslog_linux_i386.tar.gz remote_syslog_linux_i386.tar.gz.5
remote_syslog_linux_i386.tar.gz.1 remote_syslog_linux_i386.tar.gz.6
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:44
looks like it is gone
now type "df" and see how much space you have
Lin
@denglfddd
Jun 06 2017 14:47
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/root 1441648 1351752 0 100% /
devtmpfs 491400 0 491400 0% /dev
tmpfs 491704 0 491704 0% /dev/shm
tmpfs 491704 19080 472624 4% /run
tmpfs 5120 0 5120 0% /run/lock
tmpfs 491704 0 491704 0% /sys/fs/cgroup
tmpfs 491704 4 491700 1% /tmp
/dev/mmcblk0p10 1337936 808664 512888 62% /home
/dev/mmcblk0p7 32686 5234 27452 17% /boot
tmpfs 98344 0 98344 0% /run/user/0
屏幕快照 2017-06-06 22.47.34.png
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:48
wow you are still full
Lin
@denglfddd
Jun 06 2017 14:48
yes.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:48
to find your top 10 largest directories type
"find . -type d -print0 | xargs -0 du | sort -n | tail -10 | cut -f2 | xargs -I{} du -sh {}"
to find your top 10 largest files type
"find . -type f -print0 | xargs -0 du | sort -n | tail -10 | cut -f2 | xargs -I{} du -sh {}"
Lin
@denglfddd
Jun 06 2017 14:49
320K ./.npm/lodash
320K ./.npm/registry.npmjs.org/yargs
500K ./.npm/moment/2.18.1
504K ./.npm/moment
508K ./.npm/ajv/4.11.8
512K ./.npm/ajv
560K ./.npm/registry.npmjs.org/ajv
4.7M ./.npm/registry.npmjs.org
9.9M ./.npm
19M .
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:50
what is your biggest one?
wow only 19 mb
are you using an edison?
Lin
@denglfddd
Jun 06 2017 14:51
yes,edison
I think I should be all redo
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 14:56
I am not sure
looks like your problem is in your actual linux install
which means you would have to reflash it
I personally would want to find what is taking up the space so it will not happen again, and if it does I would know how to fix it
@denglfddd try this "rm -rf /var/log/journal/"
I have to go but please confirm if you tried it and if it worked
rsilvers129
@rsilvers129
Jun 06 2017 15:02
@diabetic_dude_twitter My case has the Edison heat-sink exposed. But maybe it did hit 90C and reset. I can check that. Thanks. Every time it happened it was plugged in.
Lin
@denglfddd
Jun 06 2017 15:03
root@localhost:~# rm -rf /var/log/journal/
root@localhost:~# rm -rf /var/log/journal
root@localhost:~# rm -rf /var/log/journal/*
root@localhost:~# rm -rf /var/log/journal/
root@localhost:~#
@diabetic_dude_twitter Did not seem to react
rsilvers129
@rsilvers129
Jun 06 2017 15:04
@ericfeibelman Thanks for the script ideas. As for fat when eating pizza - there was a study that my former endo did. They compared meals with 10 grams of fat, and 60 grams of fat. The meals had the same number of carbs. The high-fat meals needed not just insulin over a longer period of time, but also 42% more insulin! So do a second bolus 1.5-3 hours after the first, to account for this additional need. Bolus should be for the grams of fat divided by two, or so.
Try this Excel sheet.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 15:05
@denglfddd now run that other command again
"find . -type d -print0 | xargs -0 du | sort -n | tail -10 | cut -f2 | xargs -I{} du -sh {}"
see if it made a difference
hopefully root is no longer at 100% anymore
rsilvers129
@rsilvers129
Jun 06 2017 15:05
Protein has a much smaller effect, although I have seen people who like to account for protein but don’t account for fat. I don’t think there is science to support that protein is as significant as fat.
