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

13th
Jul 2017
Marco
@CaptainBalou
Jul 13 2017 00:02
@scottleibrand So do you think I should do nothing for now? I don’t feel that anything goes wrong.
Scott Leibrand
@scottleibrand
Jul 13 2017 00:07
I just looked in the cgm-remote-monitor package.json, and all I see for versions is "node": "0.10.x", so based on what @sulkaharo said it sounds like you're OK
although not sure what it's saying is running 6.10.0
PieterGit
@PieterGit
Jul 13 2017 00:08
@live4sw and @scottleibrand : unfortautely 0.10 is also vulnerable, see https://gitter.im/nightscout/public?at=5966b936bc46472974e3af4d
Scott Leibrand
@scottleibrand
Jul 13 2017 00:09
ah, k. so sounds like someone needs to update the NS package.json to a newer version that's compatible
if someone wants to experiment with listing something newer and see if NS still runs with that, that might inform what gets PR'd into NS dev...
but we probably should move further discussion over to https://gitter.im/nightscout/public for now, until we get to the point of having a tested fix
PieterGit
@PieterGit
Jul 13 2017 00:31
perhaps, making upgrade oref0 node to a stable/secure version after releasing 0.5.0 and after a few weeks of testing releasing a 0.6.0 with updated npm/node is also a good idea for oref0
Scott Leibrand
@scottleibrand
Jul 13 2017 01:43

We tried upgrading npm (and node) recently. It broke stuff, so I'm reflashing that rig back to a version that works.

I get upgrading node if it's running NS on a publicly accessible web server. But what's the point if you're not running any external services?

