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

30th
Apr 2017
aabramowicz
@aabramowicz
Apr 30 2017 00:06
@acezone its a 512 but will work
Billy Forester
@acezone
Apr 30 2017 00:08
Hmm.. takes the smaller res though right?
Paul Dickens
@thebookins
Apr 30 2017 00:20
Wondering if nightscout master or dev recommended for using openaps atm
Dana Lewis
@danamlewis
Apr 30 2017 00:35
master should be fine for general purpose OpenAPS usage
Paul Dickens
@thebookins
Apr 30 2017 00:52
Thanks dana ended up doing dev as that's what we used last time we upgraded. BTW autotune is still for use with 1 carb ratio only, isn't it?
susieanneh
@susieanneh
Apr 30 2017 00:53
Hi I am about to purchase a pump and I was wonderig, am i correct in thinking all minimed 522 pumps are compatible? the chart was a bit confusing
Dana Lewis
@danamlewis
Apr 30 2017 00:54
@thebookins correct
@susieanneh hey! correct, as far as we know, all 522s have been fine.
Geoff Aumaugher
@geoffro10
Apr 30 2017 00:54
Anyone ever have a problem with their OpenAPS pill not working in Nightscout? It just stopped last night and can't get it working. Pump is still interacting with the rig and setting temp basals. Just doesn't seem to be sending OpenAPS info to NS anymore. I've rebooted both rigs and the phone.
Dana Lewis
@danamlewis
Apr 30 2017 00:55
@geoffro10 did you check for future treatments in NS?
susieanneh
@susieanneh
Apr 30 2017 00:55
@danamlewis Thanks :)
Dana Lewis
@danamlewis
Apr 30 2017 00:55
:+1: :)
Geoff Aumaugher
@geoffro10
Apr 30 2017 00:56
@danamlewis Yep, checked that too. All is good.
Dana Lewis
@danamlewis
Apr 30 2017 00:56
k. how is the pill not working? disappeared, or has old info, or ? is everything else uploading to NS properly (basals rendering, etc.) or is it all missing from NS?
Geoff Aumaugher
@geoffro10
Apr 30 2017 01:01
OpenAPS pill says "unknown". Pump pill is blank. All treatments are being uploaded and appear to be working fine. NS logs even say the source is the pump.
Paul Dickens
@thebookins
Apr 30 2017 01:10
Any hacky ways of using autotune with multiple carb ratios through the day?
Dana Lewis
@danamlewis
Apr 30 2017 01:11
@geoffro10 have you dug into ns-loop yet?
@thebookins nope. options: 1) average them. 2) run multiple times w/ each of the different ratios (clearing out the profile file each time to reset it)
Paul Dickens
@thebookins
Apr 30 2017 01:11
Thinking about a "prepare" step that massages the data to look like it has a single carb ratio, and then a post-process step that transforms the results back to multiple carb ratios
Dana Lewis
@danamlewis
Apr 30 2017 01:12
3) be so annoyed that you take on multiple carb ratios with us. (I only use one, so need someone w/ good knowledge of how it should behave to help @scottleibrand and I adjust autotune to deal with multiples)
interesting
Paul Dickens
@thebookins
Apr 30 2017 01:12
kind of shoehorn it in
I am annoyed :smile: so (3) is an option
Geoff Aumaugher
@geoffro10
Apr 30 2017 01:12
@danamlewis Nope... Guess I better go read up on it.
Dana Lewis
@danamlewis
Apr 30 2017 01:13
@thebookins see openaps/oref0#326 for some background, and let’s switch over to https://gitter.im/openaps/autotune to discuss further :D
Paul Dickens
@thebookins
Apr 30 2017 01:13
OK
aabramowicz
@aabramowicz
Apr 30 2017 02:43
Trying to get BT tethering going. after using command "sudo killall bluetoothd" I get "sudo: unable to resolve host matthew" How to fix that?
Scott Leibrand
@scottleibrand
Apr 30 2017 02:44
that's because you didn't update your hostname correctly
it needs to be updated in both /etc/hosts and /etc/hostname
aabramowicz
@aabramowicz
Apr 30 2017 02:44
when I try to get there it says permission denied
Scott Leibrand
@scottleibrand
Apr 30 2017 02:44
the commands in the docs would have done that originally, but now that your rig is in an unknown state it's probably best to do so by editing those files manually
why aren't you logged in as root?
this is an Explorer board rig I presume?
aabramowicz
@aabramowicz
Apr 30 2017 02:45
I am...thats why I'm stuck. Explorer board yes
Scott Leibrand
@scottleibrand
Apr 30 2017 02:45
or are you trying to execute one of those files instead of editing it?
to edit you'd do something like nano /etc/hosts
aabramowicz
@aabramowicz
Apr 30 2017 02:49
I'll try again. Login: root Psswrd edison
Scott Leibrand
@scottleibrand
Apr 30 2017 02:55
if you're still at that stage, you shouldn't be doing bluetooth
bluetooth comes well after you've set up your proper password for root, wifi, etc.
