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

17th
Jul 2016
Jason Calabrese
@jasoncalabrese
Jul 17 2016 00:20
Welcome @OrlaWilson , try running df -h on the pi
OrlaWilson
@OrlaWilson
Jul 17 2016 00:22
thanks! looks like /dev/root is at 100% (5.7Gused).
Jason Calabrese
@jasoncalabrese
Jul 17 2016 00:38
there's usually a bunch of junk you can uninstall, there might be something in the docs but I'm not seeing it
OrlaWilson
@OrlaWilson
Jul 17 2016 00:40
OK, I'll try uninstalling the junk. Thanks!
Jason Calabrese
@jasoncalabrese
Jul 17 2016 00:41
you might also be able to resize the partition with raspi-config depending on the size of your sd card
Dana Lewis
@danamlewis
Jul 17 2016 00:42
@cjo20 oh yea. (N=1)*96!
Chris Oattes
@cjo20
Jul 17 2016 00:42
yay!
@danamlewis thanks
Dana Lewis
@danamlewis
Jul 17 2016 00:43
:)
sarahspins
@sarahspins
Jul 17 2016 00:52
@danamlewis what is your estimate on average time for loops in progress to getting to where it "counts" :)
Dana Lewis
@danamlewis
Jul 17 2016 00:52
It Depends ;)
Some people spend weeks in open loop, some people take more time to build, etc. so no good average.
sarahspins
@sarahspins
Jul 17 2016 00:53
ok :)
any records for fastest to closed? (I'm not setting a goal, just satisfying curiosity)
Chris Oattes
@cjo20
Jul 17 2016 00:54
I've been dabbling with OpenAPS for a year or more. I've been open-looping for several weeks, I could have closed it earlier but I wanted to get a better feel for how the loop worked and made suggestions
Dana Lewis
@danamlewis
Jul 17 2016 00:55
No records, that's not important :)
(Puts a ban in this channel on people touting complete times. Needs to be done based on safety, comfort, and understanding, not arbitrary time counts!)
sarahspins
@sarahspins
Jul 17 2016 00:55
well as I'm sitting here trying to flash my edison for the 29th time today, it's what pops into my head :P
and yes, I understand what you're saying :)
Dana Lewis
@danamlewis
Jul 17 2016 00:56
:)
Hannah Crabtree
@lollydaggle
Jul 17 2016 00:56
I'm just excited because I'm doing the :pizza: test
Dana Lewis
@danamlewis
Jul 17 2016 00:57
Everyone loves the (pizza/Chinese/your crazy food with high carb amount) test ;)
Chris Oattes
@cjo20
Jul 17 2016 00:58
I've been stuck at 12 mmol/l for a few hours now. Chinese seems to last a long time. But that is as high as I got, which is good for a meal like that.
fjungman
@fjungman
Jul 17 2016 01:04
Posted this over on RileyLink, but reposting here for the more-general crowd. This is Riled-Up, a small board 1.6" x 2.2" that an Intel Edison plugs into.
alt
CC1110 with chip antenna, 24MHz crystal. USB OTG. Detects if battery has three-pin connection with thermistor, and switches to high-current charging (1.5A). Solder area to connect Wixel or other things. Setup for programming the CC1110 from the Edison, if that can be done.
Jeremy M Lucas (GetRileyLink.org)
@jlucasvt
Jul 17 2016 01:05
@danamlewis @sarahspins :+1: it's the journey that's the secret to life.
Dana Lewis
@danamlewis
Jul 17 2016 01:06
So true :)
Jason Calabrese
@jasoncalabrese
Jul 17 2016 01:07
nice @fjungman, the back is smooth?
fjungman
@fjungman
Jul 17 2016 01:07
Yes, other than the through-hole solder points, of which there are few.
No components back there.
sarahspins
@sarahspins
Jul 17 2016 01:08
@fjungman that picture seems way larger than life :)
Pete Schwamb
@ps2
Jul 17 2016 01:08
Thanks! :clap: I couldn’t do near the job @fjungman did on this. Really appreciate it!
fjungman
@fjungman
Jul 17 2016 01:08
Yes, it is a little bigger than a quarter.
I'm just trying to give back, since I can't program anymore, apparently...
Now, to see if it works... :)
Jason Calabrese
@jasoncalabrese
Jul 17 2016 01:09
solder points for wixel sound great
Paul Martin
@Spazholio
Jul 17 2016 01:10
Is there any way I can ask that you post that pic on Imgur or paste it in here? I’d love to see it, but it’s making me login to Live, which is problematic...
fjungman
@fjungman
Jul 17 2016 01:10
Oh, that funny looking edge-connector is for programming the CC1110 for now. You just slide a edge-connector attached to a ribbon cable from the CC-debugger.
Can you paste an image here?
Jason Calabrese
@jasoncalabrese
Jul 17 2016 01:10
should be able to drag and drop
just doesn't work from mobile
Pete Schwamb
@ps2
Jul 17 2016 01:11
The board also has gpios hooked up to the cc1110 programming pins, so we could in theory bit-bang firmware into the cc1110 w/o a programmer.
fjungman
@fjungman
Jul 17 2016 01:11
Riled Up 3D.JPG
Jeremy M Lucas (GetRileyLink.org)
@jlucasvt
Jul 17 2016 01:11
Wow that's tiny
sarahspins
@sarahspins
Jul 17 2016 01:11
the thumbnail looks more appropriately sized :)
fjungman
@fjungman
Jul 17 2016 01:11
Damn. That would have saved me hours of trying to post pictures here....
Paul Martin
@Spazholio
Jul 17 2016 01:12
What? No banana for scale? :)
Yeah, that looks tiny as hell. Very cool, man.
fjungman
@fjungman
Jul 17 2016 01:12
For reference, the pink things on the top are the two-pin and three-pin JST battery connectors that are on all of the batteries we use for xDrip, etc. The ones from Adafruit...
Jeremy M Lucas (GetRileyLink.org)
@jlucasvt
Jul 17 2016 01:12
What's U5/U6 for?
Chris Oattes
@cjo20
Jul 17 2016 01:13
any chance of posting schematics?
fjungman
@fjungman
Jul 17 2016 01:13
U5 and U6 are level-shifters.
Yes, let me get a schematic up here.
