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

11th
Mar 2017
Scott Leibrand
@scottleibrand
Mar 11 2017 00:01
both checks are relevant, but they're two different things
the contents of cd ~/src/oref0 (particularly which branch you're on) determine which version of oref0-setup you run. that in turn installs the appropriate version of the oref0 npm package
David
@dakago
Mar 11 2017 00:05
ok ... want me to do a pr and add npm list -g oref0 to the "update your rig" page? its actually something i've been wondering for a while
Scott Leibrand
@scottleibrand
Mar 11 2017 00:10
PR's always welcome, but not sure what you want to say there. It's not necessary to check the current version before updating, if that's what you mean...
gregtcm
@gregtcm
Mar 11 2017 00:11
@scottleibrand So Scott, with my output of oref0@0.4.1 am I not on the dev branch then?
(we forgot to update the version number in the dev branch)
fixed (it's 0.4.2-dev now)
gregtcm
@gregtcm
Mar 11 2017 00:15
Thanks! I'll update and see if that makes a difference. :)
David
@dakago
Mar 11 2017 00:17
I thinks its good so you know if you have to update or not. I'm on 0.3.4-dev and had no idea... probably time to update. I think a lot of average people like me might think if it ain't broken don't fix it and hold off on updates until bigger ones
Scott Leibrand
@scottleibrand
Mar 11 2017 00:17
@gregtcm are you doing the npm run global-install stuff?
gregtcm
@gregtcm
Mar 11 2017 00:19
@scottleibrand I used the code from the doc:
mkdir -p ~/src; cd ~/src && git clone -b dev git://github.com/openaps/oref0.git || (cd oref0 && git checkout dev && git pull)
Is there a better way?
and then the oref0-runagain.sh script
I've been wondering, though, because when I run the runagain script it looks like it's checking the master branch.
gregtcm
@gregtcm
Mar 11 2017 00:48
I ran the above code again and it says Already on 'dev' and that my branch is up-to-date. But npm list -g oref0 is still showing
/usr/local/lib
└── oref0@0.4.1
PieterGit
@PieterGit
Mar 11 2017 00:53
use sudo npm install -g . or npm run global-installto update your oref0
ann oxley
@oxintrot_twitter
Mar 11 2017 02:09
trying to flash the edison and cannot progress past sudo screen /dev/tty.usbserial-115200 get error on blank screen that says cannot exec '/dev/tty.usbserial- no such file or directory. I opened a new terminal and ran jubilinux with ./flashall.sh and then returned to reboot the other terminal and get operation is not permitted....where do I go from here?
Dana Lewis
@danamlewis
Mar 11 2017 02:17
@libby4label hi! here’s good
it looks like you created a file and it was empty, right? for trying to edit networks for getting rigs on wifi?
Scott Leibrand
@scottleibrand
Mar 11 2017 02:17
@oxintrot_twitter sounds like you're pasting the command in wrong? what exactly are you entering?
ann oxley
@oxintrot_twitter
Mar 11 2017 02:18
copy and pasted it exactly from the openaps pages
sudo screen /dev/tty.usbserial-* 115200
Dana Lewis
@danamlewis
Mar 11 2017 02:18
@libby4label chances are you did a typo on the command, or were in the wrong directory (so it created a new, empty file vs opening the one you want to change)
@libby4label so if you weren’t copying/pasting that line, do it again with a copy/paste and see if that opens it for you.
Scott Leibrand
@scottleibrand
Mar 11 2017 02:19
@oxintrot_twitter your microUSB cable might not be working properly for data, then. try the stuff in the docs about troubelshooting cables by swapping them etc?
libby4label
@libby4label
Mar 11 2017 02:19
so how can i easily back up
ann oxley
@oxintrot_twitter
Mar 11 2017 02:19
I switched around the cables as well
Dana Lewis
@danamlewis
Mar 11 2017 02:20
@libby4label escape to get out of insert mode; then :q! to quit out
err, escape rather than control c to get out of insert mode
ann oxley
@oxintrot_twitter
Mar 11 2017 02:20
I have followed all of the trouble shooting information that I have found, I can see if I can find another cable
Scott Leibrand
@scottleibrand
Mar 11 2017 02:20
do both cables work for making the Edison show up in finder?
ann oxley
@oxintrot_twitter
Mar 11 2017 02:21
let me check that
libby4label
@libby4label
Mar 11 2017 02:22
changed cables and now there is "in-bulk and then dwc3 device dwc3 device.1
Scott Leibrand
@scottleibrand
Mar 11 2017 02:23
@libby4label my cables question was for @oxintrot_twitter not you
libby4label
@libby4label
Mar 11 2017 02:23
sorry
Scott Leibrand
@scottleibrand
Mar 11 2017 02:24
whenever we post here without tagging someone directly, we're usually referring to the last message immediately preceding what we said
Dana Lewis
@danamlewis
Mar 11 2017 02:25
but, never hurts to ask if you’re unsure about where something is directed.
no worries!
Scott Leibrand
@scottleibrand
Mar 11 2017 02:25
hope that didn't throw off your editing
Dana Lewis
@danamlewis
Mar 11 2017 02:25
@libby4label ok - did you do escape and :wq! to exit, or just change cables? what are you seeing currently?
libby4label
@libby4label
Mar 11 2017 02:25
I can't escape
it just dings at me
i had changed cables and then the screen filled with 677575] dwc3-device dwc3-device.1: request f5c45540 was not queued to ep6in-bulk
[ 2957.677667] dwc3-device dwc3-device.1: request f5f75780 was not queued to ep6in-bulk
[ 2957.836647] dwc3-device dwc3-device.1: request f
ann oxley
@oxintrot_twitter
Mar 11 2017 02:26
@scottleibrand so individually when I plug them in the Edison should be visible is what you are telling me correct?
Dana Lewis
@danamlewis
Mar 11 2017 02:26
@libby4label what happens if you control-c?
Scott Leibrand
@scottleibrand
Mar 11 2017 02:27
@oxintrot_twitter if you haven't flashed it yet, the Edison will show up in Finder when you connect to its OTG port
libby4label
@libby4label
Mar 11 2017 02:27
@danamlewis it just dings
Dana Lewis
@danamlewis
Mar 11 2017 02:27
@libby4label ok. let’s close this window, and open a new terminal and log back into the rig.
libby4label
@libby4label
Mar 11 2017 02:28
@danamlewis ok...
ann oxley
@oxintrot_twitter
Mar 11 2017 02:29
@scottleibrand ok so only 1 of 3 cables when plugged in individually to the OTG port show the Edison
so I am assuming that I need to get another cable before trying to progress any further
Scott Leibrand
@scottleibrand
Mar 11 2017 02:31
ok, then the other two won't work, right
or you can try to do one thing at a time
use your one good cable for flashing, and then use it for console. that is hard to get right though, so better to get another cable
ann oxley
@oxintrot_twitter
Mar 11 2017 02:32
I will go get another cable and try that route
libby4label
@libby4label
Mar 11 2017 02:32
@danamlewis it does nothing now...the screen is black with a green box and it dings
dan lightner
@danimaniac
Mar 11 2017 02:33
@ecc1 Pump still has a meter ID stored from the previous owner. I secured both from him. Meter option is turned on, sensor option is off, remote option is off. I wanted to turn off the sensor option so I only had to press Esc once from the home screen to view the status screen with current temp basal. With the sensor option on...it took 3 clicks! How completely lazy can I be? Thanks.
ann oxley
@oxintrot_twitter
Mar 11 2017 02:34
@scottleibrand Thanks for the help
Dana Lewis
@danamlewis
Mar 11 2017 02:36
@libby4label ok. time to reboot the rig. (press and hold down the little circular button until the light goes off, wait a second then press it til light turns back on )
libby4label
@libby4label
Mar 11 2017 02:38
k
@danamlewis ...done
Dana Lewis
@danamlewis
Mar 11 2017 02:40
@libby4label great. let’s log back into the rig now
once you’re in, then copy and paste the vi line in again and that should bring up you wifi screen to edit
libby4label
@libby4label
Mar 11 2017 02:51
@danamlewis small confusion over passwords...hopefully we are back in...
Dana Lewis
@danamlewis
Mar 11 2017 02:52
:+1:
libby4label
@libby4label
Mar 11 2017 02:52
@danamlewis edison@devonsolberg:~$
thats what we are seeing
Dana Lewis
@danamlewis
Mar 11 2017 02:53
Great. Did you copy and paste in the vi command for editing the wifi file?
libby4label
@libby4label
Mar 11 2017 02:55
@danamlewis yes my son got it!
Dana Lewis
@danamlewis
Mar 11 2017 02:55
Great! :clap:
libby4label
@libby4label
Mar 11 2017 02:56
@danamlewis
So now we type i right
Dana Lewis
@danamlewis
Mar 11 2017 02:56
Yep. Then you can edit.
libby4label
@libby4label
Mar 11 2017 03:07
@danamlewis How do you ucomment the auto wlan()
Dana Lewis
@danamlewis
Mar 11 2017 03:11
Just remove the #
You probably want to delete those posts too btw since it has your info in them
libby4label
@libby4label
Mar 11 2017 03:13
@danamlewis yep done and done
Dana Lewis
@danamlewis
Mar 11 2017 03:13
:+1:
libby4label
@libby4label
Mar 11 2017 03:16
@danamlewis So I went to get rid of the two lines and put in the wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf but when i did that it said that i should quit or continue with caution... what did i do wrong
Scott Leibrand
@scottleibrand
Mar 11 2017 03:19
@libby4label you may need to copy and paste in the text of the warning you got for us to know what you're referring to. My best guess is it's telling you that you were already editing the file when you rebooted the Edison, and that you can either recover where you left off, or delete the auto recovered version and start again.
libby4label
@libby4label
Mar 11 2017 03:20