Lin
@denglfddd
Jun 06 2017 15:06
root@localhost:~# find . -type d -print0 | xargs -0 du | sort -n | tail -10 | cut -f2 | xargs -I{} du -sh {}
320K ./.npm/lodash
320K ./.npm/registry.npmjs.org/yargs
500K ./.npm/moment/2.18.1
504K ./.npm/moment
508K ./.npm/ajv/4.11.8
512K ./.npm/ajv
560K ./.npm/registry.npmjs.org/ajv
4.7M ./.npm/registry.npmjs.org
9.9M ./.npm
19M .
no change
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 15:07
try "df"
Lin
@denglfddd
Jun 06 2017 15:09
100% 😅
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 15:10
:fire:
:(
@denglfddd you may want to wait for someone else to help you
or reflash
Lin
@denglfddd
Jun 06 2017 15:12
okay,reflash
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 15:12
at this point it may be faster depending on how many times you have done it in the past,
Lin
@denglfddd
Jun 06 2017 15:12
thank you `
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 15:13
anytime (sorry my terminal kungfu is not better)
fidoman
@efidoman
Jun 06 2017 15:14
@rsilvers129 - :+1:
Lin
@denglfddd
Jun 06 2017 15:14
Do not worry, you're much more powerful than me😎
dan lightner
@danimaniac
Jun 06 2017 15:22
Last night one of my local television stations ran a brief story stating the first patient in central Illinois is now officially using the MiniMed 670G "hybrid closed loop system"...SMH. I sent them a brief email this morning stating #WeAreNotWaiting and we have been building these systems at home for a while now. Thanks @danamlewis @scottleibrand @others...love you guys...yeah, that sounds a little weird.
Lin
@denglfddd
Jun 06 2017 15:23
@diabetic_dude_twitter look this
屏幕快照 2017-06-06 23.21.03.png
@danimaniac 👍
Dana Lewis
@danamlewis
Jun 06 2017 15:29
@danimaniac :heart:
Lin
@denglfddd
Jun 06 2017 15:55
I tried to flash again, install openaps, but stuck in this step:Step 2: Run oref0-setup
root@edisonhost:~/src# bash ~/myopenaps/oref0-runagain.sh
bash: /root/openaps/oref0-runagain.sh: No such file or directory
Dana Lewis
@danamlewis
Jun 06 2017 16:03
You deleted so you'll need to start from scratch, rather than run-again
Lin
@denglfddd
Jun 06 2017 16:05
Yes, I did start from flashing to Edison
Dana Lewis
@danamlewis
Jun 06 2017 16:07
Right - but instead of bash run again, you'll need to run oref0-setup from scratch, too
Scott Leibrand
@scottleibrand
Jun 06 2017 16:08
@diabetic_dude_twitter you were having @denglfddd run those commands from his homedir, so they were missing everything outside it. If you had done a ‘cd /` first they would've seen everything.
Lin
@denglfddd
Jun 06 2017 16:11
?I am confused
Scott Leibrand
@scottleibrand
Jun 06 2017 16:11
@denglfddd if you've reflashed, my help is too late: you've now deleted whatever was filling up the disk and can continue with re-running oref0-setup
I was just telling @diabetic_dude_twitter one thing I noticed, which might be why his commands weren't finding the culprit
and @diabetic_dude_twitter thanks for taking the night shift and doing what you can to pay it forward and help people out. :)
Lin
@denglfddd
Jun 06 2017 16:13
屏幕快照 2017-06-07 00.13.14.png
Scott Leibrand
@scottleibrand
Jun 06 2017 16:14
@denglfddd you can't use oref0-runagain. you can't run it "again" because you deleted the information about the previous run off the rig already.
you have to use oref0-setup like the very first time
Lin
@denglfddd
Jun 06 2017 16:17
thank you @scottleibrand
@scottleibrand What kind of CGM are you using? (i.e. G4, ShareBLE, G4-raw, G5, MDT, xdrip)
This question is the same as before
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 16:27
Ok I am back @denglfddd how are you holding up? I see you are getting help from the best currently so you should be in good hands
Lin
@denglfddd
Jun 06 2017 16:28
@diabetic_dude_twitter thanks, I have been refreshed again。
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 16:29
sweet
are you up and working?