Scott Leibrand
@scottleibrand
Jul 13 2017 03:44
@PieterGit I just checked that rig, and it has node v6.11.0 and npm 5.0.3. After re-flashing it, I might try upgrading just node and leaving npm unchanged, as I think my only issues were with npm.
katie disimone
@Kdisimone
Jul 13 2017 04:45
someone else posted a day ago that they’d been running 6.11.1 stable for NS. So, I’m going with that.
Cameron Chunn
@applehat
Jul 13 2017 04:51
Fwiw:
root@openaps:~# node -v
v8.0.0
root@openaps:~# npm -v
5.0.3
My main rig
No node or npm related issues here.
Scott Leibrand
@scottleibrand
Jul 13 2017 04:58
looks like the main problem with npm 5.0.3 is that it's not very console-friendly: the long lines with lots of ansi don't get displayed correctly, so you can't tell if it's working if you run everything via openaps-bootstrap.sh
I wonder if it has a script-friendly mode we can activate instead
Cameron Chunn
@applehat
Jul 13 2017 05:00
Ahh yeah, tons of ansi stuff.
There is --no-color but that probably doesn't strip all the screen clearing.
Scott Leibrand
@scottleibrand
Jul 13 2017 05:02
looks like --no-spin doesn't work. trying | cat
nope, | cat doesn't work either
npm run global-install --no-progress doesn't work either, hmm
looks like maybe I have to redirect stderr too: npm run global-install 2>&1 | cat
Cameron Chunn
@applehat
Jul 13 2017 05:22
--progress=false and --no-color gave me a plain text output.
Lol, there's a config called.ham-it-up not mentioned in their docs.
Scott Leibrand
@scottleibrand
Jul 13 2017 05:31
@applehat I'm using npm 3, not 5, so I'll stick with 2>&1 | cat in the script and try --no-color when I'm doing it interactively. thx
Dirk Gastaldo
@dirkgastaldo
Jul 13 2017 05:36
Evening all. Working on a second rig and am testing out BT tethering. I drove out of range of WiFi. It took about 20 minutes but rig eventually got a 172 internal IP from my phone. But it never showed the public IP like my other rig did. It seemed to be stuck in this endless loop (network filter of log).
image.png
Scott Leibrand
@scottleibrand
Jul 13 2017 05:37
looks like you're still on oref0 master w/o the IPv6 vs. IPv4 fix to oref0-online
Zakm0n
@Zakm0n
Jul 13 2017 05:38
Ok, just made the switch to decom from enlite, and I've got xdrip+ running. showing BG, and downloading from nightscout, but for some reason, my xdrip readings are not showing up in nightscout.
Dirk Gastaldo
@dirkgastaldo
Jul 13 2017 05:39
Thanks, @scottleibrand Scott
Dirk Gastaldo
@dirkgastaldo
Jul 13 2017 05:46
@scottleibrand Scott, I'm looking for that in the docs. I searched on a number of the terms you mentioned, but I'm not finding it. What should I search for to find "the IPv6 vs. IPv4 fix to oref0-online"?
Scott Leibrand
@scottleibrand
Jul 13 2017 05:47
you just need to upgrade to oref0 dev: it's not specifically documented other than that
PieterGit
@PieterGit
Jul 13 2017 05:56
@scottleibrand : the v0.10 and v0.12 are not supported and end of life. So even if you're not running services, it should be a good practice to use supported versions of nodejs and npm.
Scott Leibrand
@scottleibrand
Jul 13 2017 05:59
The following packages have unmet dependencies:
nodejs : Conflicts: npm
E: Unable to correct problems, you have held broken packages.
that's from doing this on a freshly flashed rig:
curl -sL https://deb.nodesource.com/setup_6.x | sudo -E bash -
sudo apt-get install -y git python python-dev python-software-properties python-numpy python-pip nodejs npm
unless you have a better idea, I'll try taking npm out of that list
I really wish package maintainers for things like node would update their packages in the operating system repos rather than forcing you to add theirs
a default OS install and apt-get upgrade should always get the latest patch release of the current stable release of everything important IMO
</rant>
Matthias
@ceben80
Jul 13 2017 06:07
@tepidjuice @tim2000s @colinlennon I am out for today, excursion on holiday :star2:
Cameron Chunn
@applehat
Jul 13 2017 06:38
@scottleibrand on Ubuntu (at least) they don't update to the latest stable because of support reasons.
I agree tho.
Scott Leibrand
@scottleibrand
Jul 13 2017 06:42
So looks like removing npm from that list did the trick.
After we release oref0 0.5.0 and the dev docs, we can promote the nodejs-6 version of the bootstrap/install scripts to the dev docs branch.
Or maybe we should move all these install scripts over to the oref0 repo
katie disimone
@Kdisimone
Jul 13 2017 06:44
should’ve experimented with this before…but what size of usb battery will roughly power the rig offline (receiver plugged in) for the day. No wifi around. No bt tether. 6000 mah should be more than enough yes?
Lipo still there too, of course
Cameron Chunn
@applehat
Jul 13 2017 06:45
Also fwiw, I removed npm and node using apt-get, installed npm and node using their install script, installed N with NPM (a node version manager) and use that to swap versions if needed. Let's you run multiple nodes if needed. Kinda nice.
@Kdisimone easily, I'd say.
katie disimone
@Kdisimone
Jul 13 2017 06:45
great. thanks!
Cameron Chunn
@applehat
Jul 13 2017 06:46
I could be wrong tho.
A 6000mAh should easily power the rig for over 24 hours - no clue how much the receiver draws
(assuming we are talking Edison)