Lynne Butts
@flutelynne_twitter
Jul 17 2016 01:13
@cjo20 congrats! I often do the temp target for eating soon to get through those high carb things.
fjungman
@fjungman
Jul 17 2016 01:14
Assembly Drawings.pdf
Jeremy M Lucas (GetRileyLink.org)
@jlucasvt
Jul 17 2016 01:14
It's like Christmas in July. @ps2 sends me 30 RL's
And @fjungman shows us the future of Edison coms
fjungman
@fjungman
Jul 17 2016 01:16
Riled-Up Rev A0.pdf
Ho Ho Ho... :)
Chris Oattes
@cjo20
Jul 17 2016 01:17
Which software is that?
and thanks
Jason Calabrese
@jasoncalabrese
Jul 17 2016 01:17
wow, really exciting
Chris Oattes
@cjo20
Jul 17 2016 01:19
@fjungman what is FL1?
fjungman
@fjungman
Jul 17 2016 01:20
The band filter for the radio.
Chris Oattes
@cjo20
Jul 17 2016 01:20
a chip version of the capacitor network on the CC1111 stick?
fjungman
@fjungman
Jul 17 2016 01:21
Yes, a SAW filter, which is much more repeatable and temp-stable.
Chris Oattes
@cjo20
Jul 17 2016 01:28
Why the level shifters on the CC1110 interface (sorry, last question :P)
fjungman
@fjungman
Jul 17 2016 01:28
1.8V to 3.3V shift.
Chris Oattes
@cjo20
Jul 17 2016 01:29
Ah, I thought Edison GPIOs were 3v3 capable
fjungman
@fjungman
Jul 17 2016 01:29
Oh, that is to allow the Edison to program the CC1110 and remove the need for CC Debugger.
Pete Schwamb
@ps2
Jul 17 2016 01:29
The breakout board does the gpio shifting.
The edison native gpios are 1.8v.
fjungman
@fjungman
Jul 17 2016 01:29
I found no reference that said Edison could interface 3.3V, and definitly the CC1110 couldn't handle 1.8V logic input.
Also, the other level shifting is for the Wixel.
Pete Schwamb
@ps2
Jul 17 2016 01:30
I researched that quite a bit, because I was really hoping to get away w/o them.
Chris Oattes
@cjo20
Jul 17 2016 01:34
ah, yeah, 1.89v max
Paul Martin
@Spazholio
Jul 17 2016 02:13
So, I guess the big question is - when will they be ready for shipping? Gotta get a Kickstarter/GoFundMe set up , or….? :)
Scott Leibrand
@scottleibrand
Jul 17 2016 03:07
@mhaeberli yeah, @sarahspins was right: you don't change system time. That is set at the factory and never adjusted. You just set the display time (which likely actually just sets an offset to system time internally)
Martin Haeberli
@mhaeberli
Jul 17 2016 03:37
@scottleibrand thx
Mitchell Slep
@moomoobloo
Jul 17 2016 03:53

My treatments are being uploaded to nightscout a few hours in the past. I'm guessing it's a timezone issue because in all the reports the treatments have the correct local time. Pump has the correct time, and nightscout and rpi have the correct time and timezone (America/Los Angeles). Here's my pumphsitory-zoned:

report "monitor/pumphistory-zoned.json"]
use = rezone
reporter = JSON
astimezone = False
date = ['timestamp', 'dateString', 'start_at', 'end_at', 'created_at']
adjust = missing
input = monitor/pumphistory.json
device = tz
timezone =

Anything look wrong there? Anything else I should check?

Scott Leibrand
@scottleibrand
Jul 17 2016 03:55
what do the actual .json file timestamps look like?
do they have -0700 at the end?
Pete Schwamb
@ps2
Jul 17 2016 03:56
Finally got a pebble. Been wanting one for a long time.. Well, more wanting my apple watch to have an always on screen, and frequently updating complications, but pebble is the only game in town for that right now. So I’m wondering which watchface people are using for NS + pump info & tempbasal info. Is there any face that displays all that?
Scott Leibrand
@scottleibrand
Jul 17 2016 03:56
(particularly the .json files you're uploading)
Mitchell Slep
@moomoobloo
Jul 17 2016 03:56
no, they look like this: "timestamp": "2016-07-16T19:52:00",
clock-zoned.json has the -0700 but not pumphistory-zoned
Scott Leibrand
@scottleibrand
Jul 17 2016 03:56
@moomoobloo ok, then you're not zoning them properly. is that a -zoned.json file that is missing the timezone?
k
Mitchell Slep
@moomoobloo
Jul 17 2016 03:57
yes, it's the zoned one
Scott Leibrand
@scottleibrand
Jul 17 2016 03:57
@ps2 you want Urchin probably
and maybe Cards for Pebble
Pete Schwamb
@ps2
Jul 17 2016 03:57
Thanks, I’ll look into those.
Mitchell Slep
@moomoobloo
Jul 17 2016 03:58
@scottleibrand anything look wrong with my pumphsitory-zoned config?
Scott Leibrand
@scottleibrand
Jul 17 2016 03:58
try removing the timezone = line
also, my date = line is slightly different syntax: date = timestamp dateString start_at end_at created_at
Paul Martin
@Spazholio
Jul 17 2016 04:18
For my monitor/pumphistory-zoned.json file, I have astimezone = True.
[report "monitor/pumphistory-zoned.json"]
use = rezone
reporter = JSON
astimezone = True
date = timestamp dateString start_at end_at created_at
adjust = missing
input = monitor/pumphistory.json
device = tz                                                                                                                                                                                                                                                                                                             
timezone = EDT
Unsure if I even need the timezone = EDT but it works so far, so...
Scott Leibrand
@scottleibrand
Jul 17 2016 04:20
@Spazholio do you know what add command is required to get that? I'm using https://github.com/scottleibrand/openaps-sh/blob/refresh-loops/setup.sh#L119
you actually don't want the timezone = EDT, but I'm not sure how to edit the add command to get rid of it
Paul Martin
@Spazholio
Jul 17 2016 04:20
openaps use tz rezone --timezone "EDT" --date timestamp --date dateString --date start_at --date end_at --date created_at --astimezone monitor/pumphistory.json
Scott Leibrand
@scottleibrand
Jul 17 2016 04:21
when DST ends you'll be off by an hour unless you remove it before then (or change it to EST at exactly the right time)
Paul Martin
@Spazholio
Jul 17 2016 04:21
I could always just edit the INI.
I suspect if you leave the --timezone “EDT” out of the add command, it wouldn’t put it in there?
Scott Leibrand
@scottleibrand
Jul 17 2016 04:21
unfortunately it's a default
Paul Martin
@Spazholio
Jul 17 2016 04:21
Once I finish my dev rig, I really should try all of these things.
