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

12th
Jul 2017
Miiko
@tw_miiko_twitter
Jul 12 2017 00:39
Target 0 on pump loop is already illogical, or?
Scott Leibrand
@scottleibrand
Jul 12 2017 00:40
https://github.com/openaps/oref0/blob/dev/lib/profile/targets.js#L63-L71 disallows targets below 80 or above 200. I'm trying to figure out if there's any way around that
if you can reproduce the issue, that would help tremendously
Miiko
@tw_miiko_twitter
Jul 12 2017 00:41
No body sets a target to 0, APS shoud not actually set a target to 0, or?
Scott Leibrand
@scottleibrand
Jul 12 2017 00:42
If you set a target to 0, oref0 is programmed to change that to 80
similarly if you set 400, it will change it to 200
if you can figure out how to make it report target of 0 again, I'd like to troubleshoot further and figure out what additional safety checks are needed to prevent that
Miiko
@tw_miiko_twitter
Jul 12 2017 00:43
It sounds logical-
I will do it later, now must go in Bed. thank you.
Scott Leibrand
@scottleibrand
Jul 12 2017 00:48
k. I'm also removing one code path we never use that possibly could bypass those checks if you were doing so on purpose. it would've required having an explicit target_bg set, though, and not sure how that's possible if you didn't modify the input files to determine-basal.
Miiko
@tw_miiko_twitter
Jul 12 2017 00:53
I did nothing, except via NS a normal higher temp target entered.
Scott Leibrand
@scottleibrand
Jul 12 2017 00:54
I wonder if there was an error downloading the temptarget.json from NS or something, and it got truncated.
Miiko
@tw_miiko_twitter
Jul 12 2017 00:56
One quetion: Is the the Targets on Pump-Loop floating, although he has set a fixed target.
Scott Leibrand
@scottleibrand
Jul 12 2017 00:57
what do you mean by floating?
Miiko
@tw_miiko_twitter
Jul 12 2017 01:00
I have observed that the target in the pump-loop is constantly changing, although I only set a fixed target, such as 100 is set in the pump target, but in the loop will be observed sometimes 84, sometimes 95, often change.
Scott Leibrand
@scottleibrand
Jul 12 2017 01:01
take a look at pump-loop.log: it explains why it's adjusting target.
Miiko
@tw_miiko_twitter
Jul 12 2017 01:01
What is the reason for this?
Ok
are the two settings in preferences.json that control that behavior
Miiko
@tw_miiko_twitter
Jul 12 2017 01:09
yes, i read it. Thank you, good day!
Scott Leibrand
@scottleibrand
Jul 12 2017 01:09
I just opened openaps/oref0#541 to stop using profile.target_bg, and just use min_bg and max_bg to set target_bg.
Miiko
@tw_miiko_twitter
Jul 12 2017 01:13
does it mean, i must update my rig?
Scott Leibrand
@scottleibrand
Jul 12 2017 01:14
no, unless you determine that your profile.json somehow got a target_bg value in it
when you get a chance, I would focus your efforts on trying to reproduce the issue.
Miiko
@tw_miiko_twitter
Jul 12 2017 01:18
understood. Thanks and See you!
Russ Inskeep
@RussInskeep_twitter
Jul 12 2017 03:10
Hey everyone new here I'm trying to find out if anyone has experimented with using the 630g with openaps system
Scott Leibrand
@scottleibrand
Jul 12 2017 03:15
If you're an end user, I wouldn't hold your breath: get a loopable 52x/72x instead. But if you're a hardware guru who wants to help reverse engineer the CNL and 6x0g comms protocols, there is a channel for that.
Russ Inskeep
@RussInskeep_twitter
Jul 12 2017 03:16
I'm new here and have a spare pump here so I was trying to see if it was possible and I'm no guru
Dana Lewis
@danamlewis
Jul 12 2017 03:16
Russ - not currently possible right now unfortunately
We don't have the ability to remotely set temp basals on the 6series pumps
Russ Inskeep
@RussInskeep_twitter
Jul 12 2017 03:18
The can be remotely set though already
Can't they
Like I said I'm new with pump period so these are just questions
Dana Lewis
@danamlewis
Jul 12 2017 03:20
No, they can't on the 6series.
Russ Inskeep
@RussInskeep_twitter
Jul 12 2017 03:21
I just got the 670 G today and found this site
Dana Lewis
@danamlewis
Jul 12 2017 03:22
The models we can set temp basals on remotely are listed here: http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-0/hardware/pump.html
Russ Inskeep
@RussInskeep_twitter
Jul 12 2017 03:22
I'm new in the world of pumps period so still learning
Dana Lewis
@danamlewis
Jul 12 2017 03:25
:+1: there might be some Facebook groups with other new 670G users that might be helpful to look into as you're getting started. This channel is for DIY efforts, so we don't have many (any) 670G users here.
aliceinnowonderlands
@aliceinnowonderlands
Jul 12 2017 03:27
I just googled 670G - "The first and only system that constantly self-adjusts to automatically keep your sugar levels in range" LOL
Russ Inskeep
@RussInskeep_twitter
Jul 12 2017 03:48
I appreciate you sending me somewhere else but I have an older pump I was wanting to experiment with didn't no if it was possible
aliceinnowonderlands
@aliceinnowonderlands
Jul 12 2017 03:49
what is model of your old pump?
Katie Aldridge
@kcrcgm
Jul 12 2017 03:50