Lin
@denglfddd
Jun 06 2017 16:29
I have a question, what is the difference between G4 and G4 -raw?
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 16:29
well,
I think g4 raw pulls the raw data which some say is more precise
Lin
@denglfddd
Jun 06 2017 16:30
@diabetic_dude_twitter 12:30 am🙃
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 16:31
woah getting late?
did you choose g4 or g4 raw?
Lin
@denglfddd
Jun 06 2017 16:32
If you use G4 instead of G4-raw, can the rig read the receiver's BG value?
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 16:33
it will pull the data from your NS site
it will pull the filtered data not the raw data (it was the same option I used when I had a G4)
Lin
@denglfddd
Jun 06 2017 16:35
Does the NS site pull data from DEXCOM SHARE?
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 16:35
in my experience the raw data and the regular g4 data only had small differences (but I only used if for about 2 months)
that depinds on how you set it up
it can
do you have an android or iphone
Lin
@denglfddd
Jun 06 2017 16:36
iphone
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 16:36
goog so do I
oops good so do i
if you set up your nightscout site to read your share data it will pull it from there filtered
Lin
@denglfddd
Jun 06 2017 16:38
But yesterday I set the G4, NS to this morning did not show BG data
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 16:38
what about today?
sorry man I have to go to lunch. I will be back in about 1 hour
if you are not seeing your BGs on night scout ask here
Lin
@denglfddd
Jun 06 2017 16:40
@diabetic_dude_twitter ok ··bye
Scott Leibrand
@scottleibrand
Jun 06 2017 16:52
the main difference between g4 and g4-raw is whether the rig will upload the collected data to NS or just use it locally
for that reason, we renamed them to g4-local-only and g4-upload in the newer oref0 version in dev
Lin
@denglfddd
Jun 06 2017 16:56
@scottleibrand understand,✌️
Would you like to remove your existing crontab first? y/[N]
Scott Leibrand
@scottleibrand
Jun 06 2017 16:56
safe to say yes, as you shouldn't have one
Dana Lewis
@danamlewis
Jun 06 2017 17:09
@jenm999 :+1:
Dana Lewis
@danamlewis
Jun 06 2017 17:49
I posted this on Fbk but wanted to post here, too - so thankful for everyone who helps out, answers questions, troubleshoots, edits the docs, etc. I can’t give you a hug, so instead...
Internet High Five.jpeg
Matthias
@ceben80
Jun 06 2017 18:15
Yeah @danamlewis ...so cute :cat2:
JELCRAWFORD
@JELCRAWFORD
Jun 06 2017 18:17
@annettemvalle - our son runs on cell while at school - switching back and forth to school wifi was proving too difficult - he and I share a 6g data plan - When at home he is on wifi. If he doesn't download PokemonGo updates (or the like) on cellular, we usually stay just under 2g of data. All bets are off if he or his brother (who is often tethered when with us) decide downloading game updates or watching youtube are a good idea on the way to school :)
hilarykoch
@hilarykoch
Jun 06 2017 18:18
For showers, we haven't been suspending. Should we? And for an hour at the beach... suspend or temp basal zero?
Dana Lewis
@danamlewis
Jun 06 2017 18:18
temp basal to zero and suspend.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 18:20
how do most use their pump at the beach? me I just leave it on and put it in my pocket.
If I go in the water I suspend, but normally that isnt longer than 15 mins and I am ready to get out
hilarykoch
@hilarykoch
Jun 06 2017 18:21
@danamlewis Thank you!