Mitchell Slep
@moomoobloo
Jul 17 2016 04:21
fixing date= did the trick!
Paul Martin
@Spazholio
Jul 17 2016 04:21
Is it? Huh.
Mitchell Slep
@moomoobloo
Jul 17 2016 04:21
the syntax I used came from the walkthrough
Scott Leibrand
@scottleibrand
Jul 17 2016 04:21
unless that's been changed in a more recent version than I use
the walkthrough shouldn't have "make your ini look like this" stuff
Paul Martin
@Spazholio
Jul 17 2016 04:22
Maybe @bewest can shed some light on that one?
That’s very true. I keep forgetting that aspect of things.
Scott Leibrand
@scottleibrand
Jul 17 2016 04:22
if you can PR an edit to replace that with the appropriate openaps report add command that would be good
I just do openaps report add monitor/pumphistory-zoned.json JSON tz rezone monitor/pumphistory.json
only problem with that is the extra timezone = PDT at the end that breaks things when we switch timezones (or DST)
Paul Martin
@Spazholio
Jul 17 2016 04:23
How does it know what TZ to use if it’s not in the file? System TZ?
Scott Leibrand
@scottleibrand
Jul 17 2016 04:24
yeah
Paul Martin
@Spazholio
Jul 17 2016 04:24
Wait…I think my system uses EDT. So isn’t that equivalent? Or is some magic being done whenever DST ends to make sure things are in sync?
Scott Leibrand
@scottleibrand
Jul 17 2016 04:24
when DST ends date will change to EST
if you don't have timezone = in your ini it will automatically update the timezone it adds to reflect that
Paul Martin
@Spazholio
Jul 17 2016 04:26
What happens if you just put —timezone in the command? With no value after it?
scottleibrand @scottleibrand tries
Paul Martin
@Spazholio
Jul 17 2016 04:26
Or —timezone “”
Scott Leibrand
@scottleibrand
Jul 17 2016 04:27
that does timezone = None
not sure if that's good
seems to work
"timestamp": "2016-07-16T21:25:06-07:00",
Paul Martin
@Spazholio
Jul 17 2016 04:29
Still though - the default should probably result in an entry with no value in it if you don’t explicitly set the flag.
Scott Leibrand
@scottleibrand
Jul 17 2016 04:30
agreed. want to confirm whether that behavior is still the same in latest dev and file an issue if so?
Paul Martin
@Spazholio
Jul 17 2016 04:31
Is this a feature of openaps or oref0? Probably openaps, right?
Scott Leibrand
@scottleibrand
Jul 17 2016 04:31
openaps
if it was oref0 I'd just fix it :)
Paul Martin
@Spazholio
Jul 17 2016 04:31
Is openaps not “yours” as well?
Scott Leibrand
@scottleibrand
Jul 17 2016 04:31
I don't know the python code
Paul Martin
@Spazholio
Jul 17 2016 04:31
I always thought it was. Learn something new every day. =)
Scott Leibrand
@scottleibrand
Jul 17 2016 04:32
ben has been primary maintainer there
and he wrote almost all of it
Paul Martin
@Spazholio
Jul 17 2016 04:36
Same with oref0?
Scott Leibrand
@scottleibrand
Jul 17 2016 04:37
Ben helped bootstrap that, but I've done almost all the coding there, up until recently when a lot of contributions have been coming from others like @sulkaharo, @cjo20, and @jasoncalabrese
oref0 is all javascript
Paul Martin
@Spazholio
Jul 17 2016 04:38
Oh, I know. I’ve looked at both codebases. Which is why I was curious why one was in Python and the other JS when I thought one person was sorta responsible for both. =)
Scott Leibrand
@scottleibrand
Jul 17 2016 04:39
@bewest has also continued to make a lot of contributions to oref0 since the beginning, mostly cleaning up my stuff and making things more sane. :)
the main reason for using js in oref0 was a lot of the code we started with was stuff I'd contributed to NS cgm-remote-monitor
Paul Martin
@Spazholio
Jul 17 2016 04:40
It’s been a while since I’ve done pure JS coding. My last foray into it was as a web dev using jQuery for a year or so.
Scott Leibrand
@scottleibrand
Jul 17 2016 04:41
Ben helped me bootstrap oref0 while sitting on the 5th floor "log cabin" area at Twitter HQ one evening. I remember that well. :)
Paul Martin
@Spazholio
Jul 17 2016 04:41
Wait, which one of you worked at Twitter?
Dana Lewis
@danamlewis
Jul 17 2016 04:41
Heh jQuery. Having flashbacks.
@scottleibrand worked at Twitter for a year
Based in Seattle, then flew down once a month or so
Paul Martin
@Spazholio
Jul 17 2016 04:43
The guy that taught me jQuery used to study and read the DOM specs for fun. Used to point out where jQuery would be doing things wrong, and he was almost always right. It was a little unnerving.
Dana Lewis
@danamlewis
Jul 17 2016 04:44
Lol
Scott Leibrand
@scottleibrand
Jul 17 2016 04:46
I am fortunate never to have used jQuery. :)
Dana Lewis
@danamlewis
Jul 17 2016 04:47
It wasn't that bad. I tried to teach a doc how to use it at the hospital (because he wanted rotating images), so he could swap the images out at his leisure, but he didn't like it so much.
Paul Martin
@Spazholio
Jul 17 2016 04:47
Ooooh, dare I ask why? I quite liked it, provided I could find out what was happening under the hood. It was definitely useful.
Dana Lewis
@danamlewis
Jul 17 2016 04:47
:+1:
Pete Schwamb
@ps2
Jul 17 2016 05:08
urchin is pretty awesome!
Thanks @mddub!
Dana Lewis
@danamlewis
Jul 17 2016 05:10
:+1: that! ^
jim-andrews
@jim-andrews
Jul 17 2016 05:19
@jasoncalabrese Thanks for putting together the scratch pad. I had planned on spending the weekend blissfully working on my CGM and the Edison project right up until my central air bit the dust. With 100 degree temps, that's game over for the computer projects. Will be happy to review your docs on Monday when I can stand to be in my work shop. Anyone have a block of ice I can sit on?
berat
@beratkurar
Jul 17 2016 05:36
blob
Who knows what is wrong with my deployment in azure cloud:
Because my website shows me only white blank page.
berat
@beratkurar
Jul 17 2016 06:04
It was that I used a n API_SECRET value which contains less than 12 characters
Bill Stackpole
@billstackpole
Jul 17 2016 10:33
So edison out of space is in fact 100% usage of /home directory. .git is 1.3 gig right now (so @tazitoo you're right - git is eating up the space). What's not clear is how to purge potentially un-needed git artifacts - any ideas or thoughts? (Don't want to delete things in the .git directory that will force a full reinstall <smile>).