what page (URL) and step are you at in the docs?
aabramowicz
@aabramowicz
Apr 30 2017 02:57
I've done all that and had this thing looping already. Now I'd like to connect to the iPhone hotspot via BT for use outside the house
Scott Leibrand
@scottleibrand
Apr 30 2017 02:57
but you never set a root password? that's a good way to get your medical device controlled by hackers...
seems like you missed a few steps during setup
aabramowicz
@aabramowicz
Apr 30 2017 03:12
does that mean reflashing jubilinux to set up new hostname and password?
rsilvers129
@rsilvers129
Apr 30 2017 03:29
IMG_5640.JPG.jpeg
Martin Haeberli
@mhaeberli
Apr 30 2017 03:32
@rsilvers129 new improved cooler version?
rsilvers129
@rsilvers129
Apr 30 2017 03:33
The other one is for a 2500. This is for a 2000 mAh battery.
IMG_5641.JPG.jpeg
Martin Haeberli
@mhaeberli
Apr 30 2017 03:34
cool - and posted already? (I just built one of the 2500 and am I hope building a second. Printer is across town so I cannot monitor progress)
rsilvers129
@rsilvers129
Apr 30 2017 03:41
This one is not done yet.
Martin Haeberli
@mhaeberli
Apr 30 2017 04:12
:+1:
trying to set up paper trail .but get 2017-04-30 04:12:05 CRITICAL remote_syslog.go:244 Failed to start server: No destination hostname specified
in response to remote_syslog
even though /etc/log_files.yml was set up
Martin Haeberli
@mhaeberli
Apr 30 2017 04:30
bluetooth tethering troubles … ssh into the device, fine. pinging 8.8.8.8 for example, fails. pinging www.apple.com intermittent . mostly getting “network is unreachable"
Scott Leibrand
@scottleibrand
Apr 30 2017 04:44
@aabramowicz if you want to follow the directions exactly, then you can reflash. If you think you can go back to the steps right after flashing and make sure everything is set up correctly, by figuring out what each step is trying to accomplish and either redoing it or making sure it has already been done correctly, that might be faster, by requires a bit more attention to detail.
katie disimone
@Kdisimone
Apr 30 2017 05:00
@mhaeberli can you post what you have for the log_files.yml? Maybe a cut paste error?
Martin Haeberli
@mhaeberli
Apr 30 2017 05:01
ok, the first line lacked “files"
let me try again, then post
katie disimone
@Kdisimone
Apr 30 2017 05:02
And double check both the components of the log file address that you have for papertrail. Sometimes it isn’t simply the port number than needs to change…but also that earlier part (log5)
Martin Haeberli
@mhaeberli
Apr 30 2017 05:03
better - that part was happier
still trying to parse out intermittent bluetooth / wlan routing ...
but that’s a different thing
katie disimone
@Kdisimone
Apr 30 2017 05:04
:+1:
katie disimone
@Kdisimone
Apr 30 2017 05:09
@scottleibrand I’m seeing a git corruption on someone’s logs. She’s running oref0-git-reset and the returned message is command not found. Ideas on why the command wouldn’t be found? Logged in as root.
Scott Leibrand
@scottleibrand
Apr 30 2017 05:10
It's oref0-reset-git
katie disimone
@Kdisimone
Apr 30 2017 05:13
damn it
thanks
i even copy and pasted it correctly in mine and ran it. could not figure out why her’s wasn’t working. reading…what a skill…;)
katie disimone
@Kdisimone
Apr 30 2017 05:22
@scottleibrand if that command doesn’t resolve the error (it’s still showing in the pump log after a reboot), would a runagain help?
