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

5th
May 2017
rsilvers129
@rsilvers129
May 05 2017 00:00
This would untangle variables and help autosense not be tricked by activity.
Scott Leibrand
@scottleibrand
May 05 2017 00:00
@mhaeberli after openaps ns-loop I would do openaps alias show ns-loop and start drilling down into the component aliases and report
rsilvers129
@rsilvers129
May 05 2017 00:01
You would have to run an app on an iPhone that would look at the HR and upload it to Nightscout.
Scott Leibrand
@scottleibrand
May 05 2017 00:02
from my perspective, it would be easier (better improvements per hour spent) for me to improve autosens to make better use of the CGM data we already get than to write an app to upload HR/step data.
but that might change once someone volunteers to help build the OpenAPS UI app that @danamlewis is designing and we get some progress on that...
Martin Haeberli
@mhaeberli
May 05 2017 00:07
@scottleibrand ok thx re deconstructing ns-loop; asking where to find its parent; what step is AFTER ns-loop - ; how to find sequence - because ns-loop looks good at the moment. I could reboot but would prefer to isolate root cause
Scott Leibrand
@scottleibrand
May 05 2017 00:08
you saw that ns-loop is run directly from cron, right?
Martin Haeberli
@mhaeberli
May 05 2017 00:12
Yes
And it seemed happy now standalone? I think. But I'll look at the alias as you suggest
Scott Leibrand
@scottleibrand
May 05 2017 00:13
openaps alias show get-ns-bg might be of interest
Martin Haeberli
@mhaeberli
May 05 2017 00:14
@scottleibrand and @all continuously grateful for your service and collaboration / help / guidance
K thx
Aaron Michelson
@Aaybob
May 05 2017 00:19
where is the oref1 setup guide?
Scott Leibrand
@scottleibrand
May 05 2017 00:20
there isn't one: it's still in dev and needs testing first
the testing guidelines we do have are in the PR
we need people to test whichever of those things they're comfortable with, and comment on the PR with feedback on what they tested and what their results were
rsilvers129
@rsilvers129
May 05 2017 00:22
@scottleibrand and @danamlewis - I will ask people I know to see if I can find someone to code the UI. It seems like a great summer intern project for a student who can spend every day on it.
garykidd
@garykidd
May 05 2017 00:31
@scottleibrand funny because I got home from long commute and everything was working fine...and then bammm same issue as before. I'm thinking root cause may be that the Edison CPU is not seated properly and that may have caused it....wonder if this is the reason others have encountered the issue as well....
So when I reviewed the steps necessary---and following docs verbatim (which I will PR accordingly including the compiling part)....it made me wonder if the relative references to the erase and write commands work at all....I;m sure it's that I'm doing something wrong but I could only get the commands to work by absolute reference to filepaths (still likley user misunderstanding on my part).
Scott Leibrand
@scottleibrand
May 05 2017 00:34
relative references work fine, but you (and the files) need to be in the right places
garykidd
@garykidd
May 05 2017 00:36
sure....just wanted to confirm that if I was in the exact directory where the ccprog file exists and went to run --let's just say the erase command for now----that it should work? Or is there any sort of requirement that you can't execute the command from from within its own directory??
PR to Dev?
Scott Leibrand
@scottleibrand
May 05 2017 00:59
@garykidd you need a ./ in front of it to run something that isn't in your PATH.
@johnnymule I think you have to remove something from an alias too to avoid trying to refresh one of those files.
Check the docs again for that.
tepidjuice
@tepidjuice
May 05 2017 01:01
interested in the ui app anywhere I can keep up to date on progress?
Scott Leibrand
@scottleibrand
May 05 2017 01:08
Did you confirm your change took effect?
dan lightner
@danimaniac
May 05 2017 01:12
What is this "OpenAPS App UI" you casually mention in passing? Is this something everyone knows about already?
Scott Leibrand
@scottleibrand
May 05 2017 01:29
@danimaniac no, it hasn't been fully designed or written up yet. Idea is to have an iPhone/Android app to interact with the OpenAPS rig to visualize everything, change settings, enter carbs, set temp targets, etc.