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 18:22
and to be clear I do take it off before going in the water especially salt water
hilarykoch
@hilarykoch
Jun 06 2017 18:25
@diabetic_dude_twitter We've never really spent a lot of time in the water or at the beach. We've typically just disconnected for water-play for maybe 30 minutes. Anything else (playing in sand, just getting feet wet, pump stays on.) Today will be a bit of an experiment.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 18:28
best wishes out there. My only challenge is always the heat.
rsilvers129
@rsilvers129
Jun 06 2017 18:29

So my hardware failures have been temp related when charging. This is even with my enclosure that exposes the Edison to the air. And even when sitting on my nightstand with room temp around 68F/20C. Type this and see if you have temp warnings:

grep "temperature" /var/log/syslog

Dana Lewis
@danamlewis
Jun 06 2017 18:32
@hilarykoch :+1: I always end up off longer than expected, so I usually set a long temp zero (say, 2 hours if I think I’m going to be off about an hour) and suspend. Ditto for showers if they end up long, might throw on an hour of zero instead of thirty, to cover the time it takes to get dressed and re-attach the pump and unsuspend. It’s probably minor in the minutes and the impact of the netIOB calculation, but might be more important for tinier basal’ers.
most important: have fun! :)
:palm_tree: :sunrise:
@diabetic_dude_twitter same, re: leaving it on for the most part unless I’m going into the water. the only time I worried about that was holding my sand-covered nephew on my hip and worrying about wet sand getting all over my pump.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 18:41
lol yup my weak points are the same, playing with my kids
JELCRAWFORD
@JELCRAWFORD
Jun 06 2017 19:12
@hilarykoch We like the beach and usually spend good portions of a day there when we have a chance. We have a waterproof bag (and have used a new ziplock) to keep devices in so the sand and water stays out. We have a waterproof pump bag, but really don't need it -we find our 10yr old is active enough, and wants to eat often enough that we just disconnect (with the temp zero and suspend strategy outlined above) - when he snacks we bolus and then adjust at that point, as needed. In the pool, we found that he was more likely to trend low rather than high, even if he'd been unplugged for hours. Of course, everyone is different though! Given the total amount of time our rig and pump spent in the bag when we were at the beach this past winter (yay Hawaii!) we are thinking that next time we will take the new waterproof 630g when we head out to a water/beach 'date' and leave the rig back at home/condo/hotel. Then there is less worry about wrecking the really important things, and the waterproof bag is not so necessary (don't know if anyone has run into sand troubles with their pump tho). Then we'll hook back up to the full rig when we get home. Wet Dexcom cgm sites also do not play very nice and can be a consideration with the whole rig...
Dana Lewis
@danamlewis
Jun 06 2017 19:14
FYI to all - we just got IRB approval via Johns Hopkins to do a more detailed retrospective outcomes study. We want to give people time to upload their data to the OpenAPS Data Commons, so if you’re interested in your data being included in the study, please follow the directions for uploading your data (https://openaps.org/data-commons/). I’ll do the first pull from the Data Commons for the study after June 16. (Feel free to PM or email me if you have any questions about the study or have issues uploading your data. Also note I had Open Humans add an anonymous messaging feature to the platform, so if you’ve joined the data commons, you can also anonymously message using the platform)
garykidd
@garykidd
Jun 06 2017 19:27
@danamlewis cool :clap:
garykidd
@garykidd
Jun 06 2017 19:37
thinking this would be great but until it's around..pushover alert for more carbs is not too shabby...
rsilvers129
@rsilvers129
Jun 06 2017 19:49
@danamlewis How would you compare the RF range of your TI-stick based systems to the Explorer? While my Explorer sometimes works a dozen feet, it also sometimes doesn’t work under 1 meter away. Do you find that the TI stick has a much farther range? I am wondering if I should buy a stick for an unused-PI I have, and leave it running in my house plugged in.