Apr 29 22:20:41 edisonjoe pump-loop.log:  Traceback (most recent call last):
Apr 29 22:20:41 edisonjoe pump-loop.log:    File "/usr/local/bin/openaps-report", line 82, in <module>
Apr 29 22:20:41 edisonjoe pump-loop.log:      app( )
Apr 29 22:20:41 edisonjoe pump-loop.log:    File "/usr/local/lib/python2.7/dist-packages/openaps/cli/__init__.py", line 52, in __call__
Apr 29 22:20:41 edisonjoe pump-loop.log:      self.epilog( )
Apr 29 22:20:41 edisonjoe pump-loop.log:    File "/usr/local/bin/openaps-report", line 69, in epilog
Apr 29 22:20:41 edisonjoe pump-loop.log:      super(ReportToolApp, self).epilog( )
Apr 29 22:20:41 edisonjoe pump-loop.log:    File "/usr/local/lib/python2.7/dist-packages/openaps/cli/__init__.py", line 75, in epilog
Apr 29 22:20:41 edisonjoe pump-loop.log:      self.create_git_commit( )
Apr 29 22:20:41 edisonjoe pump-loop.log:    File "/usr/local/lib/python2.7/dist-packages/openaps/cli/__init__.py", line 99, in create_git_commit
Apr 29 22:20:41 edisonjoe pump-loop.log:      self.repo.git.gc(auto=True)
Apr 29 22:20:41 edisonjoe pump-loop.log:    File "/usr/local/lib/python2.7/dist-packages/git/cmd.py", line 425, in <lambda>
Apr 29 22:20:41 edisonjoe pump-loop.log:      return lambda *args, **kwargs: self._call_process(name, *args, **kwargs)
Apr 29 22:20:41 edisonjoe pump-loop.log:    File "/usr/local/lib/python2.7/dist-packages/git/cmd.py", line 877, in _call_process
Apr 29 22:20:41 edisonjoe pump-loop.log:      return self.execute(call, **exec_kwargs)
Apr 29 22:20:41 edisonjoe pump-loop.log:    File "/usr/local/lib/python2.7/dist-packages/git/cmd.py", line 688, in execute
Apr 29 22:20:41 edisonjoe pump-loop.log:      raise GitCommandError(command, status, stderr_value, stdout_value)
Apr 29 22:20:41 edisonjoe pump-loop.log:  git.exc.GitCommandError: Cmd('git') failed due to: exit code(255)
Apr 29 22:20:41 edisonjoe pump-loop.log:    cmdline: git gc --auto
Apr 29 22:20:41 edisonjoe pump-loop.log:    stderr: 'Auto packing the repository in background for optimum performance.
Apr 29 22:20:41 edisonjoe pump-loop.log:  See "git help gc" for manual housekeeping.
Apr 29 22:20:41 edisonjoe pump-loop.log:  fatal: Unable to create '/root/myopenaps/.git/packed-refs.lock': File exists.
Apr 29 22:20:41 edisonjoe pump-loop.log:  If no other git process is currently running, this probably means a
Apr 29 22:20:41 edisonjoe pump-loop.log:  git process crashed in this repository earlier. Make sure no other git
Apr 29 22:20:41 edisonjoe pump-loop.log:  process is running and remove the file manually to continue.
Apr 29 22:20:41 edisonjoe pump-loop.log:  error: failed to run pack-refs’
Scott Leibrand
@scottleibrand
Apr 30 2017 05:24
Perhaps, but it'd be better to figure out what state the system is in that oref0-reset-git can't detect the corruption.
Also, just rm -rf ~/myopenaps/.git should allow oref0-reset-git to re-do the git repo and fix it.
katie disimone
@Kdisimone
Apr 30 2017 05:24
51522204370__6F7030E0-7808-4905-8B6C-762229F0C71D.JPG
i’ll try the rm
Scott Leibrand
@scottleibrand
Apr 30 2017 05:45
Hold on.
Martin Haeberli
@mhaeberli
Apr 30 2017 05:45
@Kdisimone how full is your openaps “disk” - try df -h
Scott Leibrand
@scottleibrand
Apr 30 2017 05:45
That is being run in the homedir
It needs to be run in ~/myopenaps
Martin Haeberli
@mhaeberli
Apr 30 2017 05:46
on my second day with this new rig (0.1.0 kernel) my ‘disk’ filled up, and I got something looking a lot like the above errors
@scottleibrand @Kdisimone fyi, the git status showed ‘not a git repository’ when the disk was full; i was about to destroy the dir and start over, then found disk was full, then fixed log overflow, then somehow git was happy, too
just sayin’ - your openaps may vary, of course
katie disimone
@Kdisimone
Apr 30 2017 05:48
oref0-reset-git needs to be run in myopenaps dir?
Scott Leibrand
@scottleibrand
Apr 30 2017 05:50
Yes
Take a look at how cron runs it.
katie disimone
@Kdisimone
Apr 30 2017 05:50
i’ll get that PR’d in. :)
Scott Leibrand
@scottleibrand
Apr 30 2017 05:50
Honestly it shouldn't need to be run manually all all
*at all
Only reason to do so is to figure out why it's not working automatically.
katie disimone
@Kdisimone
Apr 30 2017 05:51
hmmm…well, i suppose that would make sense too…this seems like it’s an error that that command isn’t fixing. So, manually running it shouldn’t be fixing it, considering.
Martin Haeberli
@mhaeberli
Apr 30 2017 05:51
@Kdisimone I’ll re-iterate - can you have her check df -h
while deferring to @scottleibrand s wisdom
katie disimone
@Kdisimone
Apr 30 2017 05:52
I need help understand the whole drill down, figure out part of troubleshooting in openaps. admittedly, i’m trying hard to get there, but it’s not quite clear to me yet...
yes @mhaeberli …will do too
Scott Leibrand
@scottleibrand
Apr 30 2017 05:53
In this case that script checks a bunch of stuff to see if the repo looks corrupt. If the commands it's checking don't indicate corruption, it won't take drastic action to fix things.