Bill Stackpole
@billstackpole
Jul 17 2016 10:41
@jasoncalabrese - what directories would you suggest be symlinked? (/home was 100% full - deleted a few installers I had there so now @ 99% but most of it is .git) Probably wouldnt make sense to send things to the bit bucket, given that the git history is intended to be available to evaluate issues that may have happened, but I've not had to interrogate (yet) my git archives. I did purchase a sparkfun daughter card that holds a microSD. I intend to add, format, and mount a card, and then might plan to symlink my .git directory there - any issues you can think of with that approach? Is there an easy or clean way to purge or cleanup the git objects > 1 month old or similar? This looks like a data aggregation issue similar to what might happen if we didn't use logrotate or similar. Some method to auto-cleanup might be in order. Thanks and Happy Sunday!
Bill Stackpole
@billstackpole
Jul 17 2016 10:54
also just changed the logrotate script to execute daily / 5M / bitbucket as you suggested to someone else (back in feb?). Rebooting now after manually running invoke-rc.d rsyslog rotate > /dev/null - looks like my ministrations only freed up 15meg of space (boo hiss)
logs only taking about 48M.
edison@billopenapse1:~$ sudo du -h /var/log
[sudo] password for edison: 
12K    /var/log/fsck
4.0K    /var/log/unattended-upgrades
4.0K    /var/log/ntpstats
4.0K    /var/log/news
4.0K    /var/log/ConsoleKit
32K    /var/log/apt
4.0K    /var/log/watchdog
48M    /var/log
edison@billopenapse1:~$
Bill Stackpole
@billstackpole
Jul 17 2016 11:00
output of sudo du -h /home/edison | more (snipping out all the extra git directories so all fits in one paste ...)
16K    /home/edison/billopenaps/nightscout
32K    /home/edison/billopenaps/settings
20K    /home/edison/billopenaps/upload
76K    /home/edison/billopenaps/pump-raw
28K    /home/edison/billopenaps/minimed-connect-to-nightscout-master/test
100K    /home/edison/billopenaps/minimed-connect-to-nightscout-master
68K    /home/edison/billopenaps/monitor
12K    /home/edison/billopenaps/enact
44K    /home/edison/billopenaps/.git/hooks
4.0K    /home/edison/billopenaps/.git/refs/heads
4.0K    /home/edison/billopenaps/.git/refs/tags
12K    /home/edison/billopenaps/.git/refs
124K    /home/edison/billopenaps/.git/objects/ab
...<snip>...
176K    /home/edison/billopenaps/.git/objects/b4
1.3G    /home/edison/billopenaps/.git/objects
576K    /home/edison/billopenaps/.git/logs/refs/heads
580K    /home/edison/billopenaps/.git/logs/refs
1.2M    /home/edison/billopenaps/.git/logs
4.0K    /home/edison/billopenaps/.git/branches
8.0K    /home/edison/billopenaps/.git/info
1.3G    /home/edison/billopenaps/.git
1.3G    /home/edison/billopenaps
12K    /home/edison/.ssh
24K    /home/edison/.npm/registry.npmjs.org/oref0
28K    /home/edison/.npm/registry.npmjs.org
32K    /home/edison/.npm
24K    /home/edison/mmeowlink-source/mmeowlink.egg-info
28K    /home/edison/mmeowlink-source
1.3G    /home/edison/
running edison@billopenapse1:~$ sudo du -h /home/edison/ | wc -l yields 285 lines (minus the 30 or so above means approx 250 .git/objects)
and that's where all the space is used...
Please (at)mention me in replies - will pop back in later in the day - tnx!
Chris Oattes
@cjo20
Jul 17 2016 12:19
I think the issue with "BG data too old" is something firmware-related in the pump. I've only really started seeing the error since switching from a 3.1 to 2.6 554
berat
@beratkurar
Jul 17 2016 13:04
Can nightscout show me an adviced value for basal rate?
Chris Oattes
@cjo20
Jul 17 2016 13:17
Yes. It's part of the device status upload
You need to have the openaps pill enabled
berat
@beratkurar
Jul 17 2016 13:18
Do I enable openaps pill, in the way I enabled iob, cob, rawbg and etc. ?
Julie
@julieraines
Jul 17 2016 13:30
I'm having some issues with upgrading NS to the DEV branch. I've followed the instructions and either end up in 0.8.5 off the MASTER (with no OPENAPS PILL) or 0.5.1 when I try to deploy to the DEV off HEROKU - any suggestions on this?
berat
@beratkurar
Jul 17 2016 13:34
Unfortunately I even don't understand what you ask about.
Chris Oattes
@cjo20
Jul 17 2016 13:44
@beratkurar yes
Julie
@julieraines
Jul 17 2016 13:44
that's ok - I don't understand half of what I ask about - LOL - so in order to use Nightscout to visualize/see what is going on in the artificial pancreas, I need a beta version of it installed which I've tried to do unsuccessfully a few times over the last few days. If you're seeing the OpenAPS pill where there is IOB and COB in your Nightscout, then you were able to do this and I haven't been. And if you look below the pill and see what version you have (0.8.5 or lower is not what it should be it should be 0.9 something or other).
berat
@beratkurar
Jul 17 2016 13:45
:+1: @cjo20 I am working on it
Chris Oattes
@cjo20
Jul 17 2016 13:46
@julieraines which branch are you trying to deploy? it should be cgm-remote-monitor/dev
Julie
@julieraines
Jul 17 2016 13:47
yes that's what I've tried deploying with no success - end up in 0.5.1 when I do that
Chris Oattes
@cjo20
Jul 17 2016 13:47
I've not used heroku before so I'm not sure how to choose which version you deploy
Julie
@julieraines
Jul 17 2016 13:48
it's a drop-down menu from a manual deploy
i've done it several times just not had it actually work
no error message
just doesn't give me the dev branch
Chris Oattes
@cjo20
Jul 17 2016 13:49
Are you trying to deploy from your own repository or the nightscout one?