Scott Leibrand
@scottleibrand
Jun 06 2017 19:52
we've found they're pretty similar. the "doesn't work under a meter away" is atypical.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 19:54
how can I test the rage of my explorer board?? I did a modified way of scraping the antenna (different from what the instructions say) and my rage is FAR.. I want to test the range in my house if it is as good as I think I want to modify the instructions and spread the wealth
yesterday for example I had my openaps charging in the kitchen, and it changed my pump across the kitchen, through the living room, through our bedroom, and into the bathroom
rsilvers129
@rsilvers129
Jun 06 2017 19:57
James C posted on Facebook "The TI stick has far batter range. I can be on any floor of my house and in any room including one with thick ceramic tile floor and walls and never need to move the pi. I still have strong signal. "
It could be my pump. I could range test my spare pump. I know my friend’s setup didn’t work well with one pump, and it did with his other one.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 20:00
did you scrape the explorer board antenna to increase the range?
rsilvers129
@rsilvers129
Jun 06 2017 20:00
Yes. That helped get me to where I am now.
Scott Leibrand
@scottleibrand
Jun 06 2017 20:00
yeah, we've seen differences between pumps too. never really dug into details though.
rsilvers129
@rsilvers129
Jun 06 2017 20:00
Mine works one meter away. Just not always.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 20:00
I have a 723
rsilvers129
@rsilvers129
Jun 06 2017 20:01
It is not the model of pump. It is individual pump variations, I think.
I have 723 and 722s
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 20:01
wow that reminds me of my carelink days.. I want to test a TI but the main thing that is holding me back is getting a debugger http://www.ti.com/tool/cc-debugger
rsilvers129
@rsilvers129
Jun 06 2017 20:01
It could even be my battery affecting the radiation pattern.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 20:02
I am going to see if I can find an old picture of what my scrape looks like
rsilvers129
@rsilvers129
Jun 06 2017 20:03
If I want to change the serial number, must I rerun setup, or is there some file I can edit?
Scott Leibrand
@scottleibrand
Jun 06 2017 20:06
it's in multiple places, so easiest to edit runagain and run that
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 20:08
0DA54EAB-11F9-4A45-BB51-B7EB8233BF0C.JPG
that is my scrape, before it did that, I was not able to get to the other side of a room,
live4sw
@live4sw
Jun 06 2017 20:15
I have had noticeably worse connectivity with the Explorer than with my TI stick based rigs. It's still way better than the Carelink stick, but the range is disappointing. I scraped one of my Explorers, but didn't really see a significant improvement, so haven't done it with my others yet.
@diabetic_dude_twitter I am a little confused, as your scrape appears to be parallel to or on the line, isn't it supposed to be perpendicular to the line, just to the left of the "915" so as to cut off any current beyond that?
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 20:16
yes it is
i did mine different but it seems to work very well
that is also why I did not add my picture to the docs yet, that is also why I want a way to officially test my range
live4sw
@live4sw
Jun 06 2017 20:19
Oh ok, sure.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 20:28
I screwed up when I read the instructions, but I have been very happy.
the way I read the instructions was to make a clean cut to the outer coating (not the actual wire)
garykidd
@garykidd
Jun 06 2017 20:29
@diabetic_dude_twitter I can flash TI for you if you send it my way---or if you want---send me a self -addressed prepaid envelope and I'll let you borrow it.
debugger that is
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 20:30
@garykidd I will take you up on that as soon as I get one.
garykidd
@garykidd
Jun 06 2017 20:31
a TI stick?
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 20:31
yes
garykidd
@garykidd
Jun 06 2017 20:31
OK; make me an offer--Ihavea couple I'm not using.
Diabetic_Dude
@diabetic_dude_twitter
Jun 06 2017 20:32
to purchase a ti stick?
pre flashed?
garykidd
@garykidd
Jun 06 2017 20:32
honestly, my expereince was that the range on ti stick was the best...really impressive
sure--both are flashed--but should be able to flash whatever you want.
fidoman
@efidoman
Jun 06 2017 20:42
@scottleibrand - any plans for carbsRequired pushover alerts going out as an ns-event?