And of course have it work fully offline as well.
I can't remember if there is an issue open for it yet or not.
Dana Lewis
@danamlewis
May 05 2017 02:07
(Not yet)
Luis Betancourt
@betluis
May 05 2017 02:47
hello all, I am having a continuous problem with a pump
tepidjuice
@tepidjuice
May 05 2017 02:47
@betluis what's the problem?
Luis Betancourt
@betluis
May 05 2017 02:48
Old pumphistory-24h refreshError, retrying
does not get a good pump history-24hr refresh
therefore never goes on to calculate a temp basal
any idea why this may happen?
thanks @tepidjuice for asking
it retries and retries endlesly
tepidjuice
@tepidjuice
May 05 2017 02:49
I've not encountered that error. So I don't know how to fix it. Sounds like it might be a communication error though? Is the pump close enough can you send a button press?
Scott Leibrand
@scottleibrand
May 05 2017 02:50
you need to run the sub-commands manually to see why it's failing. see the docs </brokenrecord>
Luis Betancourt
@betluis
May 05 2017 02:51
I have changed pumps and it works
@scottleibrand ok, I'll see those docs
Scott Leibrand
@scottleibrand
May 05 2017 02:51
:+1:
FWIW, the new version now in dev will be better about reporting some info about the error when it has to retry
but for anyone still on master, you have to do the manual drill-down to see why
Aaron Michelson
@Aaybob
May 05 2017 02:52
@betluis yes that just happened on my rig when i comented out my wifi connections and forgot i did that.
Luis Betancourt
@betluis
May 05 2017 02:52
Radio ok, Listening: .................No pump comms detected from other rigs
MDT cgm data retrieved
MDT cgm data reformatted
Old pumphistory-24h refreshError, retrying
Starting pump-loop at Thu May 4 21:44:45 CDT 2017:
Radio ok, Listening: .No pump comms detected from other rigs
MDT cgm data retrieved
MDT cgm data reformatted
Old pumphistory-24h refreshError, retrying
Listening: .^Croot@edison2:~#
root@edison2:~# ./view_log.sh
Listening: .No pump comms detected from other rigs
mmtune: pump://JSON/mmtune/monitor/mmtune.json
reporting monitor/mmtune.json
"916.660", 5, -73 Starting pump-loop at Thu May 4 21:47:20 CDT 2017:
Radio ok, Listening: ...................No pump comms detected from other rigs
MDT cgm data retrieved
MDT cgm data reformatted
Old pumphistory-24h refreshError, retrying
Listening: .No pump comms detected from other rigs
mmtune: pump://JSON/mmtune/monitor/mmtune.json
reporting monitor/mmtune.json
"916.636", 5, -59 Starting pump-loop at Thu May 4 21:50:30 CDT 2017:
Radio ok, Listening: .^Croot@edison2:~# ./view_log.sh
"916.660", 5, -73 Starting pump-loop at Thu May 4 21:47:20 CDT 2017:
Radio ok, Listening: ...................No pump comms detected from other rigs
MDT cgm data retrieved
MDT cgm data reformatted
Old pumphistory-24h refreshError, retrying
@Aaybob but this same rig works fine with other pump and I have not touched the wifi settings
Aaron Michelson
@Aaybob
May 05 2017 02:56
@betluis I just saw that in the thread above too.
Luis Betancourt
@betluis
May 05 2017 02:56
right
Aaron Michelson
@Aaybob
May 05 2017 02:56
well its next step if it worked would be to read from the pump
Dana Lewis
@danamlewis
May 05 2017 02:56
@dmdfreak if you can't find the flag in the code, you may need to spend more time getting to know the code before enabling advanced features
Aaron Michelson
@Aaybob
May 05 2017 02:56
or not
Sgoldberg89
@Sgoldberg89
May 05 2017 02:57
I am getting the following error message in my tail: File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/subg_rfspy_link.py", line 72, in check_setup
self.serial_rf_spy.sync()
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/serial_rf_spy.py", line 121, in sync
raise CommsException("Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?")
mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?
any ideas? I updated everything yesterday
Aaron Michelson
@Aaybob
May 05 2017 02:57
I have seen that. The other unit i just fixed i started over by deleting openaps and src and starting over at download supporting files or something....check the wifi and then stage 2
Luis Betancourt
@betluis
May 05 2017 02:58
I'd like to reset the pump completely, any idea how to do that?
kind of start with the pump from scratch
Aaron Michelson
@Aaybob
May 05 2017 02:58
A battery removal?