katie disimone
@Kdisimone
Jul 13 2017 06:49
yes
Cameron Chunn
@applehat
Jul 13 2017 06:50
When we say "for the day", if we are talking a standard 16 hour awake time - I'd say without question it will power it. If we are talking a 24 hour period, I'm not 100% sure :)
stefanakerblom
@stefanakerblom
Jul 13 2017 06:53
@applehat In my car I have a lithium jump starter pack with 68800 mAh!!!. Apprx. size 20102 cm, 500 grams. It also has 4 USB ports. I bought it on ebay from Malaysia for 40 $ including shipping. Works very well. If there is interest I can take a pic and post it here?
Scott Leibrand
@scottleibrand
Jul 13 2017 06:56
@applehat that's great for an actively managed environment. I'm trying to update node on everyone's mostly unmanaged new rigs. I think I got it working though.
It comes in a nice plastic case with all the cables. 955 grams is total weight.
Luis Betancourt
@betluis
Jul 13 2017 07:05
For Thise using pushover wouldn't you like the message to start with the host name of the rig that sends it?
When using various rigs it may be helpful
stefanakerblom
@stefanakerblom
Jul 13 2017 08:01
What does this exactly mean (from pump log)? Edison battery at 63% is charged (>= 98%) or likely charging (60-65%).
tepidjuice
@tepidjuice
Jul 13 2017 08:08
@tim2000s @colinlennon I'm about for the next couple of hours if someone's free to help.
Tim Street
@tim2000s
Jul 13 2017 09:44
Hi @tepidjuice
tepidjuice
@tepidjuice
Jul 13 2017 09:50
Hi @tim2000s
Miiko
@tw_miiko_twitter
Jul 13 2017 09:52
@scottleibrand hi, thank you. I have no running autotune. Max-IOB changes not from the default Running in midnight. Also 2 in Preferences.json and 3 in Autotune.
@scottleibrand And I test dev broke without SMB, looping runs, but Temp basal in NS Plug-in does not show up, basal changes not in NS, why?
Tim Street
@tim2000s
Jul 13 2017 10:07
I'm not sure what else to go through on your xdripaps setup. It sounds as though xDrip isn't communicating properly to the rig, so it might be worth a reflashed and clean reinstall, if you are sure there are no typos.
tepidjuice
@tepidjuice
Jul 13 2017 10:13
Yeah I'm thinking it's a problem with flask.
Luis Betancourt
@betluis
Jul 13 2017 10:24
please, anyone a suggestion to fix this? the loop is failing continuously here: Old pumphistory-24h refreshValueError: need more than 0 values to unpack
Unsuccessful supermicrobolus pump-loop at Thu Jul 13 05:22:27 CDT 2017
Luis Betancourt
@betluis
Jul 13 2017 10:36
I rebooted the edison and removed - inserted the battery from the pump
keep having the same issue
image.png
BG going up, Daughter sleeping... not very nice when loop does not work...
Luis Betancourt
@betluis
Jul 13 2017 10:45
image.png
tepidjuice
@tepidjuice
Jul 13 2017 11:38
@tim2000s though I'm stuck with 'captive' networks at the moment so I don't know how I can easily do a reflash?
Is there a way to get everything as close to stock as possible but keeping the bluetooth tether active?
tepidjuice
@tepidjuice
Jul 13 2017 12:18
when running git clone -b dev git://github.com/openaps/oref0.git I'm getting: Cloning into 'oref0'... error: cannot expand target domain name in SRV RR fatal: Unable to look up github.com
Oh... I was behind a proxy so I've accessed the url directly.
Miiko
@tw_miiko_twitter
Jul 13 2017 12:54
Looping status cannot in NS upload, why???
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 12:56
what do your logs say?
Miiko
@tw_miiko_twitter
Jul 13 2017 12:56
"2017-07-13T14:48:48" from clock.json, but it is 14:56
"2017-07-13T14:56:41+02:00" from clock-zoned.json
Tom Boudreau
@TomBoudr_twitter
Jul 13 2017 12:59
Anyone else get a heroku notice about their node.js?
tepidjuice
@tepidjuice
Jul 13 2017 13:06
I did a full reinstall of openaps and still have the same error message! This getting very confusing
Miiko
@tw_miiko_twitter
Jul 13 2017 13:15
what ist the second pass phrase for enabling oref1 features?
did not fond it in DOC
cameronrenwick
@cameronrenwick
Jul 13 2017 13:24
@TomBoudr_twitter yes a number of ppl did. There was some discussion last night. Today I rec'd an update explaining how to update my 2 apps. Kinda hoping there's going to be some more dialogue on this but I haven't spent the needed time to dig into it yet...
Tom Boudreau
@TomBoudr_twitter
Jul 13 2017 13:33
@tw_miiko_twitter in your openaps directory do a: nano preferences.json and you will see all the flags to edit
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 13:35
@TomBoudr_twitter The Node.js team has announced that a high severity remote Denial of Service (DoS) Constant Hashtable Seeds vulnerability in Node.js versions 4.x through 8.x has been patched in the following versions:
Tom Boudreau
@TomBoudr_twitter
Jul 13 2017 13:35
@cameronrenwick After I asked I found some earlier discussion over in another channel, seems like
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 13:35
I got an email about a vunerability in node.js
Tom Boudreau
@TomBoudr_twitter
Jul 13 2017 13:35
@diabetic_dude_twitter Thats the one... mostly interested in how to upgrade
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 13:36
if direct messabge me your email I will send you what I got
Detecting if you're vulnerable from the Heroku CLI
$ heroku run node -v -a APPNAME