Marco
@CaptainBalou
Jun 06 2017 20:43
My assurance company confirmed after one day that they will pay for the CGM. Love to hear that. BUT they want me to use a 640G. Aaaaaaaaah! Currently thinking about how to argument for the older pump which is out of warranty and which is not produced anymore...which cannot get swaped in case of a damage... OMG. :-)
garykidd
@garykidd
Jun 06 2017 20:45
YOu should be able to use MM sensors on a loopable pump...take the 640 as a backup and look to score a loopable pump somewhere
your insurance company isnt going to pay you for an out of warranty pump.
Scott Leibrand
@scottleibrand
Jun 06 2017 20:47
@ericfeibelman No, because @danamlewis and I have never used NS alarms and don't intend to. Someone else would have to take that on.
Dana Lewis
@danamlewis
Jun 06 2017 20:48
(if you want to have fun/be mean, watch me twitch when I’m in a room with other PWDs and everyone’s pumps are boop-beeping and CGMs are alarming and NS is going off and OMG THE NOISE!! :))
garykidd
@garykidd
Jun 06 2017 20:48
I just learned to tune my alarms out (unfortunately, I'm sort of not joking).
Cameron Chunn
@applehat
Jun 06 2017 20:49
I have NS on a tablet in my room, and the volume is set to 0
I hate that alarm
I use the "why am I shaking?" alert method for hypos.
garykidd
@garykidd
Jun 06 2017 20:49
+1 I like that alarm almost as much as my usual daily alarm.
@scottleibrand what are the pushover token and key entries for in latest script--figured that actually already was for NS.
Scott Leibrand
@scottleibrand
Jun 06 2017 20:52
no, oref0 directly calls the pushover API using your pushover keys
garykidd
@garykidd
Jun 06 2017 20:55
and if and when oref0 directly calls the pushover API using your pushover key....what exactly happens?
Scott Leibrand
@scottleibrand
Jun 06 2017 20:55
it sends a push notification to your pushover app on your phone
garykidd
@garykidd
Jun 06 2017 20:56
got it....I guess that should of been a clear assumption on my part.
and no separate app is needed?
on phone?
Scott Leibrand
@scottleibrand
Jun 06 2017 20:58
if that's not clear in the doc Dana linked, please PR
pushover is a separate app, which can run on any/all of your phones/tablets/browsers
or, not sure what you mean
garykidd
@garykidd
Jun 06 2017 20:59
Probably clear for almost everybody....just not knuckleheads resembling myself
Scott Leibrand
@scottleibrand
Jun 06 2017 20:59
if it's not clear to you, it won't be clear to someone else either
garykidd
@garykidd
Jun 06 2017 20:59
nope---that's exactly what I meant
I will PR then
I almost made you feel obligated to say I'm not a knucklehead didn't I...wait don't answer that....
Scott Leibrand
@scottleibrand
Jun 06 2017 21:01
I generally ignore inane comments like that. ;-P
garykidd
@garykidd
Jun 06 2017 21:03
wait looking up inane now....got it...yes stupid...wait. A little levity can go a long way I find.
especially the self-deprecating kind.
Scott Leibrand
@scottleibrand
Jun 06 2017 21:05
;-)
Tim Street
@tim2000s
Jun 06 2017 21:10
On the topic of range, I think it's may well be dependent on the rf status in a location. Two of my explorer boards reach two floors up in my house so that when I'm in bed, they talk to the pump from the ground floor. We have only the wifi 2.4 Ghz and 5Ghz in the house plus bt from Dexcom G5. There's not a lot else going on.
On the other hand I've been in a busy room with loads of Bluetooth and watched a RileyLink fail to connect to the phone effectively.
I suspect if you're in an area close to a cellular mast on an 800mHz signal you're going to see more of an impact than further away and if your uploader phone uses that frequency I guess it could affect pump response.
garykidd
@garykidd
Jun 06 2017 21:14
assuming this is the relevant app @scottleibrand https://pushover.net/
Scott Leibrand
@scottleibrand
Jun 06 2017 21:23
correct
Marco
@CaptainBalou
Jun 06 2017 21:47
@garykidd But the transmitter of a 640G doesn't work with my Veo. :-(
rsilvers129
@rsilvers129
Jun 06 2017 22:09
This board gets so hot charging it, I am not sure if I should buy the external charger and only charge when not on the Explorer. Or if I should just set up a small fan on it. But it sort of keeps me from using a USB power back when I am carryng it.