Luis Betancourt
@betluis
May 05 2017 02:58
did that for a whole day
still the same issue
Aaron Michelson
@Aaybob
May 05 2017 02:59
try a whole month
it stores electricity in capacitors and they're good capacitors
Eric
@ecc1
May 05 2017 03:00
It takes several days. When you've waited long enough, you'll get an A## error (I can't remember exactly, maybe A21?) indicating it's forgotten all the settings and you'll be prompted to set the time, etc.
Aaron Michelson
@Aaybob
May 05 2017 03:00
yes i knew it was between one day and one month
but as a solution it is unlikely
Scott Leibrand
@scottleibrand
May 05 2017 03:02
@Sgoldberg89 does a search of the docs have anything about that error?
if not, it should be added to the troubleshooting section: first step is to check the two LEDs next to the microUSB ports
Sgoldberg89
@Sgoldberg89
May 05 2017 03:03
would I search for.... "line 72, in check_setup
self.serial_rf_spy.sync()"
Luis Betancourt
@betluis
May 05 2017 03:04
@Aaybob why you think as a solution is unlikely?
Aaron Michelson
@Aaybob
May 05 2017 03:04
@Sgoldberg89 did you let it run for fifteen minutes?
Sgoldberg89
@Sgoldberg89
May 05 2017 03:04
@Aaybob yea, it was working fine last night
Aaron Michelson
@Aaybob
May 05 2017 03:05
@betluis because I've never heard of that being a solution befgore
Luis Betancourt
@betluis
May 05 2017 03:05
ok.. I will look at the documentation that Scott recommended me
Aaron Michelson
@Aaybob
May 05 2017 03:05
@Sgoldberg89 but when you use the other pump it wasn't and it has to tune up or something but the first several cycles it wont work is normal
that's five minute cycles
Sgoldberg89
@Sgoldberg89
May 05 2017 03:06
@Aaybob other pump?
Aaron Michelson
@Aaybob
May 05 2017 03:07
someone was saying a rig worked with one pump but not the other pump
Sgoldberg89
@Sgoldberg89
May 05 2017 03:08
too much?
Aaron Michelson
@Aaybob
May 05 2017 03:08
I had an easy time fixing two of my rigs, but i recognised the symptoms as a no network issue from having seen the error before. It takes a longg time to get started but it gets easier.
Sgoldberg89
@Sgoldberg89
May 05 2017 03:08
sorry, didn't realize the video was so explicit. but anyway, that wasn't me
Aaron Michelson
@Aaybob
May 05 2017 03:10
when you update make sure it updates
Sgoldberg89
@Sgoldberg89
May 05 2017 03:11

Starting pump-loop at Thu May 4 20:09:11 PDT 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Old pumphistory:
Error, retrying
Starting pump-loop at Thu May 4 20:10:19 PDT 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Old pumphistory:
Broadcast message from root@edison1 (Thu 2017-05-04 20:11:01 PDT):

Rebooting to fix radio errors!
The system is going down for reboot at Thu 2017-05-04 20:14:01 PDT!