Upgrading
If your application specifies a Node.js version in the engines section of your package.json, you will first need to update it to the appropriate version listed above. See this devcenter article on specifying a Node.js version for details.

To upgrade, you’ll need to push a new commit to your app, which will cause a deploy. If you don’t want to push any actual changes, this commit can be empty:

$ git commit --allow-empty -m “upgrade Node.js version"
$ git push heroku master

Verifying that you're running a fixed version of Node.js
After your application has been deployed, you can verify that the new version is installed by running:

$ heroku run node -v -a APPNAME

This will show one of the version strings from above.

there @TomBoudr_twitter I hope that helps you
Tom Boudreau
@TomBoudr_twitter
Jul 13 2017 13:43
@diabetic_dude_twitter I tried the link for the Heroku console in my app's admin page, but those commands didn;t work
also where would I run those deployments from exactly?
if I did an empty deployment
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 13:44
let me have a look at mine, I have only read the email at this point
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 13:49
@TomBoudr_twitter this is what I am seeing that I may be missing
CLI needs to be installed it seems
Matt Cooper
@MrCooper42
Jul 13 2017 13:51
so I spun up openaps last night and it has never gone through and actually started. It gets to mmtune: "916.732", 5 , -53(or some other number) Listening: .................. Error, retrying. I ocasionally get Starting pump-loop but then it runs again the last thing to go off before the listening is pkg_resources.DistributionNotFound: mmeowlink==0.11.0 anyone have any insights?
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 13:54
@TomBoudr_twitter I also did the same thing you did and it did not work for me also, that is why I think the CLI MAY need to be installed. but remember, I am reading and learning right along with you
@MrCooper42 what set up are you using?
do you have an explorer board?
Matt Cooper
@MrCooper42
Jul 13 2017 13:56
edison with spark board and the usp ti stick using mmeowlink setup
I do not
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 13:58
what happens when you run... what is it ???? lsusb
Matt Cooper
@MrCooper42
Jul 13 2017 13:59
sorry when I run what?
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 14:00
from your terminal try typing 'lsusb' no quotes
Miiko
@tw_miiko_twitter
Jul 13 2017 14:01
@TomBoudr_twitter I see, but what is the second pass phrase for enabling oref1. The first ist s@fety, but what is the second for Installation
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 14:02
@MrCooper42 that should show you the connected usb devices you are looking to confirm that your TI stick is connected
Matt Cooper
@MrCooper42
Jul 13 2017 14:02
i get three things back Bus 001 device 001 Linux Foundation 2.0 root hub``Bus 002 device 001 Linux Foundation 3.0 root hub``Bus 001 Device 002 OpenMoko, inc the also have id's but i figure its a little much to type
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 14:03
do you see your TI stick?
Matt Cooper
@MrCooper42
Jul 13 2017 14:06
I see it is connected in the /dev/tty area but unless it is the openmoko, inc i am not sure
i do get radio okay, listening ... no interfering pump comms detected from other tigs etc. when i refresh it... which leads me to think that it is there and connected
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 14:09
@MrCooper42 hang out for a bit and see if someone with a bit more experience comes around. I have not fully set up a TI yet so I am not the type of help you really want
Matt Cooper
@MrCooper42
Jul 13 2017 14:10
thanks for the help anyway! I also got the message from heroku about some of my apps have you guys made some head way on that? I haven't looked into it yet
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 14:11
I am working on installing CLI now and going to try and run the commands listed in the email I got and see what happens
I will keep you updated
Matt Cooper
@MrCooper42
Jul 13 2017 14:13
awesome thanks
Dirk Gastaldo
@dirkgastaldo
Jul 13 2017 14:16
Greetings. I assembled a second rig. How or where do I tell my Nightscout site to display my second rig (dgedison2)? Or do I not need to change anything?
cameronrenwick
@cameronrenwick
Jul 13 2017 14:20
@dirkgastaldo you really don't need to do anything. Assuming that you've flashed/installed oref and used your pump serial number, all the various parts for installation NS will get the info and things will be well. As Dana puts it, multiple rigs behave well together. When you look at your log, you'll see "listening" which is the rig listening for others.
if you hover your mouse over the openaps pill you'll actually see both rigs, one usually just standing by... all in it'll be seamless
Dirk Gastaldo
@dirkgastaldo
Jul 13 2017 14:24
@cameronrenwick Thank you. I see both rigs when I hover over the pill. Currently #2 is talking to a different pump. The blue basal rate display at the top of the NS screen is showing which rig? Or are both rigs issuing the same basal rate commands to both pumps?
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 14:35
@TomBoudr_twitter did you install CLI? https://devcenter.heroku.com/articles/heroku-cli I only ask because I did and was able to install it using the link I provided and confirm my version... So in short the comands i sent you around 8 am should work after you install the CLI
Luis Betancourt
@betluis
Jul 13 2017 14:41
could please somebody help me understand this problem?
Old pumphistory-24h refreshKeyError: 192
Waiting for 30s silence before mmtuning
Radio ok. Listening: .No pump comms detected from other rigs
Listening for 30s silence before mmtuning: .No pump comms detected from other rigs
mmtune: "916.660", 5, -76 waiting for 32 second silence before continuing
Radio ok. Listening: .No pump comms detected from other rigs
Unsuccessful supermicrobolus pump-loop at Thu Jul 13 09:40:15 CDT 2017
this is after changing the pump due to the support cap issue. this pump had been without use for several weeks. After starting looping (new set up with the new pump SN) this started happening continuously