I think my long term answer is to make an enclosure for 18650s and just swap those as needed and charge them externally.
rsilvers129
@rsilvers129
Jun 06 2017 22:31
Is there a single command line I can type to test radio signal strength to a pump of a specific serial number?
Scott Leibrand
@scottleibrand
Jun 06 2017 22:38
probably easiest to use mmtune for that
rsilvers129
@rsilvers129
Jun 06 2017 22:52
Says /dev/spidev5.1 already in use. Do I need to stop cron and killall openaps?
Scott Leibrand
@scottleibrand
Jun 06 2017 22:53
yes
rsilvers129
@rsilvers129
Jun 06 2017 23:23
I ate 132 grams of carb pizza and entered it with my pump, so OpenAps saw what I did, and all was fine. A few hours later I got around to logging the food in MyNetDiary. I just noticed that NightScout had this massive amount of carb twice, which is not good. Is it because I had Loop running (in open loop mode) on my iPhone? My theory is that Loop saw the food in Apple Heath and decided to tell NightScout about it. But that was already in NightScout from OpenAps reading my pump. So I was getting insulin for 264 grams of carb.
Dana Lewis
@danamlewis
Jun 06 2017 23:24
I know Loop had some issues a while back about double enters to NS from using 3rd party apps.
for that, plus other reasons, you may want to disable Loop’s ability to upload to NS
you may be able to go in and delete the duplicate NS entry
rsilvers129
@rsilvers129
Jun 06 2017 23:28
I just had to eat two ramen noodle packs and a sugar cookie to load up with another similar amount of carb.
if you notice something like that, you may want to temporarily turn off everything if you can’t delete any duplicates
rsilvers129
@rsilvers129
Jun 06 2017 23:29
I suspended the pump.
Scott Leibrand
@scottleibrand
Jun 06 2017 23:29
@rsilvers129 I hope that was a "had to" (as in you wanted an excuse) rather than really feeling like you had to
rsilvers129
@rsilvers129
Jun 06 2017 23:29
Yeah. I wanted to.
Scott Leibrand
@scottleibrand
Jun 06 2017 23:30
how long was it between the additional carb entry and when you noticed? how much extra insulin did it do in that time?
rsilvers129
@rsilvers129
Jun 06 2017 23:30
Probably only like 2 units too much. I could have suspended the pump and had one Gatorade.
But I was looking for dinner anyway.
And didn't want to waste insulin on Gatorade.
Scott Leibrand
@scottleibrand
Jun 06 2017 23:32
if you deleted the carbs in NS, OpenAPS should realize you need less insulin and zero temp for you
rsilvers129
@rsilvers129
Jun 06 2017 23:32
Yeah. Not sure how to delete in NS
Oh. Hit the x?
Scott Leibrand
@scottleibrand
Jun 06 2017 23:33
you need to enable edit mode
and then hit the edit button up top and click and drag the treatment off to the side to remove it
rsilvers129
@rsilvers129
Jun 06 2017 23:34
There is a red X and it worked
On my Mac
Scott Leibrand
@scottleibrand
Jun 06 2017 23:34
are you in the main NS view or in the treatments report?
because yeah, that will work too. I'd forgotten about that option. :)
rsilvers129
@rsilvers129
Jun 06 2017 23:38
It worked. Thanks. That was super useful to know.
Screen Shot 2017-06-06 at 7.37.53 PM.png
So that is 1/2 of a large Papa Ginos Sausage pizza - 132 grams of carb. It did a decent job of it. Never over 170 or so.