Martin Haeberli
@mhaeberli
Apr 30 2017 05:53
@Kdisimone thx - so NOW that my logs are rolling hourly, 4 logs, compressed, once per hour, yada yada (haven’t yet put in new kernel) - I get this in response to df -h:
Filesystem       Size  Used Avail Use% Mounted on
/dev/root        1.4G  795M  521M  61% /
devtmpfs         480M     0  480M   0% /dev
tmpfs            481M     0  481M   0% /dev/shm
tmpfs            481M   49M  432M  11% /run
tmpfs            5.0M     0  5.0M   0% /run/lock
tmpfs            481M     0  481M   0% /sys/fs/cgroup
tmpfs            481M   12K  481M   1% /tmp
/dev/mmcblk0p7    32M  5.2M   27M  17% /boot
/dev/mmcblk0p10  1.3G  799M  492M  62% /home
tmpfs             97M     0   97M   0% /run/user/0
key is, in failure mode, Avail in first useful row showd 0 vs say 500 - 800 M
John Males
@johnmales
Apr 30 2017 07:40
Hi all, I’m having some trouble with direct connection of a G5 receiver via USB to a raspberry pi. I’ve not been able to pull glucose values from the receiver. From the cgm-dexusb-loop.log:
return self.method(args, app)
  File "/usr/local/lib/python2.7/dist-packages/openaps/uses/use.py", line 45, in __call__
    output = self.main(args, app)
  File "/usr/local/lib/python2.7/dist-packages/openaps/vendors/dexcom.py", line 532, in main
    iter_glucose = self.get_glucose_data(params, args)
  File "/usr/local/lib/python2.7/dist-packages/openaps/vendors/dexcom.py", line 513, in get_glucose_data
    return itertools.takewhile(self.comparison, self.dexcom.iter_records('EGV_DATA'))
AttributeError: 'NoneType' object has no attribute ‘iter_records’
cameronrenwick
@cameronrenwick
Apr 30 2017 13:50
anyone have any trouble with rolling on their pump during sleep and hitting a button which put the pump in a stalled state (from a looping point of view). Woke up and looked on my phone to see I hadn't been looping for a few hours... last thing I looked at was my pump (duh) and realized it was flashing expecting a carb number input. I'm sort of thinking of installing some sort of ~2mm high plastic circle around each button so that you have to put your finger "inside" the circle to actuate the button. Anyone else have this happen? I keep my pump on the waistband of my pjs ... can't leave it under pillow or other place as I wake up tangled!
FBH1606
@FBH1606
Apr 30 2017 13:51
Hi all, I have autotune enabled as an advanced feature, I would like to look at a report to see if I should make some adjustments to my pump settings. If I use the command cd ~/myopenaps/autotune && cat autotune_recommendations.log I get a No such file or directory error. What am I doing wrong? (I cut and paste, so no typos in terminal).
cameronrenwick
@cameronrenwick
Apr 30 2017 13:58
your directory where you have openaps is called "myopenaps" correct? If no, that dir will need to be edited to reflect what you've called it during setup
FBH1606
@FBH1606
Apr 30 2017 14:00
I've tried that, and get the same error.
Actually, no, it brings up a prompt at my directory. What command do I put in at that prompt for a report?
FBH1606
@FBH1606
Apr 30 2017 14:06
root@finsedison:~/finsaps/autotune#
Matthias
@ceben80
Apr 30 2017 14:08
@FBH1606 for autotune related issues you might swap to https://gitter.im/openaps/autotune :wink:
FBH1606
@FBH1606
Apr 30 2017 14:08
Oh sorry, I thought this was the one we were supposed to use. My apologies!
cameronrenwick
@cameronrenwick
Apr 30 2017 14:09
aside from switching to AAA lithium batteries for the pump, has anyone looked at hacking a larger install of (at at minimum) AA sized battery? Admittedly I've got a large pack of AAA alkaline batteries that I'm going thru at a 1-a-day pace. I wonder if hacking a hole thru the battery cap and somehow arranging a AA lithium alongside the pump. Just wondering before I go and reinvent the wheel
Matthias
@ceben80
Apr 30 2017 14:11
@FBH1606 np maybe you'll find a solution in this room by using the search function :wink:
Lynne Butts
@flutelynne_twitter
Apr 30 2017 14:50
@cameronrenwick you could lock the pump, then buttons will not press. Act>Utilities>Lock Keypad
sysyc
@sysyc
Apr 30 2017 16:04
Hi everyone!If I used Medtronic Enlite Sensor and what can I chose when I run oref0-setup? MDT?
Eric
@ecc1
Apr 30 2017 16:07
@cameronrenwick I keep my pump in one of the slightly padded Medtronic "sports cases" 24x7, and have never had any problems sleeping on it. But Lock Keypad is probably better as long as you remember to do it each night. (I don't think there's a remote command for that, or you could do it via a cron job :-)
alternateal
@alternateal
Apr 30 2017 16:14
@sysyc - yes, MDT for Medtronic integrated CGM
sysyc
@sysyc
Apr 30 2017 16:23
@alternateal Thank you!