Luis Betancourt
@betluis
Jul 13 2017 14:56
is this a problem with the pump or the rig?
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 14:56
it is very quiet here in the morning. More help will come. Is the pump working at all ? Are you getting temp basils set one it? I have no idea what the refreshkeyerror is
Luis Betancourt
@betluis
Jul 13 2017 15:00
no temp basals or SMB, i.e. the loop is dead... NS loop is ok though we can see BG fine
thanks
I rebooted pump and rig, no improvement
but really have no clue what this error could be
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 15:02
yeah give it time people will show up that can help.... and yeah you got me, I have never seen the refreshkeyroor and checked readthedocs.io and it is not mentioned
Shawn
@dyjaks
Jul 13 2017 15:03
Has anyone here tried a durable offline solution? I'm going on a climibing trip soon and am trying to figure out a case and placement solution for the rig. I want it to house an explorer, 2x2500 batteries, and a hardwired g5. That part isn't too hard, but making something strong and not too large so if I land on it it doesn't get in the way is proving difficult.
cameronrenwick
@cameronrenwick
Jul 13 2017 15:04
@dirkgastaldo my understanding is that generally you only have 1 pump per NS instance. I think that's really prone to problems if you have 2 pumps pointing to the same site. Are these 2 pumps on different people or is it you've got a backup pump that you're running? Not too sure your setup. My earlier comments were really regarding the practice of having multiple rigs talking to a single pump. That's perfectly fine and NS works great. 2 pumps, unless one is stopped allowing the other to run and have data go to NS is the only way I can see that working. Others hopefully will chime in
Luis Betancourt
@betluis
Jul 13 2017 15:07
@diabetic_dude_twitter I see @peterfulvi had this issue (or a similar one) but can't see if or what the solution was
Tom Boudreau
@TomBoudr_twitter
Jul 13 2017 15:18
Thanks @diabetic_dude_twitter for helping me realize the Heroku CLI is a different thing from their console on the app settings screen (which lets you run commands)
Dana Lewis
@danamlewis
Jul 13 2017 15:20
@tw_miiko_twitter you should not be asking for the safety phrases, or posting them here. If you have not found it, you need the read the docs more carefully. Nobody should be posting the safety phrases in this channel.
@dyjaks back in the day, I think people were using larger durable cases they found at fishing/outdoor type stores
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 15:29
anytime @TomBoudr_twitter did you install it? I hope so, I was able to verify my version with ease and now am seeing what else I can do with it
Scott Leibrand
@scottleibrand
Jul 13 2017 16:13
@betluis the oref1 pushover stuff has the hostname at the end currently.
Luis Betancourt
@betluis
Jul 13 2017 16:16
I see that, sorry
Scott Leibrand
@scottleibrand
Jul 13 2017 16:17
no worries
on the KeyError: 192 thing, it looks like we'll need to drill down into that report and see if we can get any more details on why it's failing
Andy Sharrow
@dramageek
Jul 13 2017 16:18
@stefanakerblom When the rig charges, it'll report the battery voltage as around 66%. There's no way to tell if it's charging, or actually at 66%, so it logs this message instead.
Scott Leibrand
@scottleibrand
Jul 13 2017 16:20
@betluis I would take a look in the docs for how to drill down and run reports manually, and do that for the one that is erroring out
Luis Betancourt
@betluis
Jul 13 2017 16:22
would you mind pointing out the part of documentation that has this?
Luis Betancourt
@betluis
Jul 13 2017 16:23
ok
going there
Joey
@Cooljoe04_twitter
Jul 13 2017 16:31
@betluis Not to be pointed, but has the pump you switched to been confirmed to be compatible with openAPS? Just wanting to check the obvious off the list since it seemed like the problems occurred when you changed the pump.
Luis Betancourt
@betluis
Jul 13 2017 16:32
@Cooljoe04_twitter yes, definitely, we have used it with openaps before
Scott Leibrand
@scottleibrand
Jul 13 2017 16:32
and if it were non-loopable, it'd still be able to pull pumphistory.
not a 512 or anything like that, is it?
Luis Betancourt
@betluis
Jul 13 2017 16:32
522
it has worked before perfectly
Scott Leibrand
@scottleibrand
Jul 13 2017 16:33
:+1:
Joey
@Cooljoe04_twitter
Jul 13 2017 16:46
@applehat I have an Edison+Explorer rig running dev release with 5000mA capacity and am currently at 55% battery after 36 hours of use. Been mostly on wifi, with some Bluetooth tethering. Your 6000mA should run for multiple days between charge with Edison+Explorer only. If you are talking about powering a receiver with it as well, that's a whole different ballgame.
10-4 @betluis, just making sure! LoL
Luis Betancourt
@betluis
Jul 13 2017 16:47
sure, thanks for helping out
Brian Rabinovitz
@bdr1177
Jul 13 2017 16:49
OpenAPS is looping but I'm getting a non-specified error. Mmtune looks good. Anyone have any ideas: Jul 13 12:38:27 openaps pump-loop.log: Starting pump-loop at Thu Jul 13 12:38:27 EDT 2017: Jul 13 12:39:00 openaps pump-loop.log: Radio ok, Listening: .No pump comms detected from other rigs Jul 13 12:39:38 openaps pump-loop.log: Old pumphistory: Refresh Jul 13 12:39:39 openaps pump-loop.log: Error, retrying Jul 13 12:40:25 openaps pump-loop.log: Listening: .No pump comms detected from other rigs Jul 13 12:40:57 openaps pump-loop.log: mmtune: pump://JSON/mmtune/monitor/mmtune.json Jul 13 12:40:57 openaps pump-loop.log: reporting monitor/mmtune.json Jul 13 12:41:02 openaps pump-loop.log: "916.780", 5, -59 Starting pump-loop at Thu Jul 13 12:41:02 EDT 2017: Jul 13 12:41:38 openaps pump-loop.log: Radio ok, Listening: .No pump comms detected from other rigs Jul 13 12:43:13 openaps pump-loop.log: Old pumphistory: RefreshStarting pump-loop at Thu Jul 13 12:43:13 EDT 2017: Jul 13 12:43:46 openaps pump-loop.log: Radio ok, Listening: .No pump comms detected from other rigs Jul 13 12:44:32 openaps pump-loop.log: Old pumphistory: Refresh Jul 13 12:44:32 openaps pump-loop.log: Error, retrying