6
E325: ATTENTION
Found a swap file by the name "/etc/network/.interfaces.swp"
owned by: root dated: Fri Dec 31 17:40:07 1999
file name: /etc/network/interfaces
modified: YES
user name: root host name: edison
process ID: 453
While opening file "/etc/network/interfaces"
dated: Thu Dec 22 18:20:44 2016
NEWER than swap file!

(1) Another program may be editing the same file. If this is the case,
be careful not to end up with two different instances of the same
file when making changes. Quit, or continue with caution.
(2) An edit session for this file crashed.
If this is the case, use ":recover" or "vim -r /etc/network/interfaces"
to recover the changes (see ":help recovery").
If you did this already, delete the swap file "/etc/network/.interfaces.swp "

to avoid this message.
dan lightner
@danimaniac
Mar 11 2017 03:22
Seems like several of us new loopers are having difficulty with bluetooth tethering. I am repeatedly successful, step by step, through pairing and trusting my phone and Edison, but I repeatedly fail when attempting to establish a connection. I have no trouble tethering my laptop to phone via wifi...so I believe my phone / carrier is capable.
root@lightnerdaps:~# sudo bt-pan client 1C:AF:05:75:0E:AA
Traceback (most recent call last):
File "/usr/local/bin/bt-pan", line 228, in <module>
if name == 'main': sys.exit(main())
File "/usr/local/bin/bt-pan", line 196, in main
try: iface = net.Connect(opts.uuid)
File "/usr/lib/python2.7/dist-packages/dbus/proxies.py", line 145, in call
**keywords)
File "/usr/lib/python2.7/dist-packages/dbus/connection.py", line 651, in call_blocking
message, timeout)
dbus.exceptions.DBusException: org.bluez.Error.NotSupported: Operation is not supported
libby4label
@libby4label
Mar 11 2017 03:26
@scottleibrand how do i recover or delete it
Eric
@ecc1
Mar 11 2017 03:28

@danimaniac tethering via wifi is completely different though. Try running bluetoothctl and doing info <your phone's BT MAC address>

It should say it's paired and trusted, and there should be a UUID: NAP line in the output that indicates your phone offers the BT tethering service.

Perhaps this ^^^ should be added to the docs somewhere as a way to check whether the phone is capable of BT tethering
libby4label
@libby4label
Mar 11 2017 03:33
@scottleibrand How do i restart it or quit it
Dana Lewis
@danamlewis
Mar 11 2017 03:33
@libby4label rm <filename>
The filename is what it told you in quotes
(Don't do the < >)
libby4label
@libby4label
Mar 11 2017 03:45
@danamlewis can we backup or just recover... we have duplication of lines , can make things delete
Dana Lewis
@danamlewis
Mar 11 2017 03:48
I'm not sure I understand what you are seeing, or what you are proposing to do.
Scott Leibrand
@scottleibrand
Mar 11 2017 03:49
if you want to erase the contents of the interfaces file inside vi (while in insert mode), you can then paste in the full example from the docs
libby4label
@libby4label
Mar 11 2017 03:59
@danamlewis I believe this is the file that needs to be removed as we have awarning that another program may be editing the same file. </etc/network/.interfaces.swp>
@danamlewis I just don't know how to remove that file.....any time I try to type it splits on 2 lines or dings at me
Dana Lewis
@danamlewis
Mar 11 2017 04:00
Did you try rm <filename>?
libby4label
@libby4label
Mar 11 2017 04:01
the green box is in the middle of the page and when I type the words split between to lines in some random place on the screen
ig888ack
@ig888ack
Mar 11 2017 04:03
Hiya folks...I spun up a VM and I'm trying autotune for the first time. setup seemed to go well but I'm getting an error while it's running and I was hoping to get some help with it:

'''
oref0-autotune-prep ns-treatments.json profile.json ns-entries.2017-03-09.json > autotune.1.2017-03-09.json
Could not parse input data: [SyntaxError: Unexpected token <]
oref0-autotune-core autotune.1.2017-03-09.json profile.json profile.pump.json > newprofile.1.2017-03-09.json
/usr/local/lib/node_modules/oref0/lib/autotune/index.js:77
for (var i=0; i < basalGlucose.length; ++i) {
^