Julie
@julieraines
Jul 17 2016 13:51
My own. This is what I've done: To update your version and create a PR to your dev branch, use http://nightscout.github.io/pages/test-beta/?branch=dev and look for the “I’m ready” button. Once you have completed these steps, log on to Azure or Heroku and disconnect the deployment source. Thereafter choose your cgm-remote-monitor github repository as source again. You should take the dev branch of this repository.
and i've tried without using the links just going into github from my repository
Chris Oattes
@cjo20
Jul 17 2016 13:54
git clone -b dev https://github.com/julieraines/cgm-remote-monitor.git
cd cgm-remote-monitor
git remote add upstream https://github.com/nightscout/cgm-remote-monitor
git fetch upstream
git merge upstream/dev
git push origin dev
Try running that
Chris Oattes
@cjo20
Jul 17 2016 14:03
Then click the "Deploy to Heroku" here https://github.com/julieraines/cgm-remote-monitor/tree/dev
Julie
@julieraines
Jul 17 2016 14:03
@cjo20 OMG - that was SO cool. And I did deploy right away and it WORKS!!! So do I PR this for those like me who can't get this?!? And I had to do it twice because I put my password where the username goes - LOL - not enough coffee this morning : ^ )
you are awesome!!
Chris Oattes
@cjo20
Jul 17 2016 14:10
Not sure where you'd put it, but if you feel any documentation needs improving, feel free to make a PR
Ben West
@bewest
Jul 17 2016 16:56
@Spazholio just leave --timezone out altogether... the default is None when it's left unspecified, which ensures that the local system timezone is used
that code is in openaps-contrib
Paul Martin
@Spazholio
Jul 17 2016 16:56
Is it? I thought @scottleibrand was seeing differently. I have no idea - I didn’t try it myself.
Ben West
@bewest
Jul 17 2016 16:56
is it hard to try?
Paul Martin
@Spazholio
Jul 17 2016 16:57
Nope, not at all. But last night - it was late, and I was just having a conversation with him while he was trying it so didn’t bother to do it myself. =)
Ben West
@bewest
Jul 17 2016 16:57
https://github.com/openaps/openaps-contrib/blob/master/openapscontrib/timezones/__init__.py#L53 by default it uses system time if you leave it out altogether
Paul Martin
@Spazholio
Jul 17 2016 16:59
Yep, looks like it does what we were talking about as a default when not specified. Good call.
Jason Calabrese
@jasoncalabrese
Jul 17 2016 17:01
@billstackpole I think your main issue is "1.3G /home/edison/billopenaps/.git", I think your git repo needs a git gc, git gc --auto was added to openaps a while ago and solved that issue for me. You might be running an other version. Since the repo is so big now theres a chance git gc will fail and you'll need to use oref0-fix-git-corruption.sh to fix/reset it.
Scott Leibrand
@scottleibrand
Jul 17 2016 17:05
@bewest looks like I was stuck on a nonexistent dev branch of openaps-contrib, so wasn't updating. does timezone = do the same thing as if the timezone line is not applied, or as timezone = None ?
ugh, apparently not. now it's giving me "timestamp": "2016-07-17T08:37:32",
it's not due to the timestamp = line, though, it's because the latest openaps-contrib does date = ['timestamp', 'dateString', 'start_at', 'end_at', 'created_at'] instead of date = timestamp dateString start_at end_at created_at
I changed that back to the old format and it works again
Scott Leibrand
@scottleibrand
Jul 17 2016 17:14
just opened openaps/openaps-contrib#6 to track this
Ben West
@bewest
Jul 17 2016 17:20
timezone =is what you want
right
please use pip to install the package appropriately
that will go a long way
pip install -U openaps-contrib
Jason Calabrese
@jasoncalabrese
Jul 17 2016 17:34
I don't see why I'd ever want to install via the packages instead of source
Scott Leibrand
@scottleibrand
Jul 17 2016 18:01
@bewest I have the new version of openaps-contrib installed properly. just did it via that method, and got the same date = ['timestamp', 'dateString', 'start_at', 'end_at', 'created_at'] regression causing:
edison@edison-ti ~/openaps-dev $ grep timestamp settings/pumphistory-24h-zoned.json | head -1
    "timestamp": "2016-07-17T08:37:32",
Paul Martin
@Spazholio
Jul 17 2016 18:01
@jasoncalabrese Does pulling in via the packages ensure minimum versions of dependencies? That could be a reason?
Scott Leibrand
@scottleibrand
Jul 17 2016 18:02
package would've been more convenient for not having to pip install recurrent, if I could just remember the dumb syntax for all the different package install methods. :-)
Ben West
@bewest
Jul 17 2016 18:03
it's usually the git url-ish
Scott Leibrand
@scottleibrand
Jul 17 2016 18:03
yeah, url-ish is usually the thing that makes no sense in my head and won't store in memory :)
Ben West
@bewest
Jul 17 2016 18:04
it's the same as the github url itself
except it ends with .git and starts with git+
it installs directly from the git remote without causing modal confusion with newbies
Scott Leibrand
@scottleibrand
Jul 17 2016 18:05
unless there's a way to generate that, I won't remember the magic rules. guess we need to write it down somewhere
Ben West
@bewest
Jul 17 2016 18:05
it is written down :-)
and I keep repeating it
Scott Leibrand
@scottleibrand
Jul 17 2016 18:05
heh, where can I find it easily?
maybe my brain can store a pointer. :)
Paul Martin
@Spazholio
Jul 17 2016 18:06
Did you know you can replace use with three words to add a report?
Wait…wrong thing got hammered into memory.
Scott Leibrand
@scottleibrand
Jul 17 2016 18:06
anyway, that's my problem to solve. the more interesting question for @bewest is whether you have any idea why the date = array broke things.
openaps/openaps-contrib#6
Ben West
@bewest
Jul 17 2016 18:13
openaps/docs#133
I don't think it's broken
Ben West
@bewest
Jul 17 2016 18:18
I know exactly why stuff like remainder = [] and similar breaks things in the INI
I've been working furious on releasing many patches to fix all those issues
AFAICT they are all fixed :-)
it's generally solved by using to_ini and from_ini to prevent python from serializing the repr version (the string literals)
ParkerBenda
@ParkerBenda
Jul 17 2016 19:06
My doctor (researcher and t1d himself) was so amazed when I told him about open APS he decided to build one!