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 16:49
@Cooljoe04_twitter do you have a link to the battery you are using?
Joey
@Cooljoe04_twitter
Jul 13 2017 16:53
Wish I had more things to try Luis, just learning it myself!
@diabetic_dude_twitter Adafruit battery #328
Diabetic_Dude
@diabetic_dude_twitter
Jul 13 2017 16:56
ok yeah have that one, getting about 5 hours out of it. thing is though, if I plug it into my rileylink it gets over 16
still looking at my pump logs to find the issue think I am not getting full charges or something
Scott Leibrand
@scottleibrand
Jul 13 2017 17:00
@bdr1177 looks like you need to drill down the same way @betluis is doing
Joey
@Cooljoe04_twitter
Jul 13 2017 17:00
Tried different USB chargers just in case its an issue with low output from the USB, causing low lipo charger output too?
Luis Betancourt
@betluis
Jul 13 2017 17:01
trying to... ;)
Brian Rabinovitz
@bdr1177
Jul 13 2017 17:02
@scottleibrand I'm not with the rig and just looking at papertrail right now. Would there be any point in having the person with the rig turn it off and then back on manually?
Joey
@Cooljoe04_twitter
Jul 13 2017 17:03
Hey Scott, question you might have the answer to readily: When on wifi and you leave its range, it kicks over to Bluetooth within a couple minutes. When I get back into the range of the wifi, how long before it drops the Bluetooth connection? I did a reboot on the rig this morning to force it to re-check/re-connect to the wifi. Left for lunch and came back again and it's still not dropping the Bluetooth connection.
Luis Betancourt
@betluis
Jul 13 2017 17:19
@scottleibrand I changed the rig, same error... this should be the pump then
all these rigs have now the latest dev version
Joey
@Cooljoe04_twitter
Jul 13 2017 17:20
Update: gave it about 15 minutes of being while well within the wifi range of the building and it was still maintaining the Bluetooth connection (and draining my battery). Did another rig reboot and it connected back to wifi again and began looping (again), all without moving my physical location (aka: same wifi coverage).
Luis Betancourt
@betluis
Jul 13 2017 17:21
when running openaps refresh-pumphistory-24h manually, it does not show any error...
I wonder if I should change the pump now... but could it be a pump issue or something the software should be able to manage...
Scott Leibrand
@scottleibrand
Jul 13 2017 17:52
How far away was the pump when it was failing? What mmtune results were you getting?
Luis Betancourt
@betluis
Jul 13 2017 17:59
The pump was on the night table besides the bed, normal distance
Not sure about mmtune question how can I find out?
Scott Leibrand
@scottleibrand
Jul 13 2017 18:09
on ~25% of failures it should run mmtune and show results in pump-loop.log
Luis Betancourt
@betluis
Jul 13 2017 18:17
Ok let me check the log
Tim Street
@tim2000s
Jul 13 2017 18:47
@tepidjuice the reason I am suggesting a reflash is because xDrip writes to an SQLite database and for some reason that seems not to be populating in your case.
When you reinstall OpenAPS it doesn't reinstall the database.
You may want to try deleting that before reinstalling OpenAPS again.
Marco
@CaptainBalou
Jul 13 2017 19:00
I seem to have some issues with "the rig doesn’t see the pump sometimes“. I have no bigger distance than usual, no configuration changes except updating to latest dev. I would not think this is software related. So I inspected the rig itself and found out that one part is lose. Not soldered correctly on the ExplorerBoard v1.2. I think this is the antenna wich could be the troublemaker in this case. Can someone confirm that this is the antenna? If not what part is this?
ExplorerBoard - 1.jpg
ExplorerBoard - 2.jpg
The end with the circle around is lose...
Matt Cooper
@MrCooper42
Jul 13 2017 19:07
anyone here good with mmeowlink?
pkg_resources.DistributionNotFound: mmeowlink==0.11.0 is my error but I do have a mmeowlink-source folder in my root ~/
Andy Sharrow
@dramageek
Jul 13 2017 19:25
@CaptainBalou That's part of the antenna, along with the board trace next to it.
Marcus Lindfors
@Mackwe
Jul 13 2017 19:31
@CaptainBalou Had the same problem too. Yes. It is the antenna. Got it fixed by a local electronics guy (I don't touch surface mounts for good reason). Works fine now. :-)
Marco
@CaptainBalou
Jul 13 2017 19:51
@dramageek Thanks!
Andy Sharrow
@dramageek
Jul 13 2017 19:52
@CaptainBalou Alternatively, the Explorer board does have a connector for an external antenna.
Marco
@CaptainBalou
Jul 13 2017 19:52
@Mackwe Thanks Marcus. Why don’t you touch them? No glue how to solder or is that part especially tricky?
@dramageek It does? Oh really?
Scott Leibrand
@scottleibrand
Jul 13 2017 19:53
@MrCooper42 did you install mmeowlink manually, or did oref0-setup do it for you? if the latter, which version of oref0?
Marcus Lindfors
@Mackwe
Jul 13 2017 19:56
@CaptainBalou Too scared to mess it up. Too sensitive components.. And if I warm it just a tiny bit too long.. :-/
sdneufer
@sdneufer
Jul 13 2017 19:56
image.png
Question on oref1; I see a few microboluses, however mainly it seems to be high temp'ing. Is this was it should do?
Scott Leibrand
@scottleibrand
Jul 13 2017 19:57
@sdneufer don't guess, check the logs & pill to see what it's deciding and why
sdneufer
@sdneufer
Jul 13 2017 19:59
I know. I was having trouble finding the corresponding line in the pump log
Scott Leibrand
@scottleibrand
Jul 13 2017 20:00
fair enough. look for a pump-loop.log run that has insulinReq, and see if it says anything about microbolusing.
Andy Sharrow
@dramageek
Jul 13 2017 20:00
@CaptainBalou If you head back to the center of the board from the antenna, it's the circle with the tiny pin in the middle. Labeled A2.
if you go that route, something like this is probably your best bet to keep it portable: https://www.adafruit.com/product/1991
Joey
@Cooljoe04_twitter
Jul 13 2017 20:09
@dramageek Do you know, can that antenna be tuned/trimmed to 900mhz (instead of 800/900/1800/1900/2100 capable) to reduce overall size, while maximizing signal reception on our desired frequency?
Luis Betancourt
@betluis
Jul 13 2017 20:11
@scottleibrand , here the log, with the mmtunes part. does this say what you mean?