Hi friends, I am trying to run Autotune as a one off for the first time. I entered 2017-07-11 as the start date, and it's been running and running and running and it's up to like 2019-04-12 with no end in sight. I'm guessing you can't put today's date in that spot??? How do I make it stop?
Dana Lewis
@danamlewis
Jul 12 2017 03:50
Control c
Katie Aldridge
@kcrcgm
Jul 12 2017 03:50
Bless you, thank you
Russ Inskeep
@RussInskeep_twitter
Jul 12 2017 03:51
I guess you all aren't looking to highly on the 670 my other pump was the 630 G was already ld it can't be used here
Q
aliceinnowonderlands
@aliceinnowonderlands
Jul 12 2017 03:55
yeah, from what i read i think that you can't use this pump either
i guess if you want to try to experiment with openaps (open solutions looks safer for me) you can look for used 512/712/515/715/522/722 model somewhere ( craiglist, medwow etc)
Russ Inskeep
@RussInskeep_twitter
Jul 12 2017 03:58
Thanks I will keep looking
Katie Aldridge
@kcrcgm
Jul 12 2017 04:07
How long does Autotune take to run as a one off if I put in a week? Just wondering if something is wrong bc it's going slowly. Thanks!
Dana Lewis
@danamlewis
Jul 12 2017 04:10
Are you running on the rig? Shouldn't take terribly long, but if you're running manually it might be worth running on your computer if you want to do longer than a week next
Katie Aldridge
@kcrcgm
Jul 12 2017 04:12
Yes on the rig. Thank you for the suggestion. I'll try that!
Dana Lewis
@danamlewis
Jul 12 2017 04:15
:+1:
woodtrevor
@woodtrevor
Jul 12 2017 04:41
Can anyone help with device status not getting uploaded to NS? A few weeks ago my pump and openaps pills stopped working. basal pill does update, temp basals do render, pump treatments show up. On the rig upload/ns-status.json looks good. And when I manually run ns-upload it appears to run without error. But when I go look at the devicestatus collection in mlab it never got there. Not sure where to go from here.
Assuming you've also checked for future treatments in NS already
woodtrevor
@woodtrevor
Jul 12 2017 04:43
Grrr. Thanks. Did a search but didn't find that section in the docs. I'll check it out.
Dana Lewis
@danamlewis
Jul 12 2017 04:44
It's in dev docs - no worries
woodtrevor
@woodtrevor
Jul 12 2017 04:45
I haven't checked for future treatments. Do you have a link for that one as well...? :smile:
Joey
@Cooljoe04_twitter
Jul 12 2017 04:46
Got my Pi rig running standard release about 5 weeks ago, and updated to dev release about 2 weeks ago after some coding tweaks+testing with scott (to compensate for using a carelink stick still). Finally received my first explorer board+edison that got ordered a while back and got the newly-updated dev release installed and so-far working properly. Excited to see how everything will work when I don't have to be tethered to an outlet like with my Pi! Hoping to test out bluetooth and very interested to see what the pump-to-edison range is since the carelink stick is very short.
Dana Lewis
@danamlewis
Jul 12 2017 04:46
:+1:
What day did you install dev?
Joey
@Cooljoe04_twitter
Jul 12 2017 04:47
just finished
there wasn't passphrases when I installed it before. lol
Dana Lewis
@danamlewis
Jul 12 2017 04:48
Ok great! If works well into tomorrow, would appreciate a comment on the open PR with your setup (pump and CGM, etc) type to help us check off more setup types toward releasing
Heh ;)
Joey
@Cooljoe04_twitter
Jul 12 2017 04:48
Should be quite an interesting test this week with this rig, doing a karate camp all week. lol
Dana Lewis
@danamlewis
Jul 12 2017 04:49
Heh high temp targets are awesome for extra activity times
Joey
@Cooljoe04_twitter
Jul 12 2017 04:49
Without having it today I was high all morning, and low all afternoon.
I do need/want to figure out how to add the temp targets as well. Saw them mentioned, just not sure how/where to actvate them. Would love if that was possible via Apple Watch.
To have an eating soon and exercising soon button
Dana Lewis
@danamlewis
Jul 12 2017 04:53
You can activate them via Nightscout; and wherever you have ifttt capabilities (so probably that for your Apple Watch, plus ifttt has an iPhone widget that's nice). http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-4/ifttt-integration.html
Joey
@Cooljoe04_twitter
Jul 12 2017 04:53
Awesome. Thanks for the link, will look into it now!
Dana Lewis
@danamlewis
Jul 12 2017 04:54
:+1:
Joey
@Cooljoe04_twitter
Jul 12 2017 04:59
the tutorial for that page does need a minor tweak. Maker Webhooks is now just Webhooks. Doesn't show up with the instructed search phrase of "maker"
Dana Lewis
@danamlewis
Jul 12 2017 05:03
PRs welcome :smile:
woodtrevor
@woodtrevor
Jul 12 2017 05:17
Thanks @danamlewis. Nice write up. Size On Disk = 496.00 MB is my issue. I started the repairDatabase database command to try to get it to compact. Apparently it is running now... so fingers crossed.
Dana Lewis
@danamlewis
Jul 12 2017 05:21
@woodtrevor :+1:
tepidjuice
@tepidjuice
Jul 12 2017 05:28
I'm now completely lost with this xDripAPS problem. I have no idea what's going wrong.
Joey
@Cooljoe04_twitter
Jul 12 2017 05:36
IFTTT is all set up, and initial testing is that it works great/as I hoped with watch+NS. Thanks for the help @danamlewis. Heading to bed for tonight, I will give the dev. release a more thorough run through/testing at camp tomorrow/this week!
Dana Lewis
@danamlewis
Jul 12 2017 05:38
:+1:
Luis Betancourt
@betluis
Jul 12 2017 06:08
Hello, does anybody have a practical solution to the Minimed pump lose motor cap (cap sticking out issue)?
Tim Street
@tim2000s
Jul 12 2017 08:08
@tepidjuice What have you checked so far? Sorry on phone so searching is hard
Are you entering treatment data into xDrip?
tepidjuice
@tepidjuice
Jul 12 2017 08:13
I've reinstalled openaps, changed the ip, tried running the xdripaps process manually.
No.
Tim Street
@tim2000s
Jul 12 2017 08:14
I've just taken a look at mine and it all seems to be working okay. Which version of xDrip+ are you on?
Matthias
@ceben80
Jul 12 2017 08:15
Unfortunately I do only have mobil phone access, but @colinlennon any clue here Colin regarding xdripaps?
tepidjuice
@tepidjuice
Jul 12 2017 08:15
1704032023
Tim Street
@tim2000s
Jul 12 2017 08:16
Also how many urls in the rest api settings?
tepidjuice
@tepidjuice
Jul 12 2017 08:17
1
Matthias
@ceben80
Jul 12 2017 08:17
@tepidjuice stupid q, you have defined the REST url w/o a "s"
tepidjuice
@tepidjuice
Jul 12 2017 08:17
Yes.
If I put an s there I get a handshake error... Obviously.
Matthias
@ceben80
Jul 12 2017 08:18
Yes Tim good point a did have 2 url incl the xdrip url and this did not work
Grr I hate this mobil autocorrection :smile:
tepidjuice
@tepidjuice
Jul 12 2017 08:19
Only one URL.
Matthias
@ceben80
Jul 12 2017 08:21
Are u able to start the xdrip daemon manually on your rig (the cron line)?
Or to identify per ps ... grep xrip (something like this)
tepidjuice
@tepidjuice
Jul 12 2017 08:22
I'd can do that. Unfortunately I'm in a talk on aps at the moment.
It finishes in an hour so I can screen into my rig then.
Matthias
@ceben80
Jul 12 2017 08:23
:+1:
I would love to post the console stuff but only mobil :worried:
tepidjuice
@tepidjuice
Jul 12 2017 08:25
That's fine. If your about in an hour then I can debug properly. I'm in mobile at the moment too.
Matthias
@ceben80
Jul 12 2017 08:33
In xdrip have u checked or unchecked "skip lan upload" (extra options), I found it does not work when check this
tepidjuice
@tepidjuice
Jul 12 2017 08:37
It is unchecked.
Also I don't enter treatments. In xdrip.
Colin Lennon
@colinlennon
Jul 12 2017 08:39
Hi @tepidjuice - I should be able to get on here over the new few hours too, so hopefully I can try and help figure out what the issue is
tepidjuice
@tepidjuice
Jul 12 2017 08:39
Great thanks.
Matthias
@ceben80
Jul 12 2017 08:46
Morning Colin :wink:
Colin Lennon
@colinlennon
Jul 12 2017 08:47
Morning :)
Tim Street
@tim2000s
Jul 12 2017 09:49
I think we're both around.
Hey Colin!
tepidjuice
@tepidjuice
Jul 12 2017 09:50
I have about 10 minutes. Is there some quick debugging that can be done?
Colin Lennon
@colinlennon
Jul 12 2017 10:00
Hi Tim!
I've got a meeting, will be back in around 30 minutes... what I'd suggest checking first is if the process is running
ps aux | grep xDripAPS
There should be three lines returned - two showing the python process and one for the grep itself
tepidjuice
@tepidjuice
Jul 12 2017 10:01
Ok got to go now I might not be able to get in until tonight.
And yes the process is running.
Colin Lennon
@colinlennon
Jul 12 2017 10:02
No worries @tepidjuice - I'll try and check in again later then.
peterfulvi
@peterfulvi
Jul 12 2017 11:41
What am i missing here? im trying to setup this offline website according to the docs.
  1. I added the script by using nano editor and saving it as http.sh in the enact directory
  2. I added @reboot cd /root/myopenaps/enact && python -m SimpleHTTPServer 1337 and */5 * * * * (bash /root/http.sh) 2>&1 | tee -a /var/log/openaps/http.log to the crontab file
  3. i added http://myip:1337/index.html to http widget and also on my browser and got nothing
Tim Street
@tim2000s
Jul 12 2017 11:45
You've put the script in a different place from where the crontab runs it from.
The document says that the script is running in /root while the output is in /myopenaps/enact
So from root do 'mv myopenaps/enact/http.sh http.sh'
Tim Street
@tim2000s
Jul 12 2017 11:54
@danamlewis I've just transferred my Heroku data over to open humans as I need to clear out the database (I'm hitting my limit). When I get there next time, how do I avoid overwriting existing data in openhumans?
peterfulvi
@peterfulvi
Jul 12 2017 12:15
@tim2000s Its in my openaps directory and still no display, i,ll do a reboot
stefanakerblom
@stefanakerblom
Jul 12 2017 12:31
I expected oaps to do some microbulusing since i just ate 15 carbs and entered them including the bolus wit the bolusexpert. I can see something regarding UAMpredBG but nothing about smb. is there an easy way to check if smb is working? which report? dont have pushover.
stefanakerblom
@stefanakerblom
Jul 12 2017 12:37
Anyone who has a link how to get my lipo batteries charged to 100%. The highest I get in Nightscout is 64% regardless if it is a 2000 or 2500 maH. I have been told that the EB only provides 250 mA charging capacity and that might not be enough.
peterfulvi
@peterfulvi
Jul 12 2017 12:51
image.png
image.png

does this look right i get a directory page

then i get this output when iclick on it

Matthias
@ceben80
Jul 12 2017 13:13
@stefanakerblom correct the exb has 250mA charging current, I have rig included an external Adafruit charger w/ 500mA and I'm using a 2500mAh lipo, works great (for me) and the exb won't be hot at any time :wink:
stefanakerblom
@stefanakerblom
Jul 12 2017 13:15
@ceben80 Do you have a link to that gadget?
Matthias
@ceben80
Jul 12 2017 13:22
From sparkfun w/ 500mA http://www.exp-tech.de/lipo-charger-basic-micro-usb pretty tiny :+1:
Or from Adafruit http://www.exp-tech.de/adafruit-micro-lipo-w-microusb-jack-usb-liion-lipoly-charger default current is 100mA, to boost up to 500mA you will need to close a bridge by soldering, easy
Somewhere in my twitter you'll find my rig pic shown the Adafruit charger included in the box
Matthias
@ceben80
Jul 12 2017 13:30
@stefanakerblom my first Edison rig (w/ Intel mini breakout board) I am using a 2000mAh lipo w/ a 1A Adafruit charger, pretty good
cameronrenwick
@cameronrenwick
Jul 12 2017 13:37
@stefanakerblom probably the easiest way is to just review your log tail -n 100 -F /var/log/openaps/pump-loop.log
if you haven't done so, adding the alias' to your rig is super helpful.
stefanakerblom
@stefanakerblom
Jul 12 2017 13:38
@cameronrenwick Thanx. Well the microboluses finally appeared in nightscout. I am too impatient.
cameronrenwick
@cameronrenwick
Jul 12 2017 13:39
@stefanakerblom I'd also recommend what @ceben80 has suggsted
stefanakerblom
@stefanakerblom
Jul 12 2017 13:39
@stefanakerblom done
@ceben80 Thx. Will consider when I build my 2nd rig with silvers enclosure. Awaiting the 2nd exb from us.
cameronrenwick
@cameronrenwick
Jul 12 2017 14:01
@stefanakerblom you can also hover your mouse over the NS openaps pill and it will show you exactly what's happening. I find I use that a lot when I'm watching for SMB as it reports what time SMB occurred
Tim Street
@tim2000s
Jul 12 2017 14:12
@peterfulvi, the http.sh needs to be in you ~/ directory for the crontab to work. The index.html file needs to be in ~/myopenaps/enact
You may need to touch the index.html file to create it.
peterfulvi
@peterfulvi
Jul 12 2017 14:47
@tim2000s i have the http.sh file in the myopenaps directory. Now do i need to make an index.html file in the enact directory? How do i do that?
Scott Leibrand
@scottleibrand
Jul 12 2017 16:11
@stefanakerblom your battery will always show in the low 60% range when charging, because that's what the charging voltage corresponds to. once you stop charging, it should jump up in to the 90% range
@stefanakerblom rather than wait on Nightscout, you should be watching pump-loop.log on your rig.
@sdneufer @betluis I wouldn't contact Medtronic for support on a loopable pump. they'll want to replace it with a non-loopable one. You can try a DIY fix for the pump motor cap issue - see @xpucuto's recent posts in https://gitter.im/MilosKozak/AndroidAPS for more details. just be sure to disconnect the pump from the body before even touching a loose motor cap: pressing on it even a little bit is an easy way to deliver dozens of units of extra insulin unintentionally.
Diabetic_Dude
@diabetic_dude_twitter
Jul 12 2017 16:34
@scottleibrand do you think an external charger may get any better of a charge when charging batteries? I understand what you are saying about Nightscout, but, I also can not help but think I
am getting full charges on either of my two rigs. I have tested 4 different batteries and tried on two different rigs but cant seem to get 8 hours outta my rigs also for reference this is one of my main batteries Lithium Ion Cylindrical Battery - 3.7v 2200mAh
or do you know of a command that I could run to see what process is taking up all my battery power on my rig?
Hristina Dimova
@xpucuto
Jul 12 2017 16:41
@betluis @scottleibrand the instructions are in the Looped FB group . https://www.facebook.com/groups/TheLoopedGroup/. Two days ago we were discussing this problem in the Loop channel as well . Let me know if you need any help fixing it .
Scott Leibrand
@scottleibrand
Jul 12 2017 17:02
@diabetic_dude_twitter first, check what the charge level is when you first unplug the rig. it should be in your pump-loop.log as well. the external charger would only be useful if you're not plugging it in long enough to fully charge on the internal charger
Joey
@Cooljoe04_twitter
Jul 12 2017 17:05
@diabetic_dude_twitter I just got my new Edison rig set up last night which has 2x 2500mA batteries in it (5000mA total capacity). They charged fine last night from 70% to 99% in 5 hours as I slept on just being plugged into the Edison+Explorer with a 2.5Amp output USB charger as the source. After 4.5 hours of use they are down to 88%, meaning 600mA used in 4.5 hours with my setup using the dev release and smb. If that is a linear figure, that would be basically 37 hours of operation on a full charge for my rig. That should give you a little over 16 hours (if draw of rig/discharge of batteries are linear) with your 2200mA battery. When you disconnect your charger in the morning, does the log show your batteries at 100% on your rig?
Tim Street
@tim2000s
Jul 12 2017 17:11
To generate an index.html file in ~/myopenaps/enact siimply type:
touch ~/myopenaps/enact/index.html
@peterfulvi you either need to move the http.sh file or change the crontab entry to run the file to point at it.
To make the index.html file type:
touch ~/myopenaps/enact/index.html
Shawn
@dyjaks
Jul 12 2017 17:33
@Cooljoe04_twitter: How are you connecting 2 batteries together?
Joey
@Cooljoe04_twitter
Jul 12 2017 17:55
Wired in parallel. (Batt1 red, Batt2 red, Red to Edison plug all soldered together. Batt1 black, Batt2 black, Black to Edison plug all soldered together.) Basically make a splitter with 2 female connectors (to match up to the battery plugs) and one male connector (to plug into the Edison).
peterfulvi
@peterfulvi
Jul 12 2017 17:59
@tim2000s thanks
peterfulvi
@peterfulvi
Jul 12 2017 18:07
@tim2000s I followed the docs and directs everything to the enact directory, so i generated the index.html in there. This is the line in my cron @reboot cd /root/myopenaps/enact && python -m SimpleHTTPServer 1337
Dana Lewis
@danamlewis
Jul 12 2017 18:23
@tim2000s :+1: for donating data. Trying to get some time to update the tool so it doesn't overwrite the old data set
Miiko
@tw_miiko_twitter
Jul 12 2017 18:26
@scottleibrand Hi, I have new something: In Preferences.json set max_iob 2, and then reboot, but it is different, 2 in Prefernces.json, but 3 in autotune-profile.json. Why?
Scott Leibrand
@scottleibrand
Jul 12 2017 18:27
Have you run autotune since changing it?
By default it only runs at midnight
Miiko
@tw_miiko_twitter
Jul 12 2017 18:28
how can I run it manuel
stefanakerblom
@stefanakerblom
Jul 12 2017 18:30
I like the color scheme in the pump-loop trace. Still a novice to understand all what's going on. Still it is like watching a nice movie.
Miiko
@tw_miiko_twitter
Jul 12 2017 18:31
@stefanakerblom yes, much better!
stefanakerblom
@stefanakerblom
Jul 12 2017 18:31
is white blue and green fine?
edison on battery 83 %. Nightscout shows 10%?????
cameronrenwick
@cameronrenwick
Jul 12 2017 18:35
@stefanakerblom could that 10% be your uploader (phone) by any chance?
stefanakerblom
@stefanakerblom
Jul 12 2017 18:36
@cameronrenwick OMG. You are right!
how can I show my edison battery on nightscout. I have the uploader obviously and the pump battery
Miiko
@tw_miiko_twitter
Jul 12 2017 18:41
@scottleibrand no, not running. How can I it manuell?
Tim Street
@tim2000s
Jul 12 2017 18:42
@peterfulvi that line starts the http server.
The line in cron you need to change is:
/5 * (bash /root/http.sh) 2>&1 | tee -a /var/log/openaps/http.log
Make the the beginning (bash /root/myopenaps/http.sh)
Matthias
@ceben80
Jul 12 2017 18:47
@stefanakerblom the 10% might be the battery level of your mobil, pls verify
Scott Leibrand
@scottleibrand
Jul 12 2017 18:47
@tw_miiko_twitter you can find the autotune line in cron and run that manually
Matthias
@ceben80
Jul 12 2017 18:52
@stefanakerblom sorry forget already an answer above
Shawn
@dyjaks
Jul 12 2017 19:07
@Cooljoe04_twitter: Thanks, figured as much. I'll probably end up splicing it in so I can keep the pig tail.
stefanakerblom
@stefanakerblom
Jul 12 2017 19:09
@ceben80 Yes. How can I see the edison battery level?
Martin Haeberli
@mhaeberli
Jul 12 2017 19:09
ok i’m confused - logically, i’m aware of the idea of running an http server on openaps. but I wasn’t aware that it is common. sure, I should RT(F)D. but can someone post a 1-2 line summary of what i normally would look at through such a server?
Matthias
@ceben80
Jul 12 2017 19:16
@stefanakerblom normally mouse-over in ns or on unix cat ~/myopenaps/monitor/edison-battery.json
Cameron Chunn
@applehat
Jul 12 2017 19:16
@mhaeberli most people are just pushing up a single page with stats (loop status from enacted.json, ISF stuff, ect) so that they can either consume that data through something like http-widget or by opening a browser window.
Martin Haeberli
@mhaeberli
Jul 12 2017 19:17
@applehat thx - so I should just rtfd to figure out normal setup
Matthias
@ceben80
Jul 12 2017 19:18
@stefanakerblom are you using Android? If yes, this might be helpful https://github.com/ceben80/http-widget
Joey
@Cooljoe04_twitter
Jul 12 2017 19:25
@dyjaks I just got a couple prewired pigtails off the internet so that I can keep factory connectors on everything. Then I can run just 1 battery if I need to by just unplugging the other and/or hit swap them as necessary. Be sure to check polarity though, had to swap several of the connectors around to make it correct.
Scott Leibrand
@scottleibrand
Jul 12 2017 19:26
@mhaeberli AFAIK the http-widget stuff and similar is still very uncommon, and more importantly, is not well documented yet. if you can help figure it out and document it, that'd be great
Dana Lewis
@danamlewis
Jul 12 2017 19:29
(By reading and PRing the existing docs, which are there)
peterfulvi
@peterfulvi
Jul 12 2017 19:32
@tim2000s Got it! Thanks a bunch
Matthias
@ceben80
Jul 12 2017 19:33
Regarding the http-widget stuff I think only a few screenshots are still missing, Tim did a great job!
Scott Leibrand
@scottleibrand
Jul 12 2017 19:34
:+1:
Matthias
@ceben80
Jul 12 2017 19:34
I will do that when I am back home, reviewing rtd and will finalize it
thom1394
@thom1394
Jul 12 2017 19:38
ceben80's directions work very well. But you need to change most of the source files to smb-suggested.json or smb-enacted.json to get the appropriate information. I have been using it for some time because I cna not get the pebble link to function...
Matthias
@ceben80
Jul 12 2017 19:41
That's right, I did it before oref1 :wink:
dan lightner
@danimaniac
Jul 12 2017 19:45
@ceben80 @applehat @mhaeberli I wonder if a better goal is to integrate the http-widget information into xDrip+ which already displays CGM data on the Android homescreen? As it is now we have 2 separate widgets...one displaying CGM data and one displaying rig/OpenAPS data. Just curious what the more polished solution would be...I am no programmer and can not help in this regard. Just wondering.
Scott Leibrand
@scottleibrand
Jul 12 2017 19:51
it also sounds like y'all are building a lot of really Android-specific solutions. we should look at trying to make them a bit more cross-platform to work with iOS too
dan lightner
@danimaniac
Jul 12 2017 19:51
...sounds like we have the beginnings of an OpenAPS app. Oh, wait...
Dana Lewis
@danamlewis
Jul 12 2017 19:54
There's a repo for that ;)
Matthias
@ceben80
Jul 12 2017 19:58
You're right Scott, unfortunately http-widget stuff is only available for Android, but it was easy to implement when I was looking for an offline monitoring mechanism, kind of quick and dirty but it works :smile:
Martin Haeberli
@mhaeberli
Jul 12 2017 19:58
@danimaniac - indeed - good idea - in our case, we are rather iOS centric ...
@scottleibrand :+1:
re documenting
katie disimone
@Kdisimone
Jul 12 2017 20:06
trying new rig setup. Used the new script to install master, then updated to dev. reran script fresh. Watching pump loop and it’s stuck in a spidev already in use loop. rebooted once already about to reboot a second time
  File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/serial_interface.py", line 23, in open
    raise AlreadyInUseException("%s already in use" % self.device)
mmeowlink.vendors.serial_interface.AlreadyInUseException: /dev/spidev5.1 already in use
Usage: grep [OPTION]... PATTERN [FILE]...
Try 'grep --help' for more information.
 Listening: ..No pump comms detected from other rigs
..mmeowlink.vendors.serial_interface.AlreadyInUseException: /dev/spidev5.1 already in use
Unsuccessful supermicrobolus pump-loop at Wed Jul 12 13:00:57 PDT 2017
......
Starting supermicrobolus pump-loop at Wed Jul 12 13:01:02 PDT 2017 with 3 second wait_for_silence:
Waiting up to 4 minutes for new BG: ls: cannot access monitor/pump_loop_completed: No such file or directory
...Listening for 30s silence before mmtuning: ....
Radio error found at Wed Jul 12 13:03:01 PDT 2017


Broadcast message from root@edison1 (Wed 2017-07-12 13:03:02 PDT):

Rebooting to fix radio errors!
The system is going down for reboot at Wed 2017-07-12 13:08:01 PDT!


Starting supermicrobolus pump-loop at Wed Jul 12 13:03:02 PDT 2017 with 7 second wait_for_silence:
Waiting up to 4 minutes for new BG: ls: cannot access monitor/pump_loop_completed: No such file or directory
Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK. Old pumphistory-24h, waiting for 7 seconds of silence: Radio ok. Listening: ...
Broadcast message from root@edison1 (Wed 2017-07-12 13:04:01 PDT):

Rebooting to fix radio errors!
The system is going down for reboot at Wed 2017-07-12 13:08:01 PDT!

……
Rebooted a second time...
@scottleibrand thoughts?
Scott Leibrand
@scottleibrand
Jul 12 2017 20:08
did you end up with two pump-loops in cron?
katie disimone
@Kdisimone
Jul 12 2017 20:08
said Y to deleting existing
how could I check if there’s two?
Scott Leibrand
@scottleibrand
Jul 12 2017 20:08
ok, that plus a reboot should clear it out
katie disimone
@Kdisimone
Jul 12 2017 20:09
it’s already rebooted though.
Scott Leibrand
@scottleibrand
Jul 12 2017 20:09
yeah, that's what I mean. should be fixed by now. if it's not...
have you seen any /dev/spidev5.1 already in use since the reboot? or just Rebooting to fix radio errors!?
katie disimone
@Kdisimone
Jul 12 2017 20:10
Starting supermicrobolus pump-loop at Wed Jul 12 13:03:02 PDT 2017 with 7 second wait_for_silence:
Waiting up to 4 minutes for new BG: ls: cannot access monitor/pump_loop_completed: No such file or directory
Radio ok. Listening: .No pump comms detected from other rigs
Preflight OK. Old pumphistory-24h, waiting for 7 seconds of silence: Radio ok. Listening: ...
Broadcast message from root@edison1 (Wed 2017-07-12 13:04:01 PDT):

Rebooting to fix radio errors!
The system is going down for reboot at Wed 2017-07-12 13:08:01 PDT!

............
Broadcast message from root@edison1 (Wed 2017-07-12 13:05:01 PDT):

Rebooting to fix radio errors!
The system is going down for reboot at Wed 2017-07-12 13:08:01 PDT!

.No pump comms detected from other rigs
Old pumphistory-24h refresh
Broadcast message from root@edison1 (Wed 2017-07-12 13:06:01 PDT):

Rebooting to fix radio errors!
The system is going down for reboot at Wed 2017-07-12 13:08:01 PDT!

ed
Old pumphistory, waiting for 7 seconds of silence: Radio ok. Listening: .No pump comms detected from other rigs
RefreshValueError: need more than 0 values to unpack

Unsuccessful supermicrobolus pump-loop at Wed Jul 12 13:06:50 PDT 2017

Broadcast message from root@edison1 (Wed 2017-07-12 13:07:01 PDT):

Rebooting to fix radio errors!
The system is going down for reboot at Wed 2017-07-12 13:08:01 PDT!


Starting supermicrobolus pump-loop at Wed Jul 12 13:07:02 PDT 2017 with 14 second wait_for_silence:
Waiting up to 4 minutes for new BG: ls: cannot access monitor/pump_loop_completed: No such file or directory
Radio ok. Listening: .^[[B^[[B^[[ANo pump comms detected from other rigs
Preflight OK. Old pumphistory, waiting for 14 seconds of silence: Radio ok. Listening: .No pump comms detected from other rigs
Refresh
this was right before the second reboot
Scott Leibrand
@scottleibrand
Jul 12 2017 20:10
because the Rebooting to fix radio errors! will always show up right after the reboot, until the pump-loop.log gets enough lines to age off the last /dev/spidev5.1 already in use error from its | tail
katie disimone
@Kdisimone
Jul 12 2017 20:10
so far looks better after this second reboot though
yeah…second reboot looking better. Weird…I’ve never seen the first reboot not fix things.
Scott Leibrand
@scottleibrand
Jul 12 2017 20:11
afaict without knowing the exact timing of your first reboot, I'd suspect it fixed the problem, and you just had residual Rebooting stuff in the logs
in rare cases if the pump-loop.log doesn't get a full 10 lines of output in 5 minutes, then it will do a second reboot
katie disimone
@Kdisimone
Jul 12 2017 20:12
must have been that
Scott Leibrand
@scottleibrand
Jul 12 2017 20:12
but you should never see a new /dev/spidev5.1 already in use after the reboot unless there's something wrong with your cron or something like that
katie disimone
@Kdisimone
Jul 12 2017 20:12
there may have been up farther in the logs after the first reboot…I may just not have captured it in the copy paste
looping now
Scott Leibrand
@scottleibrand
Jul 12 2017 20:13
we could probably add some logic to the reboot check to not reboot if uptime is less than 10 minutes or something like that, to reduce confusion and eliminate the possibility of two back to back reboots
I think some of the stuff Pieter has in his open PR would serve as a good way to check for that, if anyone wants to take that on
katie disimone
@Kdisimone
Jul 12 2017 20:14
It’s just so odd that I wondered if it were a code change issue. I don’t mind the second reboot. Not a huge deal and I doubt a brand new user would totally be upset by a second one
Scott Leibrand
@scottleibrand
Jul 12 2017 20:15
k. it does seem like a common source of confusion, though, so probably worth fixing at some point
Tiberiu Feredean
@feredean
Jul 12 2017 20:59

Hello, managed to set up the loop with nightscout, except for the basal rendering everything works, the basal is a continuous line even though in logs I'm seeing changes

Jul 12 23:00:58 edison pump-loop.log: Adjusting basal from 0.342 to 0.4
Jul 12 23:01:45 edison pump-loop.log: Adjusting basal from 0.296 to 0.35

I'm starting to wonder if these changes are supposed to be reflected in nightscout

katie disimone
@Kdisimone
Jul 12 2017 21:00
go to your NS settings and turn “basal render” drop down menu to “default"
@feredean
Tiberiu Feredean
@feredean
Jul 12 2017 21:01
Screen Shot 2017-07-13 at 00.01.21.png
read all the documentation i could find on the issue, not sure where I'm making a mistake
katie disimone
@Kdisimone
Jul 12 2017 21:03
well that’s because your loop is not enacting. See that x in your openaps pill…no temp basals are being set. And your “adjusting basal” message is not the same as “setting temp basal”. Adjusting basal is what autosens is adjusting your “regular” basal rate to for the purposes of determining what temp basal to set. You have autosens enabled so your regular basal rates will regularly be reassessed before determining a temp basal
Basically, I’m not sure you’re looping yet…what’s your pump-loop log say?
Tiberiu Feredean
@feredean
Jul 12 2017 21:04
that makes a lot of sense, here is the last log, I can send you more

Jul 13 00:02:28 edison pump-loop.log: Starting pump-loop at Thu Jul 13 00:02:26 EEST 2017:
Jul 13 00:03:00 edison pump-loop.log: Radio ok, Listening: .No pump comms detected from other rigs
Jul 13 00:03:01 edison pump-loop.log: Profile less than 60m old
Jul 13 00:03:01 edison pump-loop.log: glucose.json newer than temp_basal.json
Jul 13 00:03:34 edison pump-loop.log: Temp refreshed
Jul 13 00:03:39 edison pump-loop.log: {"carbs":15,"boluses":1.5,"mealCOB":0}
Jul 13 00:03:39 edison pump-loop.log: {"ratio":1.18}
Jul 13 00:03:39 edison pump-loop.log: {"iob":0.209,"activity":0.0085,"bolussnooze":0,"basaliob":0,"netbasalinsulin":0,"hightempinsulin":0,"time":"2017-07-12T21:01:04.000Z"}
Jul 13 00:03:39 edison pump-loop.log: {"delta":-1,"glucose":111,"short_avgdelta":-1.11,"long_avgdelta":-1.67}
Jul 13 00:03:39 edison pump-loop.log: {"duration":0,"rate":70,"temp":"percent"}
Jul 13 00:03:39 edison pump-loop.log: Adjusting basal from 0.3 to 0.35
Jul 13 00:03:39 edison pump-loop.log: Adjusting target_bg from 90 to 85
Jul 13 00:03:39 edison pump-loop.log: Adjusting sens from 40.16 to 34
Jul 13 00:03:39 edison pump-loop.log: Carb Impact: -0.2 mg/dL per 5m; CI Duration: 0 hours
Jul 13 00:03:39 edison pump-loop.log: Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: 0 hours
Jul 13 00:03:40 edison pump-loop.log: determine-basal://text/shell/enact/suggested.json
Jul 13 00:03:40 edison pump-loop.log: reporting enact/suggested.json
Jul 13 00:03:40 edison pump-loop.log: {"temp":"absolute","bg":111,"tick":-1,"eventualBG":103,"snoozeBG":103,"predBGs":{"IOB":[111,109,108,107,105,104,103,103,102,102,101,101,101]},"COB":0,"IOB":0.209,"reason":"COB: 0, Dev: -1, BGI: -1.45, ISF: 34, Target: 85; Eventual BG 103 >= 85, no temp, setting 0.65U/hr","duration":30,"rate":0.65}
Jul 13 00:03:49 edison pump-loop.log: pump://JSON/set_temp_basal/enact/enacted.json
Jul 13 00:03:49 edison pump-loop.log: reporting enact/enacted.json
Jul 13 00:03:50 edison pump-loop.log: {"requested":{"duration":30,"rate":0.65,"temp":"absolute"},"bg":111,"temp":"percent","snoozeBG":103,"timestamp":"2017-07-13T00:03:48.164472","predBGs":{"IOB":[111,109,108,107,105,104,103,103,102,102,101,101,101]},"rate":70,"reason":"COB: 0, Dev: -1, BGI: -1.45, ISF: 34, Target: 85; Eventual BG 103 >= 85, no temp, setting 0.65U/hr","COB":0,"eventualBG":103,"recieved":false,"duration":0,"tick":-1,"IOB":0.209}
tepidjuice
@tepidjuice
Jul 12 2017 21:05
@ceben80 @tim2000s @colinlennon So I'm free now if any of you are about I'd really appreciate trying to debug this xdripAPS problem.
katie disimone
@Kdisimone
Jul 12 2017 21:12
@feredean You need to change the temp basal type in your pump to units/hour instead of % then it will work
Tiberiu Feredean
@feredean
Jul 12 2017 21:22
Sorry took some time to confirm, it seems to work now thank you very much @Kdisimone !
Marco
@CaptainBalou
Jul 12 2017 21:24

Does anybody know what OpenAPS want to tell me with RefreshWarning: clock input Invalid Date is unzoned; please pass clock-zoned.json instead? :-) Never seen this messages before in the logs afaik.

No smb_enact needed. Temp refreshed: monitor/temp_basal.json: {"duration":30,"rate":0.45,"temp":"absolute"}
RefreshWarning: clock input Invalid Date is unzoned; please pass clock-zoned.json instead
ed pumphistory
Checking pump clock: "2017-07-12T23:00:51+02:00" is within 1m of current time: Wed Jul 12 23:01:02 CEST 2017
and that pumphistory is less than 1m old.  Temp refreshed
Autotune exists! Hoorah! You can use microbolus-related features.

Do I have to change some Parameters or command aliases?

Scott Leibrand
@scottleibrand
Jul 12 2017 21:30
@CaptainBalou what kind of setup are you running? manual, or oref0-setup? what version of oref0?
if you aren't doing anything custom, then most likely that just means that the clock refresh failed, and it'll be fine on the next run
Marcus Lindfors
@Mackwe
Jul 12 2017 21:55
@stefanakerblom Regarding the battery charge issue. Read something about the Edisons are made for 2000mAh batteries and the (slow) built-in charger will auto-stop after 6,5 hours, which is what's needed to fill a 2000mAh battery. To fully charge and empty 2500+ you would need to unplug & replug after a few hours or it won't charge fully. Workaround: Use external charger to get quicker (and full) charge. ;-)
Scott Leibrand
@scottleibrand
Jul 12 2017 22:15
@Mackwe nope, that only applies to the old Intel breakout board, not the Explorer board
Marcus Lindfors
@Mackwe
Jul 12 2017 22:16
@scottleibrand Thanks for clarifying! But the Explorer Board charger still outputs a lot less than 500 milliamps, correct?
Scott Leibrand
@scottleibrand
Jul 12 2017 22:20
depends on the version. in the most recent Explorer board version (1.2) the charging current was dropped from 500mA to something around 300mA I think, to help prevent overheating. given the ~100mA draw of the Edison, that's still plenty fast to recharge 16h worth of usage in less than 8h.
Marcus Lindfors
@Mackwe
Jul 12 2017 22:24
Yes. But it's a bit of a drawback of you run out of battery and need a quick recharge while out.. But yes. Correct (sufficient) charge rate is safer. Just best not forget to charge overnight.. ;-)
Thanks again @scottleibrand the impeccable intel (as usual). :-)
*for the..
Scott Leibrand
@scottleibrand
Jul 12 2017 22:25
and there's always the "keep a second rig plugged in so you can swap them" approach
Marcus Lindfors
@Mackwe
Jul 12 2017 22:26
Yup. Rig #2 in the mail ATM. :-)
Marco
@CaptainBalou
Jul 12 2017 22:29
@scottleibrand I am running latest dev installed via oref0-setup. I have not customized anything during setup or later. I am just wondering because I see this message a lot of time if I grep through the logs. And the way the message is written suggests me to do something to fix it. But I am not aware of what or how... But I'll first keep watching out how often this happens.
Joey
@Cooljoe04_twitter
Jul 12 2017 22:31
Or just keep a spare battery as a lower cost alternative to a second rig. (Yes a second, backup, rig is ideal, but a spare battery is a lot cheaper!)
Scott Leibrand
@scottleibrand
Jul 12 2017 22:37
@CaptainBalou if it was complaining about a timestamp other than Invalid Date it might be. that was the assumption when I wrote the error message, but apparently it's also triggered by invalid ones
@Cooljoe04_twitter yeah, but swapping lipo's at the JST connector is a pain. :)
@Mackwe :+1: you'll love having two. :)
Joey
@Cooljoe04_twitter
Jul 12 2017 22:40
I agree, that connector is a pain, could always make a "Y" like a dual battery setup and be able to hot swap them as needed and not reboot the unit either. Don't get me wrong, I would like a second rig too, just gotta build up some savings. All the accessories add up LoL
Scott Leibrand
@scottleibrand
Jul 12 2017 22:40
:+1:
Marcus Lindfors
@Mackwe
Jul 12 2017 22:42
@Cooljoe04_twitter Yes. But I think I need one stable (don't fiddle too much) rig and one for dev and backup.. I seem to have broken Bluetooth in my main rig at the mo and I am too scared to mess things up and kill the only working rig I have to fix it.. Waiting for the "backup" to arrive before doing something stupid..
Joey: A spare, charged battery is also easy to carry with you at all times. Thanks for that idea. I do have some 720mAh batteries laying around, so I'll just charge and carry one of those just in case!
Tiberiu Feredean
@feredean
Jul 12 2017 23:05
What causes this? It's been like this for 20 minutes, I've seen it before, restarting the board fixed it.
Jul 13 01:58:33 edison pump-loop.log: + ./change_setting.py /dev/spidev5.1 0x0B 0x38
Jul 13 01:58:33 edison pump-loop.log: RileyLink OK
Jul 13 01:58:33 edison pump-loop.log: Version: subg_rfspy 0.8
Jul 13 01:58:33 edison pump-loop.log: SUCCESS
Jul 13 01:58:33 edison pump-loop.log: + sleep 0.5
Jul 13 01:58:33 edison pump-loop.log: + exit 0
Jul 13 01:58:54 edison pump-loop.log: /usr/local/bin/oref0-subg-ww-radio-parameters-timeout: 8: kill: No such process
Jul 13 01:59:26 edison pump-loop.log: mmtune: pump://JSON/mmtune/monitor/mmtune.json
Jul 13 01:59:26 edison pump-loop.log: reporting monitor/mmtune.json
Jul 13 01:59:32 edison pump-loop.log: Starting pump-loop at Thu Jul 13 01:59:31 EEST 2017:
Jul 13 02:00:04 edison pump-loop.log: Radio ok, Listening: .No pump comms detected from other rigs
Jul 13 02:00:35 edison pump-loop.log: Old pumphistory: 
Jul 13 02:00:35 edison pump-loop.log: Error, retrying

Radio ok, Listening: .No pump comms detected from other rigs

Did the edison lose the connection to the pump?

Jake Punshon
@JakePunshon
Jul 12 2017 23:13
@Cooljoe04_twitter I have two 2000mAh batteries in parallel that I use for my APS. gives me lots of extra juice in case i have a late night, and it's always charged by morning.
Tim Street
@tim2000s
Jul 12 2017 23:38
@Mackwe You're better off with at least two Pis and at least two Edisons. Then you've got plenty of redundancy 😉
PieterGit
@PieterGit
Jul 12 2017 23:42
back to a pi3 rig :worried: because usb otg micro usb of explorer board has loose contact. Creating a case with USB protector probably needs to be higher on my todo list.
will see if I can solder it in place again
cameronrenwick
@cameronrenwick
Jul 12 2017 23:43
@Mackwe before I got my second rig set up I used one of these https://www.amazon.com/Docooler-Portable-Emergency-Flashlight-Cellphone/dp/B00DU4F8DM?SubscriptionId=AKIAILSHYYTFIVPWUY6Q&tag=duckduckgo-ffsb-20&linkCode=xm2&camp=2025&creative=165953&creativeASIN=B00DU4F8DM or similar. It's much easier to just plug this into your rig vs removing the battery from the jst terminal. That terminal is pretty fragile too so having a battery pack is a cheaper, easier way to go IMO
did anyone get this msg from Heroku? "[ACTION REQUIRED] Node.js security update on Heroku" looks legit but I've been getting a ton of spam lately and am just skeptical of everything...
live4sw
@live4sw
Jul 12 2017 23:51
I got it also
Scott Leibrand
@scottleibrand
Jul 12 2017 23:52
what version are they claiming is problematic, and what actions do they recommend?
Joey
@Cooljoe04_twitter
Jul 12 2017 23:52
@PieterGit I wonder if (either once you repair it, or even on a new unit) you were just epoxy the crap out of the connector to the board if it would help keep it from damaging the connectors when things aren't plugged/unplugged as delicately as we hope for?
live4sw
@live4sw
Jul 12 2017 23:52
I tried re-deploying but it still shows an old version, it says that it's currently running 6.10.0 but needs to be running 6.11.1
cameronrenwick
@cameronrenwick
Jul 12 2017 23:52
here's what the email said Scott

he 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:

4.8.4
6.11.1
7.10.1
8.1.4

LilDucky
@LilDucky
Jul 12 2017 23:53

I am in the process of getting bits together to help a friend to start looping. I am after some feedback if possible to confirm my intended direction. Currently using a Medtronic 640G with Enlite CGM. I am monitoring progress in the hacking of the newer 802.15.4 based Comms but want to get things going faster.
We have access to a carelink USB and a 754 pump on 2.6a which I have confirmed will talk to OpenAPS via an ancient Pi.

I'm debating whether to use an RPi3 or Edison. The Edison is great for the 915MHz gear but harder to modify for potential 2.4GHz use. I also propose to add a Qi charger into the micro USB to protect the port.

Has anybody used the NanoPi Neo Air yet? Thanks in advance and I hope to be able to add more to OpenAPS into the future.

live4sw
@live4sw
Jul 12 2017 23:53

You are receiving this email because you run at least one Node.js application on Heroku.

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:

4.8.4
6.11.1
7.10.1
8.1.4

Detecting if you're vulnerable from the Heroku CLI
$ heroku run node -v -a APPNAME

If your version is less than what's listed above, you will need to upgrade.

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.

I see people are talking about it in the CGM in the Cloud Facebook group, but no resolution yet. It sounds to me like there may need to be an update to Nightscout itself to specify an updated version of Node.js
@sulkaharo said in the main NS gitter "The 0.10 series used in most installs isn't listes as vulnerable, so most people should be ok"