That's pretty much the ultimate compliment of yalls project;)
Martin Haeberli
@mhaeberli
Jul 17 2016 19:09
:=1:
:+1:
Scott Leibrand
@scottleibrand
Jul 17 2016 19:15
:-D
if anyone here has time to work on some documentation stuff, we have four open issues tagged with Help Wanted: https://github.com/openaps/docs/labels/help%20wanted
Scott Leibrand
@scottleibrand
Jul 17 2016 19:23
if any of them are unclear as to what needs done, feel free to ask questions, either here or on the issue itself
Julie
@julieraines
Jul 17 2016 19:57
I am at the point where I don't quite know what to do next. I think I am having trouble with the concepts of reports, alias, ref0, and JSON. I understand what they are, I've created a report, invoked a report, created a JSON file, and that's about it. I'm not sure what reports I need to create because when I list what I have, there's a lot there. And I'd like to get to the stage where I can just run it without it being attached to my daughter to see what it does (I have water in the pump right now because she wears a TSlim now and will wear this at night at first once I get it working). I don't really even know what question to ask - I get how to make the pump suspend and resume and how to get glucose readings off the CGM manually and I've played around with this. I think I need to understand how to customize the oref0 so that my settings/parameters are in there?
Martin Haeberli
@mhaeberli
Jul 17 2016 20:00
@julieraines I’m with you, including wanting to run without the patient being near or attached, just to learn more about how the system works while ramping up. There was a post yesterday about using fake devices, but I haven’t invested in trying to understand that part yet ...
Jason Calabrese
@jasoncalabrese
Jul 17 2016 20:02
when I was first testing I ran with the pump on my desk with no cartridge, and pulled BG data from NS, it didn't know about boluses, but still would still set temps and upload to a different NS site
for the list of reports the templates are the way to go
Scott Leibrand
@scottleibrand
Jul 17 2016 20:03
There is stuff in the walkthrough for pulling all the settings and reports off the pump and using those to make a suggested.json.
Jason Calabrese
@jasoncalabrese
Jul 17 2016 20:04
but thats not using oref0 template mint reports medtronic-pump | openaps import etc right?
Julie
@julieraines
Jul 17 2016 20:05
Ahhh. @jasoncalabrese I guess since I've been having NS trouble, I didn't consider that. @scottleibrand I've copied all the reports over from the walkthrough. And I've read the suggested.json a few times and since I don't really work in this environment it is not fully clear how I make the suggested.json
@jasoncalabrese I did use that template
I feel a bit like I'm in another country ordering off the menu and I don't exactly speak the language - I know enough to have food and drink brought to the table but God only knows what will show up (LOL)
Julie
@julieraines
Jul 17 2016 20:36
I have NS version 0.9.0-beta3 but not pulling the latest numbers - we use Share G5 - any ideas?
Martin Haeberli
@mhaeberli
Jul 17 2016 20:48
@julieraines when was this last working for you? I had some trouble setting it up, had to remember to add configuration string, etc. and restart the NightScout server. See, for example, http://www.nightscout.info/wiki/welcome/nightscout-with-ios-and-dexcom-share and http://www.nightscout.info/wiki/welcome/nightscout-with-ios-and-dexcom-share/nightscout-bridge-for-dexcom-g5
sarahspins
@sarahspins
Jul 17 2016 21:09
So since I've upgraded my NS site to dev from master and I'm actually using the pillboxes and tooltips that come up when you tap on them, I'm having a big problem on mobile where I can't get the tooltip to go away
I don't know if it just wasn't a problem before or I simply never noticed it
David Cintron
@loudestnoise
Jul 17 2016 21:11
My trick for that is tap on one of the times 3HR, 6HR etc. and it makes whatever tooltip you have up go away
on mobile at least
Julie
@julieraines
Jul 17 2016 21:13
@mhaeberli THANKS! - needed the username password in HEROKU and bridge in ENABLE
works like butta now
sarahspins
@sarahspins
Jul 17 2016 21:14
@loudestnoise sometimes that works and other times it takes multiple clicks
or taps, whatever :P
Martin Haeberli
@mhaeberli
Jul 17 2016 21:15
:+1:
Chris Oattes
@cjo20
Jul 17 2016 21:37
Hmm, going on holiday. Currently got a sensor that's pretty much spot-on, but it's 8 days old. Don't know whether to change it or not
Julie
@julieraines
Jul 17 2016 21:38
what's your average lifetime on a sensor? we're at about 10 days
sarahspins
@sarahspins
Jul 17 2016 21:49
I am with @julieraines - what is your average? That should lead you in the right direction. I commonly get 20+ from mine.. so I'd leave it on, but pack a spare (or two, just to be safe)
Julie
@julieraines
Jul 17 2016 21:51
My 12 year old is the diabetic and she's tiny - and doesn't hydrate much which has an impact AND she is constantly exercising (gymnastics, dance) - we've gotten 20+ days before several times. Swimming and exercise are factors for sure . . . more water, less adhesive life and we don't use a lot of extras to keep that sucker on.
sarahspins
@sarahspins
Jul 17 2016 21:54
@julieraines all of that makes a difference :)
so as I'm progressing along with my loop.... is "COB: undefined" in my openAPS pill on NS something I should be worried about?
I do have carbs set on my NS profile, do I need to configure something else somewhere?
Chris Oattes
@cjo20
Jul 17 2016 22:00
@sarahspins it means it doesn't have meal.json from openaps
My average is 8-10 days; I'm on medtronic CGM
sarahspins
@sarahspins
Jul 17 2016 22:00
@cjo20 okay, but how is that report generated?
Julie
@julieraines
Jul 17 2016 22:01
ahh - medtronic. we're dexcom
i guess changing the sensor usually is an issue with how long it takes to "get in the groove" with accuracy and when i need the accuracy to be spot on - like everything it's timing
@sarahspins did you get all the NS configured and setup for the loop? i'm working through that right now and it's a little slice of heaven (not)
sarahspins
@sarahspins
Jul 17 2016 22:04
@julieraines yes with the exception that it's not uploading missing CGM entries (my NS site is a bridge, so sometimes my phone misses readings that the receiver picks up) and I can't figure out why
Julie
@julieraines
Jul 17 2016 22:06
my NS is a bridge too and i believe i've always had that issue
sarahspins
@sarahspins
Jul 17 2016 22:22
what, the missing data or openaps not uploading what it might grab that isn't there?