Starting supermicrobolus pump-loop at Thu Jul 13 15:03:01 CDT 2017 with 18 second wait_for_silence:
MDT CGM configured; not waiting
Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK.
Attempting to retrieve MDT CGM data from pump
MDT CGM data retrieved
No new MDT CGM data to reformat

Old pumphistory-24h, waiting for 18 seconds of silence: Radio ok. Listening: .No pump comms detected from other rigs
Old pumphistory-24h refreshKeyError: 192
Waiting for 30s silence before mmtuning
Radio ok. Listening: .No pump comms detected from other rigs
Listening for 30s silence before mmtuning: .No pump comms detected from other rigs
mmtune: "916.636", 5, -83 waiting for 46 second silence before continuing
Radio ok. Listening: .No pump comms detected from other rigs
Unsuccessful supermicrobolus pump-loop at Thu Jul 13 15:06:46 CDT 2017

unfortunately I don't have another pump available now and I am stuck with this failure
Scott Leibrand
@scottleibrand
Jul 13 2017 20:17
ok. any chance you can temporarily move the rig closer to the pump? -83 might work, but it's a bit weak.
Luis Betancourt
@betluis
Jul 13 2017 20:17
I can
let me do this
Luis Betancourt
@betluis
Jul 13 2017 20:31
pump sensor and rig all in contact with each other... let's see what the log says.
Marcus Lindfors
@Mackwe
Jul 13 2017 20:49
@dramageek So A2 is the TI antenna? Cool. you know what the antenna connector on the Edison is for? WiFi?
tepidjuice
@tepidjuice
Jul 13 2017 20:55
@tim2000s ok How do I delete the database?
As I can't reflash.
Luis Betancourt
@betluis
Jul 13 2017 21:17
@scottleibrand here the log with the rig in contact with the pump

Starting supermicrobolus pump-loop at Thu Jul 13 16:08:02 CDT 2017 with 7 second wait_for_silence:
MDT CGM configured; not waiting
Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK.
Attempting to retrieve MDT CGM data from pump
MDT CGM data retrieved
No new MDT CGM data to reformat