I did eating soon, then I took a full bolus for the carb and ate right away. Then 2 hour later, I did a bolus for the fat and protein using that XLS calculator. I noticed that the rig was also giving me about 4.7 units/hour of basal, which made me glad that I did the second dose as a bolus because I knew OpenAPS would not be able to dose enough basal to keep up if it was doing 4.7 an hour on top of of my second bolus.
Scott Leibrand
@scottleibrand
Jun 06 2017 23:41
unless, perhaps, you were running oref1. ;-)
rsilvers129
@rsilvers129
Jun 06 2017 23:43
This was oref0. Do you mean oref1 would have been able to dose enough with just me logging the carbs and not doing a second manual bolus?
The most I saw oref1 do was 0.4 unit SMBs
Scott Leibrand
@scottleibrand
Jun 06 2017 23:44
oref1 could've done maxBolus-size SMBs in addition to running a high temp if it needed to
rsilvers129
@rsilvers129
Jun 06 2017 23:45
If it did one of those every 3 minutes for 30 minutes, that is just 4 units in 30 minutes. I gave myself 9 units in maybe 2 minutes, and the rig was giving another 2.35 over 30 minutes. So that is 11.35 over 30 minutes vs 4 for oref1.
Scott Leibrand
@scottleibrand
Jun 06 2017 23:45
you can certainly bolus faster, but it will usually start earlier than you would
rsilvers129
@rsilvers129
Jun 06 2017 23:46
If oref1 could handle the tail part automatically that would be nice.
I just don’t know the timing of when I should enter the fat as carbs. I am concerned that if I do it when I start eating, it will consider them gone after 2-3 hours
Scott Leibrand
@scottleibrand
Jun 06 2017 23:47
usually oref1 does SMBs every 5m, so if it did 0.4U every 5m, it'd do up to 2.4U per 30m as SMBs, plus the high temp (another 2.4U in 30m in this case). but more likely it'd do it more gradually over an hour or more
rsilvers129
@rsilvers129
Jun 06 2017 23:47
I tried this same thing with the same food yesterday and oref0. I simply logged the fat as carbs at 2 hours.
Scott Leibrand
@scottleibrand
Jun 06 2017 23:47
(starting earlier, and ending later)
rsilvers129
@rsilvers129
Jun 06 2017 23:48
That should be good. The XLS formula actually wanted me to do a 5-hour square wave with the 9 units
Scott Leibrand
@scottleibrand
Jun 06 2017 23:48
you're right that it will "forget" about entered carbs (or carb equivalents) eventually. I think the oref1 code has it remember them for DIA hours right now, and oref0 might be 1.5 * DIA hours
rsilvers129
@rsilvers129
Jun 06 2017 23:49
From past pizza tests, it takes about six hours for the large high fat meals to be processed.
Scott Leibrand
@scottleibrand
Jun 06 2017 23:49
if you enable UAM, that can also help it know to dose more aggressively for any meal-related deviations after the carbs expire
but it sounds like entering the carbs at two different times might be a good approach for that kind of meal
rsilvers129
@rsilvers129
Jun 06 2017 23:50
So would I be better off enabling UAM, or custom coding the carb processing to last 5 or 6 hours?
My DIA is set for 3 now
I think I need 2*DIA for pizza
Scott Leibrand
@scottleibrand
Jun 06 2017 23:52
someone still needs to figure out the reappearing carbs bug and/or make some other adjustments to allow carbs to live for 2 * DIA
I know one person using oref1 set it back up to 1.5 * DIA and set his min_5m_carbimpact a bit higher to mitigate any reappearing carbs
but all of these solutions are workarounds right now
rsilvers129
@rsilvers129
Jun 06 2017 23:56
I had 14 pieces of sushi and weighed it on a gram scale to know the carb and protein better. And it was 165 grams of carb. It was hard to believe it was so much, so I took less insulin. I ended up neeing more insulin than what 165 grams of carb would be. It makes me think the large amount of protein really did make me need more insulin than rice alone. So many things to test - rice with and without fish.
Any heat-sink ideas for the charging chip on the Explorer?
People have melted their wires. Mine thermal shutsdown.