Julie
@julieraines
Jul 17 2016 22:23
IDK because share definitely fills in the blanks once it's reconnected
Chris Oattes
@cjo20
Jul 17 2016 22:25
@sarahspins using the "meal" command for oref0
sarahspins
@sarahspins
Jul 17 2016 22:29
@cjo20 is there a reference for that somewhere?
the only mention I've found in the docs has no information on how to set it up, it basically looks like it assumes that was already there?
and this doesn't really tell me what I need to know either
pi@openAPS:~/sarah $ oref0 meal -h
usage:  [ 'node', '/usr/local/bin/oref0-meal' ] <pumphistory.json> <profile.json> <clock.json> <glucose.json> <basalprofile.json> [carbhistory.json]
sarahspins
@sarahspins
Jul 17 2016 22:36
I mean, I'm trying, but I don't get what I'm missing.. do I need ALL the things?
pi@openAPS:~/sarah $ openaps report add monitor/meal.json text meal shell monitor/pump-history.json settings/profile.json monitor/clock.json
added meal://text/shell/monitor/meal.json
pi@openAPS:~/sarah $ openaps report invoke monitor/meal.json
meal://text/shell/monitor/meal.json
usage:  [ 'node', '/usr/local/bin/oref0-meal' ] <pumphistory.json> <profile.json> <clock.json> <glucose.json> <basalprofile.json> [carbhistory.json]
reporting monitor/meal.json
pi@openAPS:~/sarah $ cat monitor/meal.json
{ "carbs": 0, "reason": "Insufficient arguments" }
Chris Oattes
@cjo20
Jul 17 2016 22:40
you need all of the ones that aren't in []
so pumphistory, profile, clock, glucose and basalprofile
sarahspins
@sarahspins
Jul 17 2016 22:40
ok, so I did that and I have this now
pi@openAPS:~/sarah $ openaps report add monitor/meal.json text meal shell monitor/pump-history.json settings/profile.json monitor/clock.json monitor/glucose.json settings/basal_profile.json
added meal://text/shell/monitor/meal.json
pi@openAPS:~/sarah $ openaps report invoke monitor/meal.json
meal://text/shell/monitor/meal.json
14 0 Sun Jul 17 2016 16:05:45 GMT-0500 (CDT)
++-+--+++-+>>>-+--+--+++>>++-!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!+++-+-++--+-+++++------+-+--+++-+-+>++++-+++=+--=+===++==++-==----++----------+-------===++==-=--+-----+=-+-+---=--=-----++----------=----+-+!!!!!!!!!!!!!!!!!!!+--+-+++-+--+-+=+=+-+->-+-------------+++>>>---+--->>>+=++++-+=++= 52% of non-meal deviations negative (target 45%-50%)
Excess insulin sensitivity detected: Ratio limited from 0.5842857142857143 to NaN
ISF adjusted from 28 to NaN
NaN
reporting monitor/meal.json
pi@openAPS:~/sarah $ cat monitor/meal.json
{"carbs":14,"boluses":6.4,"mealCOB":null}
Scott Leibrand
@scottleibrand
Jul 17 2016 22:40
@johners97: you can set up OpenAPS to work offline without Nightscout, but you'll also want Nightscout set up so you can check what the loop has been up to when you do happen to have Internet.
Chris Oattes
@cjo20
Jul 17 2016 22:41
I suspect that the '!' means that it can't find glucose data for those periods
sarahspins
@sarahspins
Jul 17 2016 22:42
yes, I had a sensor warmup this morning :)
is "mealCOB":null a problem?
Scott Leibrand
@scottleibrand
Jul 17 2016 22:43
@sarahspins that most likely means you don't have meal assist set up yet
that is an optional feature you can worry about later
sarahspins
@sarahspins
Jul 17 2016 22:43
well I wanted the COB: undefined to go away :P
on the openAPS pill
Scott Leibrand
@scottleibrand
Jul 17 2016 22:44
yeah, those are the same thing
sarahspins
@sarahspins
Jul 17 2016 22:44
well phooey
John Dodds
@johners97
Jul 17 2016 22:44
@scottleibrand Oh ok! Thanks
Scott Leibrand
@scottleibrand
Jul 17 2016 22:45
@johners97 no problem, and welcome! :)
@sarahspins the NaN in the mea.json report output is a pointer to your problem. check the input files and see if they all look like valid data?
sarahspins
@sarahspins
Jul 17 2016 22:47
I can check them again.. am I looking for something specific?
Scott Leibrand
@scottleibrand
Jul 17 2016 22:49
not sure. usually NaN comes from data being completely the wrong format, or a divide by zero
feel free to paste one record from each file and we can look for anything obvious there
sarahspins
@sarahspins
Jul 17 2016 22:50
so this is probably it then
pi@openAPS:~/sarah $cat settings/basal_profile.json
[
  {
    "i": 0, 
    "start": "00:00:00", 
    "rate": 1.0, 
    "minutes": 0
  }
let me go track that one down and see if I can fix it :)
Scott Leibrand
@scottleibrand
Jul 17 2016 22:50
is your basal profile 1U/hr at midnight?
sarahspins
@sarahspins
Jul 17 2016 22:50
no
it might have had it set that way when I was mucking with the pump initially... I didn't set my own basal profile since it wasn't attached to me
Scott Leibrand
@scottleibrand
Jul 17 2016 22:51
k, try invoking a fresh version of that report from the pump.
and then make sure your loop does that automatically at some point
sarahspins
@sarahspins
Jul 17 2016 22:51
hey, it has data now :D
Scott Leibrand
@scottleibrand
Jul 17 2016 22:51
and re-generates settings/profile.json
John Dodds
@johners97
Jul 17 2016 22:51
I feel foreign to all this complex looking code...be prepared for questions in the next few weeks haha
Scott Leibrand
@scottleibrand
Jul 17 2016 22:51
:-)
@johners97 you won't feel out of place long: there are lots of newbies here. :)
sarahspins
@sarahspins
Jul 17 2016 22:52
nope, that didn't help it
@johners97 it will start to make sense soon.. I'm just copying terminal output :)
Julie
@julieraines
Jul 17 2016 22:52
I'm a newbie @johners97 and not at all understanding the code but I've made excellent progress
Scott Leibrand
@scottleibrand
Jul 17 2016 22:53
@sarahspins most likely culprit is the glucose.json I'd guess.
sarahspins
@sarahspins
Jul 17 2016 22:53
hmm, ok
Scott Leibrand
@scottleibrand
Jul 17 2016 22:53
but that's just a guess
sarahspins
@sarahspins
Jul 17 2016 22:54
so does that mean not enough data, or something else?
John Dodds
@johners97
Jul 17 2016 22:55
is this python?
Dana Lewis
@danamlewis
Jul 17 2016 22:57
@johners97 depends on what you are asking about. openaps toolkit is Python. Oref0 is JavaScript.