TypeError: Cannot read property 'length' of undefined
at tuneAllTheThings (/usr/local/lib/node_modules/oref0/lib/autotune/index.js:77:39)
at Object.<anonymous> (/usr/local/lib/node_modules/oref0/bin/oref0-autotune-core.js:59:27)
at Module._compile (module.js:410:26)
at Object.Module._extensions..js (module.js:417:10)
at Module.load (module.js:344:32)
at Function.Module._load (module.js:301:12)
at Function.Module.runMain (module.js:442:10)
at startup (node.js:136:18)
at node.js:966:3
Could not run oref0-autotune-core autotune.1.2017-03-09.json profile.json profile.pump.json
'''

libby4label
@libby4label
Mar 11 2017 04:03
I can never arrow to the bottom of the screen....it just keeps adding numbers on the left and right sides of the screen
Dana Lewis
@danamlewis
Mar 11 2017 04:03
@ig888ack let's hop into the autotune channel for that troubleshooting
Scott Leibrand
@scottleibrand
Mar 11 2017 04:04
@ig888ack [SyntaxError: Unexpected token <] is your clue
scottleibrand @scottleibrand moves there
Scott Leibrand
@scottleibrand
Mar 11 2017 04:05
@libby4label did you ever close out your terminal window and open a new one? did you ever resize the terminal window after opening it?
libby4label
@libby4label
Mar 11 2017 04:05
@scottleibrand yes and yes
Scott Leibrand
@scottleibrand
Mar 11 2017 04:05
ok, let's do that again, but don't resize it this time
libby4label
@libby4label
Mar 11 2017 04:06
so close out and jump back in?
Scott Leibrand
@scottleibrand
Mar 11 2017 04:06
resizing messes up the line wrapping sometimes
libby4label
@libby4label
Mar 11 2017 04:06
gotcha
Scott Leibrand
@scottleibrand
Mar 11 2017 04:06
once you get wifi set up and can ssh in, you can resize all you want. this is just a limitation of serial console
(which is still heavily based on technology from before I was born, when computer monitors were green on black, 80 columns by 24 rows of text)
so resizing to anything other than 80 columns often messes it up. :)
libby4label
@libby4label
Mar 11 2017 04:28
@danamlewis @scottleibrand When i put the :wq and hit enter this comes up E45: 'readonly' option is set (add ! to override)
Dana Lewis
@danamlewis
Mar 11 2017 04:31
Add ! to your :wq then
libby4label
@libby4label
Mar 11 2017 04:32
@danamlewis ok did that now it sayys cant open file for writing press enter or type command to continue
Scott Leibrand
@scottleibrand
Mar 11 2017 04:33
you're not logged in as root. is this an Explorer board?
libby4label
@libby4label
Mar 11 2017 04:34
yes
Scott Leibrand
@scottleibrand
Mar 11 2017 04:35
you skipped that part of the directions then. I'd go back and read it through more carefully so you don't miss anything else
libby4label
@libby4label
Mar 11 2017 04:40
@scottleibrand sorry to have taken your time
Scott Leibrand
@scottleibrand
Mar 11 2017 04:46
no worries. you identified some things that aren't very well documented
once you get up and running, you might also want to do a PR to add in a comment about not resizing the console window, so no one else trips over that same issue...
at this point you should be logged back in as root and able to do your edits. let us know if you get stuck again
libby4label
@libby4label
Mar 11 2017 04:48
@scottleibrand it has been impossible for us to find time to devout 3-6 hours to this project so it has been done over a series of evenings since Christmas. ....but now my T1 has had enough again tonight....we've rebooted and re-done over and over and he's thrown his hands in the air
i don't know why we weren't or aren't logged in as root
Scott Leibrand
@scottleibrand
Mar 11 2017 04:50
when you first connect to the console, it asks what user to log in as: you probably typed "edison" instead of "root" there
you can also log out by typing "exit" and then log back in as root
This message was deleted
scottleibrand @scottleibrand has flashbacks to typing my password into the chat window at work... :-)
libby4label
@libby4label
Mar 11 2017 04:51
except that now it says devonsolberg login: and won't accept exit as a choice....of course
Scott Leibrand
@scottleibrand
Mar 11 2017 04:52
if it says login: that's where you'll want to type root
that means you already logged out
libby4label
@libby4label
Mar 11 2017 05:06
@scottleibrand grand....apparently the password for root is not the same as it is for edison and now we are skunked because neither my son nor I can recall that password! Good grief! This may just never happen!
Scott Leibrand
@scottleibrand
Mar 11 2017 05:06
ah, that's fixable
if you log in as edison, you should be able to do sudo passwd root to change the root password
libby4label
@libby4label
Mar 11 2017 05:08
how do i make it stop asking for the login so i can do the sudo passwd root
Scott Leibrand
@scottleibrand
Mar 11 2017 05:09
if it's asking for a login, you need to log in. that means typing the username and password
it sounds like the edison username and password are the ones you know, so you'd need to log in with those
glenrob1940
@glenrob1940
Mar 11 2017 05:23

Need some ideas. I have 2 raspberry Pi 3’s with TI sticks. They essentially were set up exactly the same. I had them in separate areas in the house and openaps worked well on both. I was running closed loop with max_iob of 4 and autosense. I decided to start using autotune and ran oref0-setup on one of them to try. Within about 3 minutes the Pi locked up (couldn’t ping it, ssh window was unresponsive) and the normally blinking (green) LEDs would flicker every now and then. Kind of looked like it might be still alive, but perhaps a kernel driver was locked up. The lockups happened every time I reapplied power about the time an initial mmtune was running. I swapped TI sticks with no change in behavior. I reflashed (dd'd the sd) the Pi and set everything up to run autotune. The hang/lockups occurred as before. I killed the crontab entries, rebooted and everything (except of course openaps) worked fine – that is, no hang. The next thing I tried was a dd copy of the working system micro sd card to another card, put it in the previously failing system and rebooted. Openaps worked fine (though this was just the closed loop with autosense). Running uname -a shows that the working sd cards have an earlier debian release.
Linux rpi2 4.4.38-v7+ #938 SMP Thu Dec 15 15:22:21 GMT 2016 armv7l GNU/Linux
Linux raspberrypi 4.4.50-v7+ #970 SMP Mon Feb 20 19:18:29 GMT 2017 armv7l GNU/Linux

Note that the hangs occurred even when I re-ran oref0-setup with no advanced options – during or shortly after the first mmtune in the loop.

Anyone have ideas for further troubleshooting. I can, of course, continue running the Pi now with the older kernel. I am concerned, however, that there may be a driver issue.

Scott Leibrand
@scottleibrand
Mar 11 2017 05:31
@libby4label closing and reopening the terminal doesn't log you out. You'll need to exit
libby4label
@libby4label
Mar 11 2017 05:32
done!
thanks
Scott Leibrand
@scottleibrand
Mar 11 2017 05:33
@glenrob1940 I would run oref0-setup without doing cron so you can debug further.
It sounds like maybe you might be using the wrong port for mmeowlink?
glenrob1940
@glenrob1940
Mar 11 2017 05:35
I am using /dev/mmeowlink and when oref0-setup runs, the mmtune then works correctly. It is just when the loop is running from cron that it hangs. So, to debug further should I go to the manual steps, or should I enter a line of cron at a time, or, ...
I will go back and take a close look at the udev stuff for mmeowlink though in case I messed something up there.
Scott Leibrand
@scottleibrand
Mar 11 2017 06:14
I would try running the openaps pump-loop alias manually and watch what that does first
glenrob1940
@glenrob1940
Mar 11 2017 06:15
okay I'll give it a try
glenrob1940
@glenrob1940
Mar 11 2017 06:22
mmtune succeeded, but some error messages of fatal loose object stored in git ...
Scott Leibrand
@scottleibrand
Mar 11 2017 06:22
maybe just blow away .git and run oref0-reset-git to re-initialize it?
glenrob1940
@glenrob1940
Mar 11 2017 06:22
okay
glenrob1940
@glenrob1940
Mar 11 2017 06:29
Results
pi@raspberrypi:~/myopenaps$ openaps pump-loop
Starting pump-loop at Fri Mar 10 23:24:43 MST 2017:
Radio ok, Listening: ....No pump comms detected from other rigs (this is a good thing!)
Profile less than 60m old
temp_basal.json less than 5m old
pumphistory more than 5m old: Refreshed pumphistory
determine-basal://text/shell/enact/suggested.json
Could not parse input data: { [Error: Cannot find module '/home/pi/myopenaps/monitor/glucose.json'] code: 'MODULE_NOT_FOUND' }
reporting enact/suggested.json
cat: enact/enacted.json: No such file or directory
Settings refreshed
Pumphistory-24 less than 20m old
Completed pump-loop at Fri Mar 10 23:25:49 MST 2017
Shouldn't oref0-setup have set those files up?
Scott Leibrand
@scottleibrand
Mar 11 2017 06:33
get that rig some glucose (data), stat! :)
if you've never run the stuff in cron, you'll just need to run the cgm-loop first
Dana Lewis
@danamlewis
Mar 11 2017 06:34
::cheers at seeing her new code addition in the logs ;) ::
Scott Leibrand
@scottleibrand
Mar 11 2017 06:35
that's openaps get-bg if you have a plugged-in CGM
for getting BG from NS it's openaps ns-loop
glenrob1940
@glenrob1940
Mar 11 2017 06:39
okay
Eventually, to get it into cron, do i need to re-run oref0-setup, or can I get at least autosense going by copying the crontab from my other Pi
My guess is that if I get autosense and loop running under cron, I can run autotune manually
glenrob1940
@glenrob1940
Mar 11 2017 06:51
The okay above, was that the pump loop has run successfully 2 times. thanks.
Scott Leibrand
@scottleibrand
Mar 11 2017 06:58
At this point I would re-run oref0-setup to add everything into cron. The one on the other Pi may not match these settings.
glenrob1940
@glenrob1940
Mar 11 2017 07:14
Did that and it hung

pi@raspberrypi:~/myopenaps$ tail -f /var/log/openaps/p
tail: cannot open ‘/var/log/openaps/p’ for reading: No such file or directory
tail: no files remaining
pi@raspberrypi:~/myopenaps$ tail -f /var/log/openaps/pump-loop.log
Radio ok, Listening: .No pump comms detected from other rigs (this is a good thing!)
Profile less than 60m old
temp_basal.json less than 5m old
Pumphistory less than 5m old
Settings less than 10m old
Pumphistory-24 less than 20m old
Completed pump-loop at Sat Mar 11 00:05:44 MST 2017

Starting pump-loop at Sat Mar 11 00:06:10 MST 2017:
Radio ok, Listening: .
Had about 5 pings after the secomd dot above

it's after midnight in colorado, i'll do some more troubleshooting tomorrow
Scott Leibrand
@scottleibrand
Mar 11 2017 08:17
Odd. But running the pump-loop on its own doesn't hang?
What happens if you unplug the TI stick and reboot?
PieterGit
@PieterGit
Mar 11 2017 08:25
@glenrob1940 : i had similar experiences that a pi3 rig can sometimes hang with a TI USB stick. i noticed there are two possible problems:
  • the pump communication / mmtune that excesses most cpu resources
  • a broken git repository can lead to high cpu usage as well
I would not only disable the pump loop, but temporary disable all cron jobs. easiest way is /etc/init.d/cron stop That will disable cron until you rebout
are you sure it's a high cpu uage problem (you can check that with uptime or top) or hanging problem, because i also had network problems in the past (where the cpu level was low, but the network connections to the rig were not working properly). Please tag me if you need any help.
PieterGit
@PieterGit
Mar 11 2017 08:34
Does the pi get hotter than normal?
PieterGit
@PieterGit
Mar 11 2017 09:13
@scottleibrand (and all others). I created an issues for the logging stuff we discussed yesterday. openaps/oref0#420 . i don't have time for openaps this weekend, but will work on my devww branch and improving openaps/logging next week. input about the way we want to improve logging is welcome.
Mike
@libxmike
Mar 11 2017 10:09
@PieterGit you committed your devww to dev correct?
@PieterGit I noticed that git is taking up to 300% CPU in top, so this may be one of the problems (on Rpi3 latest dev and Ti USB)
so I disabled both in crontab
Mike
@libxmike
Mar 11 2017 10:15
now I have to understand what the git part in crontab does and why we need it
Mike
@libxmike
Mar 11 2017 10:28
disabled git part in crontab but still runs hogging the CPU 388% for a while
pi        2140     1  0 11:26 ?        00:00:00 git gc --auto
pi        2141  2140  0 11:26 ?        00:00:00 git repack -d -l
pi        2142  2141 99 11:26 ?        00:00:32 git pack-objects --keep-true-parents --honor-pack-keep --non-empty --all --reflog --unpacked --incremental --local --delta-base-offset .git/objects/pack/.tmp-2141-pack
PieterGit
@PieterGit
Mar 11 2017 13:04
@libxmike: no, the devww branch (https://github.com/pietergit/oref0 , branch devww, pull request ) is not comitted to dev yet. @scottleibrand first wants to do a minor release of all the stuff that's currently on dev (see openaps/oref0#414 ). He's planning to do that sunday. After that we'll focus on getting all the world-wide pump (devww) stuff ok working well on the dev branch.
PieterGit
@PieterGit
Mar 11 2017 13:10
@libxmike (and maybe also @glenrob1940 ) did your pi come back to normal cpu usage? what i think is happening is a git corruption that is not fixed by oref0-reset-git or oref0-truncate-git-history . @libxmike : i really think this is a bug. can you create an issue for this?
a workaround if the CPU stays at > 200% is:
mv .git ../git-corrupted
git init

and a reboot to get rid of all the hanging openaps/git processes

I think the PI's are way more sensitive for git corruption (sd card vs eMMC memory), maybe we should make it possible to rotate/move the .git directory (outside of the openaps dir!) and recreate a new git repo ( git init) each boot and/or each day.

PieterGit
@PieterGit
Mar 11 2017 13:29
Can people with PI3/TI USB rig confirm that the latest dev is working without problems? @scottleibrand wants to release the dev to master, but I want to make sure that there are no regressions with PI3/TI USB on current dev branch. There are some known problems with WW-pumps, but I want to make sure there are no regressions for the PI3/TI USB users. @libxmike has some issues with dev, but has a WW-pump. @glenrob1940 : do you use a WW-pump or a US-pump? Do you still encounter issues with the dev branch?
Discussion is at openaps/oref0#414
Mike
@libxmike
Mar 11 2017 13:51
@PieterGit git taking up to 400% recovers after few minutes but comms after few minutes again
@PieterGit I'm bit confused, which version should I install to test yours or the official dev?
I have now openaps 0.2.0-dev
Terri Lyman
@tatigger
Mar 11 2017 14:23
Good morning - around midnight my OpenAPS and Pump pills stopped updating and are not providing any status on my OpenAPS system. The Basal pill is still updating and is correctly reflecting the temp rate OpenAPS has set on my pump. Nightscout is getting the pump-entered updates from OpenAPS, and OpenAPS is getting BG data from Nightscout. Anyone have any ideas on how to fix this? I also have started an inquiry on the CGM in the Cloud Facebook Group.
tekrulik
@tekrulik
Mar 11 2017 14:25
Tonight will be our 1st time change with rigs. What is the best strategy to keeping things working overnight tonight? Should we just power down wifi and run offline until we can make sure all device times are synced in the morning?
@tatigger Delete devicestatus records from mongo.
Terri Lyman
@tatigger
Mar 11 2017 14:26
Thanks!
Terri Lyman
@tatigger
Mar 11 2017 14:38
@tekrulik thanks that fixed the issue.
PieterGit
@PieterGit
Mar 11 2017 15:41
@libxmike : thanks for letting me no that in your case the comms recover after a few minutes from git recovery. i assume your git recover jobs are commented out in your cronjob now.
Mike
@libxmike
Mar 11 2017 15:42
@PieterGit yes, they are commented out
PieterGit
@PieterGit
Mar 11 2017 15:43
@libxmike: about the versions. Current dev is ready to be released tommorow. I have quite some fixes for TI USB and WW-pump, but unfortunately they seem to have some drawbacks as well. So I talked to @scottleibrand and my changes need some more love and attention after the next oref0 release. We must make sure that the fixes for TI USB and WW-pumps don't break stuff for other people without TI USB or WW-pump
Mike
@libxmike
Mar 11 2017 15:44
@PieterGit I'll reinstall the latest oref for ww, which one should I take?
as I'm using TI USB and WW-pump
PieterGit
@PieterGit
Mar 11 2017 15:45
if you wait 5-10 minutes I can give you a special branch, with one easy fix that all WW-pump users encounter now
Mike
@libxmike
Mar 11 2017 15:45
sure
good that I have few working rigs so I play around
PieterGit
@PieterGit
Mar 11 2017 15:47
if you can test that, it might be included into the next release, but that's up to @scottleibrand
the devww branch needs more love and attention, and i don't want to let others test stuff that doesn't work for me yet
PieterGit
@PieterGit
Mar 11 2017 15:58
@libxmike @scottleibrand : here's the pull request openaps/oref0#421
the one thing you need to check after install is whether openaps.ini has subg-ww-radio-parameters in the mmtune alias https://github.com/openaps/oref0/pull/421/files#diff-704a2fd9a43fd79bf52037b1668c146dR549
Mike
@libxmike
Mar 11 2017 16:01
do I have to remove the old oref0/ src/ ?
PieterGit
@PieterGit
Mar 11 2017 16:02
@libxmike : let me give you a hand.
cd ~/src ; mv oref0 oref0.default ; git clone https://github.com/PieterGit/oref0.git ; cd oref0 ; git checkout devww_cherrypick
the proceed with normal oref0-setup.sh stuff
Mike
@libxmike
Mar 11 2017 16:03
excellent!, thx
PieterGit
@PieterGit
Mar 11 2017 16:04
please report your findings here openaps/oref0#421
this fixes the most blocking issues for ww pump users (with a fresh install), but there may be more ti usb and/or ww pumps issues that need more work
Mike
@libxmike
Mar 11 2017 16:05
will see what the log says then
PieterGit
@PieterGit
Mar 11 2017 16:06
i'm away for the rest of the afternoon.
Eric
@ecc1
Mar 11 2017 16:53
@tekrulik There's a script oref0-set-device-clocks.sh that sets the pump and cgm clocks based on the rig's time, using NTP. So if that's not already part of your loop, you could schedule it for 2am tonight via cron.
Actually, better do it at 3am, since 2am won't actually exist tonight :-)
hilarykoch
@hilarykoch
Mar 11 2017 17:16
When I setup OpenAPS, I turned off autosens and autotune. How do I turn autosens back on?
Terri Lyman
@tatigger
Mar 11 2017 17:24
I have a silly question. When you tether an edison to an android phone via bluetooth, is that then the only bluetooth connection that is allowed? Or will my Pebble and Dexcom still communicate with the android phone?
@hilarykoch when you go back through the setup script, you have the option to turn those on as part of setup. Then you have to go back in and fix your preferences if you made any changes to that file
hilarykoch
@hilarykoch
Mar 11 2017 17:28
@tatigger Your other devices will still work
@tatigger Yes. I just wasn’t sure if there was something other than the setup script. Thank you!
Terri Lyman
@tatigger
Mar 11 2017 17:28
@hilarykoch Thanks. I am still struggling to make this work, but not ready to ask questions yet about it. I have more resources to exhaust first. sigh.
You're very welcome :)
PieterGit
@PieterGit
Mar 11 2017 17:37
@libxmike : saw your comment in openaps/oref0#421 and give some explanation of what's going on. @scottleibrand : could you also help in this discussion?
hilarykoch
@hilarykoch
Mar 11 2017 17:46
Second try. I want to enable autosens but wanted to avoid running the script all over again and having to input all pump info, BT info, etc. . If I want to edit my settings to rerun scrip, I do this: "nano oref0-runagain.sh" But I’m not clear how to add autosens. Do I just add it behind “enable”? Do I add single quotations?
PieterGit
@PieterGit
Mar 11 2017 18:54
@hilarykoch : you can look at the usage string:
Usage: oref0-setup.sh <--dir=directory> <--serial=pump_serial_#> [--tty=/dev/ttySOMETHING] [--max_iob=0] [--ns-host=https://mynightscout.azurewebsites.net] [--api-secret=myplaintextsecret] [--cgm=(G4|shareble|G4-raw|G5|MDT|xdrip)] [--bleserial=SM123456] [--blemac=FE:DC:BA:98:76:54] [--btmac=AB:CD:EF:01:23:45] [--enable='autosens meal dexusb'] [--radio_locale=(WW|US)]
so you can use --enable='autosens meal' if you want autosens and advanced meal assist
hilarykoch
@hilarykoch
Mar 11 2017 18:56
That’s exactly what I needed. Thanks! @PieterGit
Another question. I’m trying to setup new rigs. I can’t seem to get the Edison in screen mode using "sudo screen /dev/tty.usbserial-* 115200” It says “cannot find a PTY” and then terminates the screen. Any ideas?
PieterGit
@PieterGit
Mar 11 2017 19:01
Not much edison experience yet. What does ls -al /dev/tty.usb*do?
dan lightner
@danimaniac
Mar 11 2017 19:02
My vented OpenAPS Edison / Explorer board case with power button is ready for print. You can view at the following link. Constructive criticism always welcome! https://github.com/danimaniac/OpenAPS-Explorer-Board-Edison-vented-case/blob/master/OpenAPS%20Explorer%20Edison%20Case%202017-03-11.stl
hilarykoch
@hilarykoch
Mar 11 2017 19:03
@PieterGit it says: crw-rw-rw- 1 root wheel 18, 2 Mar 11 13:42 /dev/tty.usbmodem1423
crw-rw-rw- 1 root wheel 18, 4 Mar 11 13:51 /dev/tty.usbserial-DO009M5T
PieterGit
@PieterGit
Mar 11 2017 19:04
@danimaniac : congrats. did you already printed it and does the explorer board fit (with two usb connectors)?
@hilarykoch : sorry, can't help you with that
hilarykoch
@hilarykoch
Mar 11 2017 19:04
@PieterGit no worries.
Dana Lewis
@danamlewis
Mar 11 2017 19:05
for @tekrulik and all who are wondering about time changes tonight: http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-4/Usability-considerations.html#what-do-you-do-with-your-loop-when-you-travel-across-timezones <--how to change times on your rigs and devices
hilarykoch
@hilarykoch
Mar 11 2017 19:07
@PieterGit unplugging USB ports and changing them, did the trick. :)
PieterGit
@PieterGit
Mar 11 2017 19:08
it's just like windows and rebooting stuff :smile:
dan lightner
@danimaniac
Mar 11 2017 19:27
@PieterGit No, I have not printed yet. I did break down and purchase a cheap caliper from Menards to measure my rig and the location of usb ports, etc. I have never designed or printed any 3d things, so it could certainly be wrong! I am looking for a local place to print and further revise (the power button in particular).
Terri Lyman
@tatigger
Mar 11 2017 19:42
I am having trouble connecting the Edison rig to my Android phone. I get this error
root@localhost:~# sudo bt-pan client 88:79:7e:c1:63:04
Traceback (most recent call last):
  File "/usr/local/bin/bt-pan", line 228, in <module>
    if __name__ == '__main__': sys.exit(main())
  File "/usr/local/bin/bt-pan", line 188, in main
    dev_remote = find_device(opts.remote_addr, devs.values()[0])
  File "/usr/local/bin/bt-pan", line 53, in find_device
    return find_device_in_objects(get_manager().GetManagedObjects(), device_address, adapter_pattern)
  File "/usr/local/bin/bt-pan", line 68, in find_device_in_objects
    raise BTError('Bluetooth device not found')
__main__.BTError: Bluetooth device not found
I have run through the process several times. The phone seems to feel that the rig is connected.
gregtcm
@gregtcm
Mar 11 2017 19:54
While looping, when I inject insulin via syringe, what is the best way to enter that so OpenAPS knows about it? xDrip+? Careportal?
Diabetic_Dude
@diabetic_dude_twitter
Mar 11 2017 19:57
I am no expert and YDMV but I would expect careportal
Dana Lewis
@danamlewis
Mar 11 2017 19:57
@gregtcm most important thing to do will be to set a high temp target. OpenAPS won't be able to track it because it'll overwrite it when it pulls pump history if you enter it as a bolus. So setting a high target will help if you have extra iob coming from elsewhere so it doesn't also try to correct what you've chosen to correct for manually.
Or disconnect from your site and do a manual bolus for the same amount if you really want to track it
@gregtcm is this an occasional thing or do you try this regularly?
Diabetic_Dude
@diabetic_dude_twitter
Mar 11 2017 20:00
So even if you manually log it in careportal pump history will over ride? Good to know
Terri Lyman
@tatigger
Mar 11 2017 20:01
ok, will check back later on the Android bt tethering thing. I need a break :)
Eric
@ecc1
Mar 11 2017 20:12

Here's what I've added to my rig's crontab to catch changes to and from DST:

# change to daylight time on second Sunday of March
0 3 8-14 mar * test $(date +\%u) -eq 7 && set-clocks.sh

# change to standard time on first Sunday of November
0 1 1-7  nov * test $(date +\%u) -eq 7 && set-clocks.sh

set-clocks.sh is my own script; oref0-set-device-clocks.sh should work on a standard openaps rig. Won't be tested until 3am tonight, I'm afraid.

hilarykoch
@hilarykoch
Mar 11 2017 20:14
Just about finished setting up a second rig. If both rigs are in the same room, is that problematic?
Dana Lewis
@danamlewis
Mar 11 2017 20:15
Nope
That's why you'll see it listening for other rigs, so it doesn't stomp on each other
You can have upward of 7 rigs running and it won't be a problem ;)
hilarykoch
@hilarykoch
Mar 11 2017 20:16
Cool!
Mike
@libxmike
Mar 11 2017 20:27
@scottleibrand with the new dev from today I get this error, never got it before, HW Rpi3/Ti USB/WW pump

mmtune: monitor/mmtune.json  raised  [Errno 2] could not open port /dev/ttyACM0: [Errno 2] No such file or directory: '/dev/ttyACM0'
pump://JSON/mmtune/monitor/mmtune.json
Traceback (most recent call last):
  File "/usr/local/bin/openaps-report", line 6, in <module>
    exec(compile(open(__file__).read(), __file__, 'exec'))
  File "/home/pi/src/openaps/bin/openaps-report", line 82, in <module>
    app( )
  File "/home/pi/src/openaps/openaps/cli/__init__.py", line 51, in __call__
    self.run(self.args)
  File "/home/pi/src/openaps/bin/openaps-report", line 75, in run
    output = app(args, self)
  File "/home/pi/src/openaps/openaps/cli/subcommand.py", line 52, in __call__
    return self.method.main(args, app)
  File "/home/pi/src/openaps/openaps/reports/invoke.py", line 40, in main
    output = task.method(args, app)
  File "/home/pi/src/openaps/openaps/uses/use.py", line 44, in __call__
    self.before_main(args, app)
  File "/home/pi/src/openaps/openaps/vendors/medtronic.py", line 62, in before_main
    self.setup_medtronic( )
  File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/mmeowlink.py", line 92, in setup_medtronic
    setup_medtronic_link(self)
  File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/mmeowlink.py", line 75, in setup_medtronic_link
    link = builder.build(radio_type, port)
  File "/usr/local/lib/python2.7/dist-packages/mmeowlink/link_builder.py", line 16, in build
    return SubgRfspyLink(port)
  File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/subg_rfspy_link.py", line 55, in __init__
    self.open()
  File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/serial_interface.py", line 30, in open
    self.serial = serial.Serial( self.device, self.speed )
  File "build/bdist.linux-armv7l/egg/serial/serialutil.py", line 236, in __init__
  File "build/bdist.linux-armv7l/egg/serial/serialposix.py", line 268, in open
serial.serialutil.SerialException: [Errno 2] could not open port /dev/ttyACM0: [Errno 2] No such file or directory: '/dev/ttyACM0'
Usage: grep [OPTION]... PATTERN [FILE]...
Try 'grep --help' for more information.
 Listening: ........
same every time after reboot ttyACM0 after a while is gone
gregtcm
@gregtcm
Mar 11 2017 20:32
@danamlewis With my old pump, a tslim, it was pretty regular since I was able to recover quite a bit of insulin with every cartridge change/reload. I would just save it in an old vial and use it up via manual injections. My limited experience with a Medtronic, though, seems like this might not be quite the same.
Eric
@ecc1
Mar 11 2017 20:40
Why not just fill the next reservoir from your extras, rather than injecting?
Dana Lewis
@danamlewis
Mar 11 2017 20:41
Yea ^ or save the bit in old reservoirs for when you need a reservoir change but aren't ready to swap a site and that gives you a few more hours
Eric
@ecc1
Mar 11 2017 20:43
Then there are those of us who just use the "no delivery" alarm as a "reminder" to refill the pump ...
Same approach I use for refilling my gas tank :-)
gregtcm
@gregtcm
Mar 11 2017 20:45
@ecc1 @danamlewis I was thinking the same thing. I never wanted to do that with my tslim as I didn't want to "contaminate" the cartridge with older insulin, but I think it'll work fine in this instance with an old reservoir! Thanks for confirming my thoughts. :)
Dana Lewis
@danamlewis
Mar 11 2017 20:45
Lol @ecc1 ;)
gregtcm
@gregtcm
Mar 11 2017 20:46
@ecc1 I've been doing the same with the "no delivery" alarm. :)
Dana Lewis
@danamlewis
Mar 11 2017 20:46
Be aware though by time actual no delivery, you're probably a unit or two short on insulin given the bubbles at the bottom of a reservoir
So yes going below 0 is fine (there's about ~10 u before no delivery) but often I'll start seeing a rise by time I get a no delivery which means I'm an hour or two behind on basal
gregtcm
@gregtcm
Mar 11 2017 20:48
Good to know. I noticed a few small air bubbles last time, but haven't been paying too close attention.
@ecc1 Did you share your set-clocks.sh script?
Dana Lewis
@danamlewis
Mar 11 2017 20:50
And if you DO decide to do a manual injection for whatever reason - always remember to set that high temp target and keep an eye on things. You can drop a note in care portal to help remind you of time/amount, but rig won't know about it.
Mike
@libxmike
Mar 11 2017 20:51
@danamlewis wont openaps take the insulin you input in NS?
Dana Lewis
@danamlewis
Mar 11 2017 20:52
It won't get read by the rig, it'll just show in NS
KS615
@KS615
Mar 11 2017 20:52
Hi all, we are building our initial rig (and just joining gitter). We are using Windows and are on the set that recommends we download from here https://cdn.sparkfun.com/assets/learn_tutorials/3/3/4/dfu-util-0.8-binaries.tar.xz To extract libusb-1.0.dll and dfu-util.exe.
Mike
@libxmike
Mar 11 2017 20:52
haa, I always tought it will add to IOB
gregtcm
@gregtcm
Mar 11 2017 20:52
@danamlewis Ok, thanks for the advice. BTW, I still owe you some PR from my initial setup. I seem to remember someone putting together a doc for setting up an Edison via Windows. Is that out there yet? Otherwise, I'll just PR in the main docs.
KS615
@KS615
Mar 11 2017 20:53
We can't seem to find anything to extract. Should we just copy the downloaded file to our Jubilinex folder? And, I'm sorry, Just realized I cut in on another thread, not sure how to start another one
Dana Lewis
@danamlewis
Mar 11 2017 20:54
@gregtcm haven't seen it come through yet, so just do a fresh pull and pr what's there - thanks!
gregtcm
@gregtcm
Mar 11 2017 20:54
@KS615 I'm relatively new to Gitter as well and I think that's just how it works! Keeps the mind agile to sort the conversations. :)
Mike
@libxmike
Mar 11 2017 20:55
@KS615 try to change .xz to .zip
Dana Lewis
@danamlewis
Mar 11 2017 20:55
(You're not cutting into a thread that's how it works :smile: no worries)
Mike
@libxmike
Mar 11 2017 20:55
this is zip compressed file
gregtcm
@gregtcm
Mar 11 2017 20:55
@KS615 And I used 7zip
@KS615 I believe it is doubly compressed. First to extract from the .xz, then from the .tar file.
KS615
@KS615
Mar 11 2017 20:58
what is 7zip? Is that a program to extract?
gregtcm
@gregtcm
Mar 11 2017 20:59
Like WinZip and WinRAR. I prefer 7zip, though. You can use ninite.com to download and install it and other useful utilites.
KS615
@KS615
Mar 11 2017 21:00
got it, downloading it now
Eric
@ecc1
Mar 11 2017 21:02
@gregtcm I don't use openaps, so my set-clocks.sh script calls different programs to set the time on the pump and the CGM. For openaps, use the oref0-set-device-clocks.sh script instead.
Mike
@libxmike
Mar 11 2017 21:04
@KS615 or rename the extention to .zip and use winrar or winzip but 7zip is better anyway
gregtcm
@gregtcm
Mar 11 2017 21:04
@ecc1 I got it. Thanks!
KS615
@KS615
Mar 11 2017 21:17
@libxmike @gregtcm, thank you both! I think its working now
KS615
@KS615
Mar 11 2017 21:34
Ok, back again with the next trouble. In my flash window, we typed, "flashall.bat" and got the response " 'flashall.bat' is not recognized as an internal or external command, operable program or batch file."
We have the PUTTY window open, provided the edison login, "root" and it displayed, "root@edison:~#" Any idea what we're doing wrong?
KS615
@KS615
Mar 11 2017 21:46
nevermind, we had two jubilinux folders so now i think we're back in business
hilarykoch
@hilarykoch
Mar 11 2017 21:54
trying to setup bluetooth. After the killall command, I’ve done: sudo /usr/local/bin/bluetoothd --experimental &
And I get: "D-Bus setup failed: Name already in use"
jcorbett80
@jcorbett80
Mar 11 2017 22:01
@gregtcm that is me working on it. However I need to figure out 1 more step before I start typing in the pr. I can not find a way to do multiple networks. I have checked online and I have watched in here. It seems to be a hard struggle. But I want it in the docs when I do the pr
Dana Lewis
@danamlewis
Mar 11 2017 22:03
I'd suggest pr'ing what you have now so we can start reviewing it and making it available to people to use! Can always add to it
jcorbett80
@jcorbett80
Mar 11 2017 22:03
@danamlewis I will get on the rest of what I have tommorrow and try to get it to you by night
@danamlewis sorry for the delay. long long week.
however, while I am in here for a short time. I am trying to run both my home wifi and my mobile hotspot. I want it to auto swop. I can not make it recognize 2 networks and autoswop. I have worked with networks most of my career. Has someone worked this out yet that can fill me in
Scott Leibrand
@scottleibrand
Mar 11 2017 22:07
@gregtcm @ecc1 even if you don't have bubbles in your reservoir, the plunger hits the angled part of the end of the reservoir (about 10U past empty) and starts compressing the rubber end of the plunger. because of that, the last few U you ask the pump to do, before it hits No Delivery, end up actually delivering progressively less and less of what you asked. It's not until it's basically delivering nothing that you get No Delivery, and at that the pump has actually delivered about 2U less than what the pump thinks it did, so you end up rising from the missed insulin. Much better to alert on empty reservoir (0 U left) and change it before you get to 10U past that.
@tekrulik and anyone else worried about time changes: if you want you can set your rig to a non-DST timezone during the day today, so it doesn't shift overnight. I'm going to switch ours from PST to Arizona time (MST) today, which will be the same as PDT tomorrow. If you're on the east coast, you can use a timezone like Panama, Jamaica, or Puerto Rico to match either EST or EDT.
Dana Lewis
@danamlewis
Mar 11 2017 22:16
@jcorbett80 no worries - just didn't want anyone to hold back because they were missing one piece, overall anything you can add will help improve things.
KS615
@KS615
Mar 11 2017 22:40
Hi all, back again. We are using Windows PC to flash the edison and are stuck on "ready to receive application" on the PUTTY console. Its been 30 min. Not using a Pi, what should we do?
John Benjamin
@Bender1061
Mar 11 2017 22:47
Hey what would everyone think about this 5 3 11 3 * * cd /root/<yourAPSDirectory> && oref0-set-device-clocks that way at 3:05 AM on March 11 it would then set your time on your pump for Spring forward?
opps I meant 5 3 12 3 * * cd /root/<yourAPSDirectory> && oref0-set-device-clocks forgot it's on the 12th
I figure that 5 mins would give it the time, for the Edison to have changed it's time before setting the pump
could do the same thing at for the fall too when daylight savings time ends, granted there is issues with you lucky countries that don't use daylight savings time.
maybe we could write another batch file that would look at your local timezone and decide if you need it or not
Eric
@ecc1
Mar 11 2017 22:52
That requires that you put the exact dates in twice a year. The version I posted earlier uses the "2nd Sunday of March / 1st Sunday of November" rule so you don't have to.
John Benjamin
@Bender1061
Mar 11 2017 22:53
wait, I must have missed that, did you post that today?
John Benjamin
@Bender1061
Mar 11 2017 22:54
@ecc1 yeah I like yours better!
cool I'm going to put that in my rigs and test that tonight
make the change for the standard setup, but awesome
Eric
@ecc1
Mar 11 2017 22:55
Your idea of 5 minutes after the changeover is a good one though -- mine is exactly at :00 so it might get missed somehow
John Benjamin
@Bender1061
Mar 11 2017 22:56
you should also delete your monitor folder during the end of daylight savings time, otherwise you have to wait an hour before it starts collecting new data
at least on the standatd setup
John Benjamin
@Bender1061
Mar 11 2017 23:01
# change to daylight time on second Sunday of March
5 3 8-14 mar * test $(date +\%u) -eq 7 && oref0-set-device-clocks