Old pumphistory-24h, waiting for 7 seconds of silence: Radio ok. Listening: .No pump comms detected from other rigs
Old pumphistory-24h refreshKeyError: 192
Waiting for 30s silence before mmtuning
Radio ok. Listening: .No pump comms detected from other rigs
Listening for 30s silence before mmtuning: .No pump comms detected from other rigs
mmtune: "916.684", 5, -38 Unsuccessful supermicrobolus pump-loop at Thu Jul 13 16:10:47 CDT 2017

Starting supermicrobolus pump-loop at Thu Jul 13 16:11:03 CDT 2017 with 3 second wait_for_silence:
MDT CGM configured; not waiting
Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK.
Attempting to retrieve MDT CGM data from pump
MDT CGM data retrieved
MDT New cgm data reformatted
Old pumphistory-24h, waiting for 3 seconds of silence: Radio ok. Listening: .No pump comms detected from other rigs
Old pumphistory-24h refreshKeyError: 192
Unsuccessful supermicrobolus pump-loop at Thu Jul 13 16:12:27 CDT 2017

Starting supermicrobolus pump-loop at Thu Jul 13 16:13:02 CDT 2017 with 7 second wait_for_silence:
MDT CGM configured; not waiting
Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK.
Attempting to retrieve MDT CGM data from pump
MDT CGM data retrieved
MDT New cgm data reformatted
Old pumphistory-24h, waiting for 7 seconds of silence: Radio ok. Listening: .No pump comms detected from other rigs
Old pumphistory-24h refreshKeyError: 192
Unsuccessful supermicrobolus pump-loop at Thu Jul 13 16:14:30 CDT 2017

Starting supermicrobolus pump-loop at Thu Jul 13 16:15:02 CDT 2017 with 7 second wait_for_silence:
MDT CGM configured; not waiting
Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK.
Attempting to retrieve MDT CGM data from pump
MDT CGM data retrieved
MDT New cgm data reformatted
Old pumphistory-24h, waiting for 7 seconds of silence: Radio ok. Listening: .No pump comms detected from other rigs
Old pumphistory-24h refreshKeyError: 192
Unsuccessful supermicrobolus pump-loop at Thu Jul 13 16:16:30 CDT 2017

Starting supermicrobolus pump-loop at Thu Jul 13 16:17:02 CDT 2017 with 29 second wait_for_silence:
MDT CGM configured; not waiting

Tim Street
@tim2000s
Jul 13 2017 21:22
@tepidjuice I don't know. Probably worth checking the SQLite documentation.
Marco
@CaptainBalou
Jul 13 2017 21:50
@dramageek Oh that’s cool. I have never realized that A2 connector. Perhaps @rsilvers129 already has a case designed to use that external antenna?! :-)
@tim2000s
@tepidjuice I don't know. Probably worth checking the SQLite documentation.
@CaptainBalou
@Mackwe @dramageek Oh that’s cool. I have never realized that A2 connector. Perhaps @rsilvers129 already has a case designed to use that external antenna?! :-)
Dirk Gastaldo
@dirkgastaldo
Jul 13 2017 22:49
@cameronrenwick Cameron, thank you. Yeah, I won't normally have two pumps involved. Implementing ew backup pump and new backup rig today, somewhat rushed, and didn't think through the ramifications. I will make both rigs point to the same pump, which I'm wearing. I was an idiot in setting it up. Waited until the last minute this morning, and I forgot to review and set my basal rates on the pump. Unfortunately the basal rates for the previous pump user were much higher than mine. I experienced some uncomfortable lows this morning and afternoon, the first urgent low alarm since I began looping on Monday, and an "EAT SOMETHING NOW" text from my wife.
PieterGit
@PieterGit
Jul 13 2017 23:41

installing latest dev on rpi3 with a bluetooth tethering and pebble option fails with

checking linux/if_alg.h presence... yes
checking for linux/if_alg.h... yes
checking for GLIB... no
configure: error: GLib >= 2.28 is required
Couldn't make bluez

I see two possible options:

  1. encourage people to upgrade their Rasperry to stretch (which has bluez 5.43, which is good enough for openaps afaict),
  2. or add the missing packages so that bluez is manually compiled
    I think I prefer 1. (because that will save compiling bluez and have more upgrades included and uses packages instead of manual compiling).

What are the opinions on this one?

See https://packages.qa.debian.org/b/bluez.html for current debian versions of the bluez pacakge
cameronrenwick
@cameronrenwick
Jul 13 2017 23:52
@dirkgastaldo so this is the point where everyone here will remind you that high bloodsugar will kill you in 15 years but low bloodsugar can kill you today... please be careful. Don't wanna read the riot act but make sure you're checking things out. Glad your wife was on top of your low :clap: