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

27th
Dec 2017
Paul Dickens
@thebookins
Dec 27 2017 00:09
@PieterGit good point about splitting up the CR and ISF. My instinct is that there is something peculiar about breakfast though - something to do with the carbs all hitting after a night of fasting and very low basal need overnight, which is why I was wondering about associating the CR of 13 with the breakfast carbs, until all decayed. I am aware of the autotune limitations - we don't use it automatically, but from time-to-time only as a suggestion for what to change.
alimhassam
@alimhassam
Dec 27 2017 00:23
be ware that splitting the cr might make cob jump up and down more, depending on the time of the carbs
Paul Dickens
@thebookins
Dec 27 2017 00:37
ok, thanks @alimhassam
Craig Brenner
@cbrenner
Dec 27 2017 01:10

After updating one of our rigs to v0.6.0, I get this fatal effort message after running the running the oref0-setup script. Any help would be appreciated.

Removing any existing git
Removed any existing git
Checking openaps 0.2.1 installation with --nogit support
openaps 0.2.1-dev
Checking /root/myopenaps: mkdir: cannot create directory ‘/root/myopenaps’: File exists
/root/myopenaps initialized
Cloning decocare 0.1.0-dev: Cloning into 'decocare'...
remote: Counting objects: 11544, done.
fatal: write error: No space left on deviceMiB | 1.16 MiB/s s
Couldn't clone decocare 0.1.0-dev

Dana Lewis
@danamlewis
Dec 27 2017 01:11
No space left on device - search for disk space or disk full for the troubleshooting docs
Brian Cadre
@bcadre
Dec 27 2017 01:14
@PieterGit and @alimhassam thanks for the detailed info. Does the carb absorption rate (the parameter in NS) affect this calculation? It is a parameter used by oref0 to determine fractionCarbsLeft or anything else?
Dana Lewis
@danamlewis
Dec 27 2017 01:15
no. the NS one is an old remnant (also) of pre-looping days and does not impact OpenAPS in any way
can leave it blank
Brian Cadre
@bcadre
Dec 27 2017 01:19
@danamlewis thanks.
Craig Brenner
@cbrenner
Dec 27 2017 01:52

@danamlewis Ok, do ran through the disk space troubleshooting docs and found logs files were taking up 770M of space. Ran "logrotate -f /etc/logrotate.conf" and that didn't really do much to reduce size. Here are the largest files within the log directory in question. Any other thoughts on freeing up space?

-rw-r----- 1 root adm 212720653 Dec 26 17:46 syslog
-rw-r----- 1 root adm 209069527 Dec 26 17:40 kern.log
-rw-r----- 1 root adm 126825427 Dec 18 06:25 kern.log.1
-rw-r----- 1 root adm 125064311 Dec 18 06:25 syslog.1
-rw-r----- 1 root adm 9791730 Dec 26 17:45 user.log
-rw-r----- 1 root adm 7942556 Sep 17 06:17 kern.log.3.gz
-rw-r----- 1 root adm 3210527 Dec 26 17:46 messages
-rw-r----- 1 root adm 2555873 Dec 26 17:46 auth.log
-rw-r----- 1 root adm 1895223 Sep 16 06:25 kern.log.4.gz
-rw-r----- 1 root adm 1842723 Dec 26 17:40 daemon.log
-rw-r----- 1 root adm 1331200 Dec 26 15:08 debug
-rw-r----- 1 root adm 1134672 Oct 26 23:37 kern.log.2.gz
-rw-r----- 1 root adm 1092635 Nov 19 06:25 syslog.16.gz
-rw-r----- 1 root adm 1051847 Nov 2 06:25 syslog.24.gz
-rw-r----- 1 root adm 1049907 Nov 8 06:25 syslog.21.gz
-rw-r----- 1 root adm 1014128 Nov 21 06:25 syslog.15.gz
-rw-r----- 1 root adm 973457 Nov 15 06:25 syslog.18.gz
-rw-r----- 1 root adm 958688 Nov 25 06:25 syslog.13.gz
-rw-r----- 1 root adm 942228 Sep 12 06:25 messages.19.gz
-rw-r----- 1 root adm 938942 Nov 6 06:25 syslog.22.gz
-rw-r----- 1 root adm 917034 Nov 17 06:25 syslog.17.gz
-rw-r----- 1 root adm 901141 Dec 15 06:25 syslog.3.gz
-rw-r----- 1 root adm 884579 Nov 10 06:25 syslog.20.gz
-rw-r----- 1 root adm 877027 Nov 12 06:25 syslog.19.gz
-rw-r----- 1 root adm 857365 Dec 13 06:25 syslog.4.gz
-rw-r----- 1 root adm 856574 Nov 23 06:25 syslog.14.gz
-rw-r----- 1 root adm 846977 Dec 5 06:25 syslog.8.gz
-rw-r----- 1 root adm 835532 Nov 4 06:25 syslog.23.gz
-rw-r----- 1 root adm 801231 Nov 27 06:25 syslog.12.gz
-rw-r----- 1 root adm 795546 Dec 11 06:25 syslog.5.gz
-rw-r----- 1 root adm 792852 Dec 3 06:25 syslog.9.gz
-rw-r----- 1 root adm 792282 Dec 17 06:25 syslog.2.gz
-rw-r----- 1 root adm 786929 Nov 29 06:25 syslog.11.gz
-rw-r----- 1 root adm 776409 Nov 1 06:25 messages.11.gz
-rw-r----- 1 root adm 771923 Dec 1 06:25 syslog.10.gz
-rw-r----- 1 root adm 766829 Dec 7 06:25 syslog.7.gz
-rw-r----- 1 root adm 743057 Oct 10 06:25 daemon.log.5.gz
-rw-r----- 1 root adm 737753 Oct 31 06:25 daemon.log.4.gz
-rw-r----- 1 root adm 732404 Oct 26 06:24 messages.12.gz
-rw-r----- 1 root adm 729035 Dec 9 06:25 syslog.6.gz
-rw-r----- 1 root adm 701675 Sep 19 06:25 messages.18.gz
-rw-r----- 1 root adm 685784 Dec 14 06:25 daemon.log.2.gz
-rw-r----- 1 root adm 678249 Oct 7 06:25 messages.15.gz
-rw-r----- 1 root adm 655894 Sep 25 06:25 messages.17.gz
-rw-r----- 1 root adm 655429 Oct 19 06:25 messages.13.gz
-rw-r----- 1 root adm 651392 Nov 6 06:25 messages.10.gz
-rw-r----- 1 root adm 650652 Oct 1 06:25 messages.16.gz
-rw-r----- 1 root adm 639959 Oct 13 06:25 messages.14.gz
-rw-r----- 1 root adm 639749 Nov 14 06:25 daemon.log.3.gz
-rw-r----- 1 root adm 632858 Sep 27 06:23 user.log.5.gz
-rw-r----- 1 root adm 627555 Dec 17 06:25 messages.2.gz
-rw-r----- 1 root adm 622750 Nov 15 06:25 auth.log.13.gz
-rw-r----- 1 root adm 617964 Nov 8 06:25 auth.log.15.gz
-rw-r----- 1 root adm 617891 Nov 5 06:25 auth.log.16.gz
-rw-r----- 1 root adm 617072 Oct 18 06:25 auth.log.22.gz
-rw-r----- 1 root adm 613069 Nov 17 06:25 messages.8.gz
-rw-r----- 1 root adm 612577 Nov 25 06:25 messages.6.gz
-rw-r----- 1 root adm 609226 Nov 18 06:25 auth.log.12.gz
-rw-r----- 1 root adm 606548 Oct 15 06:25 auth.log.23.gz
-rw-r----- 1 root adm 605221 Nov 11 06:25 messages.9.gz
-rw-r----- 1 root adm 604884 Dec 15 06:25 auth.log.3.gz
-rw-r----- 1 root adm 604612 Nov 24 06:25 auth.log.10.gz

alimhassam
@alimhassam
Dec 27 2017 03:12
@cbrenner you could probbaly delete all these files, or you could start with everything older than say 1 week
Chris L
@woodduckie
Dec 27 2017 03:41

I have stubborn ISF values that remain after changing parameters in pump, deleting profile.json, pumpprofile.json, autotune folder, and autotune.json.
After finding stray values remaining, I am trying to revert to a single ISF aligned with autotune, but still see the remnant parameters in the profile.json
profile.json:

"sens": 130,
"isfProfile": {
"units": "mg/dL",
"user_preferred_units": "mg/dL",
"sensitivities": [
{
"i": 0,
"start": "00:00:00",
"sensitivity": 120.833,
"x": 0,
"offset": 0,
"endOffset": 360
},
{
"i": 12,
"start": "06:00:00",
"sensitivity": 120,
"x": 1,
"offset": 360
},
{
"i": 30,
"start": "15:00:00",
"sensitivity": 125,
"x": 2,
"offset": 900
},
{
"i": 38,
"start": "19:00:00",
"sensitivity": 130,
"x": 3,
"offset": 1140,
"endOffset": 1440
}

pumpprofile.json

"sens": 120,
"isfProfile": {
"units": "mg/dL",
"user_preferred_units": "mg/dL",
"sensitivities": [
{
"i": 0,
"start": "00:00:00",
"sensitivity": 120,
"x": 0,
"offset": 0,
"endOffset": 1440
}

how do I clear errant ISF paramters & force OpenAPS to read (& use) current pump settings?

Chris L
@woodduckie
Dec 27 2017 04:11
FYI - I've completely removed my autotune & settings folders again tonight and forced a refresh of pump settings. Errant ISF is gone for now. Will dig back in tomorrow if they reappear after nightly autotune run
 cd ~/myopenaps && rm -rf settings && rm -rf autotune

cd ~/myopenaps && killall -g oref0-pump-loop && openaps get-settings
Darryl Schick
@darrylschick
Dec 27 2017 04:39
Update to 0.6.0 and getting what looks like valid papertrail logs but no loop enactment, and every few minutes I get a flood (hundreds) of papertrail log entries like "Dec 26 23:34:55 ls9299 pump-loop.log: Warning; adjusting DIA from 2 to minimum of 3 hours" Any idea what is going on?
Dana Lewis
@danamlewis
Dec 27 2017 04:43
your DIA on the pump is 2; the rig is setting it to 3 which is the minimum now
if you want to cut down on logs, change it to 3 in your pump
Jacob H
@jdhigh
Dec 27 2017 04:49
I'm getting a bunch of npm WARN deprecated messages and npm WARN engine messages as I do the setup. Is this normal?
Jacob H
@jdhigh
Dec 27 2017 05:17
Thanks @darrylschick . I see it and sometimes wonder if I should do anything about it. But then I don't want to bother reflashing a rig or making things right if everything goes haywire after my attempted fix.
Scott Leibrand
@scottleibrand
Dec 27 2017 06:24
The npm deprecation warnings are normal. They’re downstream dependencies we have no real control over.
fidoman
@efidoman
Dec 27 2017 12:36
@PieterGit - I had one of my ports on the explorer board come off 2 days ago. Everything still works and I still have one functional port for charging and it happens to be the port for connecting directly to the computer. I would be interested in more details on how you fix the port. Perhaps a troubleshooting doc PR? I would also be interested in a case design that protects the ports better. Thanks for your post, it is good timing for me.
Milos Kozak
@MilosKozak
Dec 27 2017 17:43
does oref support all CGMs (over xdrip)?
or only Dexcom, Enlite 1?
Dana Lewis
@danamlewis
Dec 27 2017 17:46
Yes xdrip works @MilosKozak
Milos Kozak
@MilosKozak
Dec 27 2017 17:46
thanks
Tim Street
@tim2000s
Dec 27 2017 18:10
Basic question. Are these directories overly large?
125M ./root/.npm
118M ./root/src/decocare
223M ./root/src
351M ./root
188M ./var/cache/apt
191M ./var/cache
317M ./var
733M .
alimhassam
@alimhassam
Dec 27 2017 18:12
yes @tim2000s the decocare repo has a lot of data files which are not needed.
rm -rf ~/src/decocare/analysis would save a bunch of space.
maybe another branch with just the code would be useful.
Tim Street
@tim2000s
Dec 27 2017 18:14
Hmmm. Not so much space.
alimhassam
@alimhassam
Dec 27 2017 18:15
similarly you can run apt-get clean to remove space from /var/cache/apt
Tim Street
@tim2000s
Dec 27 2017 18:15
I’m struggling, as I’ve an Edison which is maxed out on disk space, but the usual suspects (autotune, logs, etc) aren’t causing an issue.
alimhassam
@alimhassam
Dec 27 2017 18:16
hmm after removing analysis from decocare my decocare repo went from 118M (like you) to ~19M
not sure why it didn't work for you
Tim Street
@tim2000s
Dec 27 2017 18:16
Saved 66M on mine
alimhassam
@alimhassam
Dec 27 2017 18:16
ok
still an improvement.
Tim Street
@tim2000s
Dec 27 2017 18:17
apt-get clean should be cronned as a normal item though. Huge saving.
alimhassam
@alimhassam
Dec 27 2017 18:17
try apt-get clean
right
someone suggested adding this at the end of the setup script
not sure if they raised the pr
Tim Street
@tim2000s
Dec 27 2017 18:18
I think we should include it to run weekly as part of the regularly installed cron
alimhassam
@alimhassam
Dec 27 2017 18:18
i don' think that's necessary, it only grows when you install something
so it should just be done after setup or run-again script
but definitely would be a good change to add.
ah, i just looked, and i had also removed logs from decocare, that might be why it saved a bit more than you.
Tim Street
@tim2000s
Dec 27 2017 18:29
That would explain the apt-get growth. Every dev update will have likely grown it, and as that’s not done from a set up script, even if it was in there, I wouldn’t have taken advantage.
Carl
@c-robertson
Dec 27 2017 18:59
About to setup my rig BT tethering to my phone (Galaxy S7 Android 7.0). Docs suggest 'BTAutoTething' app to do the auto tether thing, but I noticed there are a plethora of other apps available on Google play. Anyone have recommendations?
infoneo
@infoneo
Dec 27 2017 19:13
@c-robertson I use Automate + "Keep Bluetooth tethering alive" script
Carl
@c-robertson
Dec 27 2017 20:00
@infoneo Thx, do you know how the script works (i.e. does it fail benignly) when the phone picks up a known WiFi connection? Manually on the phone, it won't turn on BT tether when connected to WiFi Iwhich is desired behavior because the rig connects to WiFi too...)
infoneo
@infoneo
Dec 27 2017 20:35
@c-robertson no, it's just checks every 5 min if bt tethering is active, but the script (flowchart) may be easily modified - in Automate there is condition "Wi-Fi connected?"
Carl
@c-robertson
Dec 27 2017 20:39
@infoneo cool, Automate looks generally useful, thanks!
Jacob H
@jdhigh
Dec 27 2017 20:53
To anyone considering a Pixel 2 cell phone: I've been using it all day today for bluetooth tethering and it's working well. My carrier is Google Project Fi. (I mention carrier because someone else had probs with Pixel2, but this might be because some carriers lock out tethering if you have an unlimited plan. (Project FI isn't unlimited.)
alimhassam
@alimhassam
Dec 27 2017 21:05
@jdhigh did you go through places of low connectivity?
i don't have a pixel2, only a pixel, and seems like BT tethering (on the phone) stops when the phone's internet connection goes down.
Jacob H
@jdhigh
Dec 27 2017 21:19
I can't say that I have— I've only been around town and in the city today. Perhaps it would be different if I were hiking in a remote area.
alimhassam
@alimhassam
Dec 27 2017 21:24
or in a tunnel, or basement or something, or maybe i just have something to figure out with my setup.
Jacob H
@jdhigh
Dec 27 2017 21:30
Well, if your cell phone can't get data, bluetooth sharing isn't going to be useful. :smile:
philipgo
@philipgo
Dec 27 2017 22:00
@jdhigh It can be extremely useful for SSH access to the rig or offline transmission of CGM data from X Drip+ to the rig
Jacob H
@jdhigh
Dec 27 2017 22:05
^ This is true. But if your phone supports wifi hotspot as well, you can find your phone's IP address and then SSH in on port 22.