Patrik Nilsson
@ellabellakaramella
Apr 30 2017 16:29
Open A.P.S question: how often is it supposed to udate status in NS? every 5 min? Because mine does it very infrequent, from 5 to 20 min so far. It is in range all the time. Just wondering.
cameronrenwick
@cameronrenwick
Apr 30 2017 16:33
@ecc1 @flutelynne_twitter admittedly I have not tried the button lock; great idea... that won't stop any looping/basal aspects then?
Scott Leibrand
@scottleibrand
Apr 30 2017 17:01
Being in the bolus wizard carb entry screen should not interfere with looping.
Jeffrey Davis
@LunarseaNY_twitter
Apr 30 2017 17:17
Hi! Quick question: if we want to connect the Dexcom 4 receiver to the rig, do we need a USB-B to USB-B male/male cord and if so where's the best place to get it. On line? Didn't see one at BestBuy or Target and Radio Shacks are closing all over.
katie disimone
@Kdisimone
Apr 30 2017 17:25
We were never successful in finding a male/male cord. Instead look for an OTG USB-b cord. You'll plug that into the receiver and then a regular microB from OTG cable to board
John Kelleher
@jkelleher
Apr 30 2017 17:33
@ellabellakaramella If you look in the crontab -l you'll see ns-loop updating the NS site every minute. In practice I find it hard to tell if it keeps to that 'schedule' as there are often 'errors' communicating with the pump. This could be our PWD leaving the rig a little too far away from her pump.
Also away from home when using bluetooth tethering you're relying on a longer 'chain' - CGM (Medtronic)->pump<->rig<->iPhone->NS Site. If you're monitoring the other end (NS site) you may find you have to wait 10-15 minutes or longer sometimes.
@sysyc we use Enlite too (not many do) so happy to share any experiences.
Scott Leibrand
@scottleibrand
Apr 30 2017 17:35
MicroUSB OTG male-male: BestGameSetups Micro USB to Micro USB OTG (On-The-Go) 12" (30cm) Data Cable https://www.amazon.com/dp/B00TQOEST0/ref=cm_sw_r_cp_api_1aIbzb6D6SFF7
Or adapters to use a regular microUSB cable: CHENYANG Ultra Mini DM Micro USB 5pin OTG Adapter Connector for Cell Phone Tablet & USB Cable & Flash Disk - 5pcs https://www.amazon.com/dp/B015GZOHKW/ref=cm_sw_r_cp_api_GbIbzbATPF7KX
Someone want to PR those (and any other) links into the docs please?
sysyc
@sysyc
Apr 30 2017 17:57
@jkelleher :+1:
katie disimone
@Kdisimone
Apr 30 2017 18:08
@scottleibrand will do
just got hotspot setup too, had to switch from tracfone(verizon) to boom-mobile(verizon) to get hotspot, its working also! good deal: http://www.boom.us/1gb-data-w/unlimited-talk-text-r4g (only verizon works at his school)
overnight went ok again, testing daytime now (w/pancakes too ;)
susieanneh
@susieanneh
Apr 30 2017 19:41
Hi I'm new to looking into looping. With open aps is there an iPhone app that you can use to control/view settings?
kenkotch
@kenkotch
Apr 30 2017 19:44
@susieanneh you will use nitescout to visualize what is happening and some ittt you can use to control
Dana Lewis
@danamlewis
Apr 30 2017 19:44
@susieanneh no, you can use a different kind of app to look at the logs/change settings remotely on the rig; and use Nightscout to most easily visualize everything that's going on.
susieanneh
@susieanneh
Apr 30 2017 19:46
Thank you all:)
Milos Kozak
@MilosKozak
Apr 30 2017 19:48
anyone tested to increase max autosens multiplier (1.2) for child ?
susieanneh
@susieanneh
Apr 30 2017 19:49
So with this you set up your rig and set it what you need and let it go all day. Making changes from your computer if needed. It acts on its own, no need to tell it ok to make changes to basals?
PieterGit
@PieterGit
Apr 30 2017 19:52
@Kdisimone @scottleibrand : i got these otg cables:
http://www.ebay.com/itm/301228419677?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT (10cm Right Angle Micro USB B to Micro USB B at both ends Host OTG Adapter Cable) and
http://www.ebay.com/itm/301470446035?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT (20cm Right Angle)
Dana Lewis
@danamlewis
Apr 30 2017 19:53
@susieanneh correct, basal adjustments are automated. You can also use ifttt widgets on your watch or phone to easily tell it about carbs (if you don't want to enter them on pump), or easily change targets (i.e. If you're planning to go for a long run). But otherwise, it'll just be responding with basal adjustments as needed to changes in BGs without you needing to do anything.
susieanneh
@susieanneh
Apr 30 2017 19:54
Awesome! I'm so thankful for people like you helping me ❤
Dana Lewis
@danamlewis
Apr 30 2017 19:55
:smile: :+1:
@MilosKozak I think @Kdisimone might have adjusted hers (for her 14yo)
Milos Kozak
@MilosKozak
Apr 30 2017 20:01
our is 7yo and sometimes we'd need something like 2 .... i'd like to know what is tested as safe
Dana Lewis
@danamlewis
Apr 30 2017 20:04
Not sure we have any data or experiences to point to what's safe beyond the standard cap - going to vary so much person to person.
katie disimone
@Kdisimone
Apr 30 2017 20:13
@MilosKozak yup…I’m having good (actually, great) results with my multipliers set to 1.4 and 0.5. We have a 14-yo girl and she swings pretty wide. We are only running autosens, not autotune.
@MilosKozak I’d gone up to 1.6 too for awhile…just found we weren’t using it that much. But, I do feel it actually would have some added benefit at certain times of the month. I just haven’t adjusted back to it yet since the last time I updated preferences. We have had some really nice results with autosens.
Milos Kozak
@MilosKozak
Apr 30 2017 20:23
thanks. currently we have huge insulin resistance. it happens really fast. from great BGs to highs which are not possible to lower down within a hour
thinking about a shorter time range for detecting sensitivity
Milos Kozak
@MilosKozak
Apr 30 2017 20:28
@danamlewis btw new code for deviation calculation now works for a small child really well
original oref0 code was useless for us
tynbendad
@tynbendad
Apr 30 2017 20:45
so, what happens with autosens and autotune in the face of multiple rigs? does each rig only adjust for the times it controls you? so each rig could have its own basal rates &etc?
PieterGit
@PieterGit
Apr 30 2017 20:52

@Bender1061 Did you get a solution for the BUG: sleeping function called from invalid context at kernel/mutex.c issue,
https://gitter.im/nightscout/intend-to-bolus?at=58703d0c7a3f79ef5de69d91

I get a lot of:

[  159.166679] BUG: sleeping function called from invalid context at kernel/mutex.c:95id="MY_SSID_NAME"
[  159.166766] in_atomic(): 1, irqs_disabled(): 0, pid: 76, name: krfcommd

with a wpa_roam wpa_supplicant config in /etc/network/interfaces on a Jubilinux 0.1.1 Edison

auto wlan0
iface wlan0 inet dhcp
    wpa-roam /etc/wpa_supplicant/wpa_supplicant.conf

when i replace wpa-roam to wpa-conf it does not seem to occur. Does anybody use a wpa-roam profile with their jubilinux edison image? It works fine an RPI

Dana Lewis
@danamlewis
Apr 30 2017 20:53
@tynbendad it calculates based on the data it has. So yes, if you have a rig turned off for days and turn it on, it may not have the same autosens that your other rig has
If they're all running and all have data, should be close to each other
katie disimone
@Kdisimone
Apr 30 2017 20:59
@MilosKozak we have same issues…isf turning on a dime. hormones swing us UP in insulin needs. her track practice after school sinks us into very little insulin needs for several hours and into night.
Dana Lewis
@danamlewis
Apr 30 2017 21:01
@MilosKozak which?
tynbendad
@tynbendad
Apr 30 2017 21:03
@danamlewis but if you use one always plugged in, in the bedroom and another always mobile e.g., for when away from the house, then they'll each have a profile only for their own general timeframes... e.g., bedroom one will mainly know about nighttime basals and mobile one will only know about daytime basal...
or do they each keep up with everything as long as they are connected to NS?
tynbendad
@tynbendad
Apr 30 2017 21:12
the bluetooth tethering worked great when we left the house, but when we came back it seems stuck on... rig isn't switching back to wifi - i guess this would be a good use of papertrail (which i didn't set up)
katie disimone
@Kdisimone
Apr 30 2017 21:16

@tynbendad
If you toggled the hotspot switch to on in order for it to connect when you left the house…then you didn’t really use a BT tether…you used a wifi-tether and that will not automatically turn off and go back to home wifi.

If you used the BT-tether (didn’t toggle hotspot and the rig just automatically connected when you left home) and it didn’t automatically connect to home wifi, usually you’ll see a spidev already in use error in papertrail. That happens sometimes when the rig switches from BT to wifi for us. There’s an automatic reboot that will happen after a certain period of time when that error happens. But, it takes some time for the rig to detect the error and reboot itself. So, let it sit for 15-20 min and see if the rig catches the error itself. (if it doesn’t, there’s a backup reboot script that you can add to help with that).

get that papertrail up and running…it’s invaluable for a new setup

katie disimone
@Kdisimone
Apr 30 2017 21:16
@tynbendad our mulitple rigs are plugged in all the time and all sync to same nightscout. they are all in sync as a result, yes.
rsilvers129
@rsilvers129
Apr 30 2017 21:20
@Kdisimone how do you keep multiple rigs from sending commands to the pump?
Scott Leibrand
@scottleibrand
Apr 30 2017 21:20
@tynbendad as long as all your rigs are on at midnight, they'll all get the same data from NS and will all end up with approximately the same autotune results
@rsilvers129 we've taught them good southern manners
Dana Lewis
@danamlewis
Apr 30 2017 21:21
@tynbendad yes, if it has wifi it’ll pull the data to the rig
Scott Leibrand
@scottleibrand
Apr 30 2017 21:21
@rsilvers129 if another rig is talking to the pump, then the rig will wait-for-silence before taking its turn
Dana Lewis
@danamlewis
Apr 30 2017 21:21
@rsilvers129 it’s designed for multiple rigs to be able to talk to the pump. that way if you walk into range of one and out of the other, you don’t just stop looping. and if you’re in range of multiple at the same time, that’s why you see it “listening” for other pump comms, so it waits-for-silence before trying to talk to the pump
rsilvers129
@rsilvers129
Apr 30 2017 21:24
That is very nice. I think I read all the docs and I didn't know any of this.
Dana Lewis
@danamlewis
Apr 30 2017 21:26
sounds like you have some PRs in mind to make, then ;)
rsilvers129
@rsilvers129
Apr 30 2017 21:31
Once I use it for a while, sure.
(for those of you who’ve seen us mention in passing “SMB” or “oref1” - you can also check out the OpenAPS reference design page, https://openaps.org/reference-design/, which now reflects the changes to include the differences between oref0 and oref1. If you’re interested in testing SMB, we’ll let everyone know once it’s in dev (soon) for testing.)
rsilvers129
@rsilvers129
Apr 30 2017 21:37
Is there any way to get ultra fast acting insulin?
Dana Lewis
@danamlewis
Apr 30 2017 21:38
Fiasp if you want to drive to Canada or go to europe?
tynbendad
@tynbendad
Apr 30 2017 21:41
@Kdisimone @scottleibrand @danamlewis thanks - multi rigs makes more sense now ;).
using just bluetooth tethering, not hotspot. but, i just saw it on wifi and it went away, so it looks like its having trouble sticking on our wifi at home, i'll have to figure that out. thanks
@danamlewis holy cow, oref1 sounds intriguing, will be reading!
katie disimone
@Kdisimone
Apr 30 2017 21:43
@tynbendad we have one room in our house with dicey wifi stability. When anna hangs out in that room, the rig that's on her will periodically toggle between home wifi and BT tether. Just a bad spot in the house for home wifi
Dana Lewis
@danamlewis
Apr 30 2017 21:46
@tynbendad :D it’ll be a week or two before it’s in dev for testing more broadly
@tynbendad also - thanks, I knew we had put a note somewhere. to @rsilvers129 point about not seeing it , just stuck it up higher in the “understanding your rig” page to make it more obvious. Additional PRs welcome w/ further notes around that. http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-0/understanding-your-Explorer-Board-rig.html#what-happens-if-you-have-multiple-rigs
tynbendad
@tynbendad
Apr 30 2017 21:49
@danamlewis great, i've found the search feature for the docs invaluable ;)
Dana Lewis
@danamlewis
Apr 30 2017 21:49
:D :D :D
Milos Kozak
@MilosKozak
Apr 30 2017 21:50
@danamlewis openaps/oref0#430
Dana Lewis
@danamlewis
Apr 30 2017 21:51
@MilosKozak gotcha - wanted to make sure I was thinking of the right thing. yes, wanting someone to help port it back to js so we can give it a spin :)
Milos Kozak
@MilosKozak
Apr 30 2017 21:54
+++(21)++++++++++++(22)+++++-CCCCCC(23)CCCCCCCCCCCC(0)CCCCCCCCCCCC(1)CCCCCC++++++(2)++++++++--++(3)++++++++++++(4)-++++++++++-(5)+++--+-+----(6)--+=----+++-(7)+-+-----++++(8)++----------(9)----+CCCCCCC(10)CCCCCCCCCC++(11)++++++++++++(12)+CCCCCCCCCCC(13)CCCCCCCCCCCC(14)++++++CCCCCC(15)CCCCCCCCCCCC(16)C+++++++++++(17)+-----++++++(18)CCCCCCCCCCCC(19)CC++++++++++(20)++++++++++++(21)++++++++++++(22)++++++++++++(23)+---------
oref0 would treat this all as carbs absorption
Martin Haeberli
@mhaeberli
Apr 30 2017 22:03
@PieterGit India Times article seems to be about a University of Virginia research system - no direct connection to openaps as far as I can tell.
rsilvers129
@rsilvers129
Apr 30 2017 22:04
What is the general consensus as to how many grams of carb one can have with a single meal and have the algorithm handle it nicely and not spike over 200?