# change to standard time on first Sunday of November
5 1 1-7  nov * test $(date +\%u) -eq 7 && oref0-set-device-clocks && sudo rm -rf monitor/*
is what I'm thinking
Scott Leibrand
@scottleibrand
Mar 11 2017 23:02
you'll need to kill off any pump comms first to make sure you can set the pump clock successfully
(killall -g openaps; oref0-set-device-clocks)
Dana Lewis
@danamlewis
Mar 11 2017 23:03
Fwiw, since my chief concern is sleeping :smile: , I think I'd still prefer the manual change the next day vs risk it breaking the loop. But like where this is heading as an option, so do please let us know how it goes tomorrow
Scott Leibrand
@scottleibrand
Mar 11 2017 23:03
might also need some way to verify that the clock was set successfully
John Benjamin
@Bender1061
Mar 11 2017 23:03
true
Scott Leibrand
@scottleibrand
Mar 11 2017 23:03
I'm leaning towards always using non-DST timezones on our rigs...
John Benjamin
@Bender1061
Mar 11 2017 23:05
that probably would be easier overall, till they finally kill off daylight savings time.
but there might be ways in linux to tell if the time changed or not
but I'm thinking I'm going to try the above with the killall, and just find out if it works or not, If not, I'll just go on basic basal so it's not like any harm will be done.
hmm... zdump -v /etc/localtime | grep 2017
I've done very little with linux before OpenAPS, I'm really starting to like it now.
gregtcm
@gregtcm
Mar 11 2017 23:29
@jcorbett80 I'm struggling with multiple networks, as well. Mostly with the phone hotspot (wifi or bluetooth). I had the bluetooth working great a few days ago, but once it disconnected, it wouldn't come back. Haven't had a chance to work with it any more. My PR was going to be with hidden wifi networks. I had to add scan_ssid=1 to get it to connect to my hidden wifi.
jcorbett80
@jcorbett80
Mar 11 2017 23:30
@gregtcm I can get my hidden but I use a different command than in current docs. I will try to get my previous in tommorrow.
@gregtcm however if I add a second and I pull an ifconfig it shows multi broadcast but will never pull info and the rig will never connect to anything. I can remove either and the other will instantly start working and pull an address
gregtcm
@gregtcm
Mar 11 2017 23:47
@jcorbett80 Also trying to find an easy way to switch to different networks w/o always having to change ip addresses. Another reason I was hoping the Bluetooth tether would work and I'd always use that connection for my rig. Then I wouldn't have to deal with all the switching.
jcorbett80
@jcorbett80
Mar 11 2017 23:50
@gregtcm I am talking with Samsung. Monday we are supposed to talk again. It is definitely gone from all Galaxy s7 in the US if they are att. They are trying to fix.
I also want to BT tether
gregtcm
@gregtcm
Mar 11 2017 23:54
I'm on an S7 as well, but with a local company (Inland Cellular). G5 to the xDrip+ on my S7 and using both the Mongo upload and REST to my OpenAPS rig that is setup with the "xdrip" option. The double uploader seems like overkill, but its working and seems to work the best. I need to test more, though.
Going off my home wifi again, so I get to test some more with my hotspots. :)