John Dodds
@johners97
Jul 17 2016 22:59
Oh right:) That makes sense, good luck everyone doing this :) @danamlewis
Julie
@julieraines
Jul 17 2016 23:00
I'm not sure what reports I need to create because when I list what I have, there's a lot there.
openaps report show
cgm://JSON/oref0_glucose/raw-cgm/glucose-raw.json
pump://JSON/read_bg_targets/raw-pump/bg-targets-raw.json
units://JSON/bg_targets/settings/bg-targets.json
pump://JSON/read_insulin_sensitivities/raw-pump/insulin-sensitivities-raw.json
units://JSON/insulin_sensitivities/settings/insulin-sensitivities.json
pump://JSON/read_clock/raw-pump/clock-raw.json
tz://JSON/clock/monitor/clock.json
pump://JSON/read_temp_basal/monitor/temp-basal-status.json
determine-basal://text/shell/oref0-predict/oref0.json
pump://JSON/iter_pump_hours/raw-pump/pump-history-raw.json
tz://JSON/rezone/monitor/pump-history.json
pump://JSON/model/model.json
pump://JSON/reservoir/monitor/reservoir.json
pump://JSON/read_status/monitor/status.json
pump://JSON/read_battery_status/monitor/battery.json
pump://JSON/set_temp_basal/oref0-enacted/enacted-temp-basal.json
oref0://JSON/shell/settings/settings.json
pump://JSON/read_settings/raw-pump/settings.json
oref0://JSON/shell/settings/selected-basal-profile.json
pump://JSON/read_selected_basal_profile/raw-pump/selected-basal-profile.json
pump://JSON/read_temp_basal/raw-pump/temp-basal-status.json
pump://JSON/reservoir/raw-pump/reservoir.json
pump://JSON/read_status/raw-pump/status.json
pump://JSON/read_battery_status/raw-pump/battery.json
pump://JSON/iter_pump_hours/last_four_pump_hours.json
tz://JSON/clock/monitor/clock-zoned.json
pump://JSON/read_bg_targets/settings/bg_targets_raw.json
pump://JSON/read_insulin_sensitivies/settings/insulin_sensitivities_raw.json
pump://JSON/read_basal_profile_std/settings/basal_profile.json
get-profile://text/shell/settings/profile.json
pump://JSON/iter_pump_hours/monitor/pumphistory.json
calculate-iob://JSON/shell/monitor/iob.json
pump://JSON/read_temp_basal/monitor/temp_basal.json
cgm://JSON/iter_glucose/monitor/glucose.json
determine-basal://text/shell/enact/suggested.json
cgm://JSON/iter_glucose_hours/last_four_cgm_hours.json
ns://JSON/shell/nightscout/uploaded.json
ns://JSON/shell/nightscout/recent-treatments.json
Scott Leibrand
@scottleibrand
Jul 17 2016 23:02
@julieraines to put some structure around it, you might want to work up to whatever report you're trying to do next, like @sarahspins did above to figure out what inputs the meal.json report needed
you'll then need to make sure you have those reports defined, and invoke them, to generate the files that the oref0 reports need as input in order to suggest a temp
looks like you already have a oref0-predict/oref0.json report defined, so you can try invoking that and see what it complains about :)
sarahspins
@sarahspins
Jul 17 2016 23:05
yes, looking for complaints is usually the first step :)
sometimes things aren't missing, the names just aren't consistent
so you have to look for that too
Scott Leibrand
@scottleibrand
Jul 17 2016 23:05
yeah
Julie
@julieraines
Jul 17 2016 23:11
Ahhh 🤓 Lovely
sarahspins
@sarahspins
Jul 17 2016 23:23
a perfect example of that.. I used settings/profile.json earlier when that wasn't what I was using in my loop, I needed oref0-monitor/profile.json
not sure I'm making progress though... I've got part of it now, but still not what I want :(
pi@openAPS:~/sarah $ openaps report invoke monitor/meal.json
meal://text/shell/monitor/meal.json
14 0 Sun Jul 17 2016 16:05:45 GMT-0500 (CDT)
>++-->+++++-+---++-->>>-+--+--+++>>++-!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!+++-+-++--+-+++=+------+-+--+++-+-+>++++-+++=+--=+===++==++-==----++----------+-------===++==-=--+-----+=-+-+---=--=-----++----------=----+-+!!!!!!!!!!!!!!!!!!!+--+-+++-+--+-+=+=+-+->-+-------------+++>>>---+--->>>+=++++-+=++=++- 50% of non-meal deviations negative (target 45%-50%)
Excess insulin sensitivity detected: Ratio limited from 0.6703296703296704 to 0.7
ISF adjusted from 28 to 40
NaN
reporting monitor/meal.json
pi@openAPS:~/sarah $ cat monitor/meal.json
{"carbs":14,"boluses":6.4,"mealCOB":null}
sarahspins
@sarahspins
Jul 17 2016 23:29
for the record, I'm determined to get this because if I don't it's going to haunt me :P
Chris Oattes
@cjo20
Jul 17 2016 23:47
what happens if you try to invoke it with openaps use?
Julie
@julieraines
Jul 17 2016 23:50
@sarahspins I'm with ya, babe. Gotta get this done for my little one. She wants sleepovers at friend's houses and there's nothing more important than THAT - ha
sarahspins
@sarahspins
Jul 17 2016 23:51
@cjo20 how?
Terri Lyman
@tatigger
Jul 17 2016 23:53
I don't know what terms to use to search this question, maybe you can help. I was trying to create an instance of open aps on my pi using the terminal on my mac laptop. Most of the time, when I do everything right, this is the prompt I have and everything works ok: pi@raspberrypi:~ $ However, some of the time, obviously when I have typed something wrong, I have just: [ When I type anything there, it just thinks for a minute and then repeats my typing on the next line. Where do I look to figure out what I did, and what that is, and how to go back to the correct prompt without shutting down the terminal app on my Mac?
Julie
@julieraines
Jul 17 2016 23:54
when I invoke oref0-predict/oref0.json then I get "AttributeError: 'Namespace' object has no attribute 'iob'" so does this mean I need to add and invoke an iob report
sarahspins
@sarahspins
Jul 17 2016 23:56
iob is a process
@tatigger can you copy/paste an example?
Terri Lyman
@tatigger
Jul 17 2016 23:57
@sarahspins pi@raspberrypi:~ $ cd faith
pi@raspberrypi:~/faith $ ls
openaps.ini
pi@raspberrypi:~/faith $ cat
cat
cat
My open aps is "faith" as in Faithfully giving me info I need
sarahspins
@sarahspins
Jul 17 2016 23:58
try ctrl^c to get out of that
in terminal
usually when something gets "stuck" that's the first thing to try :)
Terri Lyman
@tatigger
Jul 17 2016 23:59
Yes! Writing THAT one down. So how do I use cat correctly?