Dana Lewis
@danamlewis
Apr 30 2017 22:04
@rsilvers129 general consensus says: you need to do a meal bolus, plus eating soon mode to have lesser meal spikes. you might also want to use AMA, as well
Martin Haeberli
@mhaeberli
Apr 30 2017 22:04
is there an “Explorer” channel? I have a question about charging; I have a 2000 mAH battery steady-state showing 67% but charging has stopped.
tynbendad
@tynbendad
Apr 30 2017 22:08
@danamlewis i know AMA can increases temp basal after you enter carbs, but not really sure what it bases that on - is there a writeup of that somewhere? it seems like it could do something similar to oref1 with just temp basals? e.g., it could lower target by the amount of "catch-up" 0-temping that you think you can handle later.
Dana Lewis
@danamlewis
Apr 30 2017 22:08
exactly - it’s temp basal improvements during post-meal time since it knows about carbs.
realized recently that’s what in the docs isn’t enough detail (http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-4/advanced-features.html#advanced-meal-assist-or-ama) and not everyone loves reading our old issues about the original development idea of it :D (openaps/oref0#68, for example)
so it’s on our to-do list to improve what’s in the docs with regards to how AMA works
SMB, though, builds upon what we created for AMA, which was dynamic carb absorption.
tynbendad
@tynbendad
Apr 30 2017 22:13
thanks - the issue is great for me, you could never write up that progression of ideas as well ;)
Dana Lewis
@danamlewis
Apr 30 2017 22:13
so even if you did put in a carb amount, but your BGs were rising or dropping faster than it expected based on that amount at that time, it would adjust temps as needed, vs relying on a fixed notion of “here’s exactly going to happen w/ this amount of carbs over the next X hours"
:+1: also search in the issues repo for “advanced meal assist” and “AMA” to see other iterations on it, too
tynbendad
@tynbendad
Apr 30 2017 22:19
i haven't finished reading it but i was thinking i'd like an AMA that high temp'd the maximum possible, i.e., until just before 0-temping wouldn't be possible to recover from low. (ok, maybe not quite that much, but something i could adjust... automatically setting target low, so no intervention needed).
(this is because i have a kid that won't prebolus of course ;)
Scott Leibrand
@scottleibrand
Apr 30 2017 22:23
sounds like you want to read up on the new oref1 / smb stuff.
Dana Lewis
@danamlewis
Apr 30 2017 22:25
It will be good for you to get comfy w oref0 before you test oref1 features - but ya, I think they'll work well for some of the scenarios you've described
Martin Haeberli
@mhaeberli
Apr 30 2017 22:34
@danamlewis thx :+1: so shows 93% when unplugged - d'oh
tynbendad
@tynbendad
Apr 30 2017 22:49
:+1:
PopcornPam
@PopcornPam_twitter
Apr 30 2017 23:32
@danamlewis , oref1 sounds great! Eating soon works great, but not so great for the unexpected high carbs (AKA when there is a random cupcake etc at work, or or the craving for that cookie is just overpowering)
I am getting a hold of some FIASP, and curious about how to use this with openaps
Though I am bionic til MAy 25, so I won't be using it before then
Dana Lewis
@danamlewis
Apr 30 2017 23:34
@PopcornPam_twitter ya, I’ve found it great in scenarios like that. May have had some spike, but that sharp rise is often blunted by 160 for me when SMB is active.
PopcornPam
@PopcornPam_twitter
Apr 30 2017 23:35
Great! I can't wait to try it out
Dana Lewis
@danamlewis
Apr 30 2017 23:35
For Fiasp, I think different people have adjusted their DIA for looping, not sure if any other changes end up happening. (I’d ask @tim2000s closer to when you are able to try it to see if he’s changed anything else beyond DIA at that point, since he’s done the most Fiasp+looping testing)
PopcornPam
@PopcornPam_twitter
Apr 30 2017 23:37
I am also super curious to see if it makes much of a difference, since bg has been pretty flat since getting my settings set in openaps
BTW, day 3 of beta bionics, I am finding that while I am going higher after food than I do with openaps, the advantage is not having to do a gosh darn thing about it... and not constantly second guessing problem BGs of did i mess up something with my install
But also, I assume that many of the issues I have now would not be an issue if I could change target bg, which the current study does not allow
Dana Lewis
@danamlewis
Apr 30 2017 23:41
:+1: so cool for you to be able to have both kinds of experiences
(also have some upcoming oref1 features that also might make it an even more interesting comparison for you in terms of ease of use around mealtime)
PopcornPam
@PopcornPam_twitter
Apr 30 2017 23:41
I have been running my openaps rig un connected at the same time, but i am not sure it is getting any worthwhile data
Lynne Butts
@flutelynne_twitter
Apr 30 2017 23:49
Wow! Just read about oref1. Amazing.