Aaron Michelson
@Aaybob
May 05 2017 03:12
sounds good maybe it will fix radio errors
Luis Betancourt
@betluis
May 05 2017 03:12
@Aaybob my rig works with one pump and not with the other
Sgoldberg89
@Sgoldberg89
May 05 2017 03:12
It keeps doing it over and over
Aaron Michelson
@Aaybob
May 05 2017 03:12
@betluis did you let it run for a while on the new pump to retune itself
Luis Betancourt
@betluis
May 05 2017 03:13
yes
for hours
Aaron Michelson
@Aaybob
May 05 2017 03:13
is the correct pump serial number in pump.ini
Luis Betancourt
@betluis
May 05 2017 03:13
this rig-pump was working perfectly
well yes, it gets BG fine and updates NS but can't go on with temp basal calculations
Aaron Michelson
@Aaybob
May 05 2017 03:15
check serial number of pump
che4ck what you input for setup or look at pump.ini file
Luis Betancourt
@betluis
May 05 2017 03:18
ok, one sec
Aaron Michelson
@Aaybob
May 05 2017 03:22
@Sgoldberg89 don't give up, try using a different computer maybe the file is corrupted and isn't downloading the file but using one in cache. Keep trying, start with the easier things you can change. I am going to buy a new edison board to get offline use working. I think I've tried everything I can think of.
dmdfreak
@dmdfreak
May 05 2017 03:24
@danamlewis I honestly feel like I've found it but I'm not using it correctly. I'll keep looking and I won't ask again. I'm sorry
Aaron Michelson
@Aaybob
May 05 2017 03:24
Is there something not in the documentation to do for offline mode? Does the charging symbol still have to be there? Why would it not?
Luis Betancourt
@betluis
May 05 2017 03:26
@Aaybob checked, SN in pump.ini fine
Aaron Michelson
@Aaybob
May 05 2017 03:28
Well, @betluis buy a rig for that pump and have existing rig for the other pump and keep them seperated.
Luis Betancourt
@betluis
May 05 2017 03:29
the issue is that at this point I can't use this pump. I will try to set it up with a different rig
Aaron Michelson
@Aaybob
May 05 2017 03:29
@dmdfreak keep asking somebody else may have ideas
Luis Betancourt
@betluis
May 05 2017 03:29
@Aaybob thanks
@scottleibrand Scott, would you mind pointing me to the part of the documents you mentioned? i can't find it <brokenrecord>
Aaron Michelson
@Aaybob
May 05 2017 03:31
@betluis troubleshooting section comes right after "flash edison" section
or right after "setup" section
Dana Lewis
@danamlewis
May 05 2017 03:35
@dmdfreak this is a great place to ask those uestions
SMB requires command line - plus some preferences changes.
So if you did command line already, check out your preferences next
dmdfreak
@dmdfreak
May 05 2017 03:37
The preferences I got. It's the command line I haven't figured out for some reason. I feel like it's something simple that I'm just using incorrectly.
Scott Leibrand
@scottleibrand
May 05 2017 03:38
+1 to this being the right place to learn how this all works so you can test it safely
so let's start with this: what is it doing now, and what is it not doing that you'd like it to?
@betluis lol, sure. one moment.
Luis Betancourt
@betluis
May 05 2017 03:39
ok
Luis Betancourt
@betluis
May 05 2017 03:40
thanks!
Scott Leibrand
@scottleibrand
May 05 2017 03:41
@dmdfreak and more specifically, what are you seeing in the logs, vs. what would you expect to see if SMB were enabled and working?
dmdfreak
@dmdfreak
May 05 2017 03:43
I'm not the greatest at counting carbs. I generally under guess. SMB feels like it could help with that.
Scott Leibrand
@scottleibrand
May 05 2017 03:43
yep, that's exactly the kind of thing we designed it to help with. :)
dmdfreak
@dmdfreak
May 05 2017 03:47
I've already dropped from 7.2 to 6.2 after using the rig. And my 90 day numbers in xdrip say I'm at 5.7 now and getting better. Just trying to use all the tools. This looked exciting. ☺️
Scott Leibrand
@scottleibrand
May 05 2017 03:48
:-D
is now a good time for you to dive into the logs and code?
dmdfreak
@dmdfreak
May 05 2017 03:49
I'm in the code right now. I'll find it 😀
Scott Leibrand
@scottleibrand
May 05 2017 03:49
happy to help point you in the right direction. guidance is good, just no shortcuts.
dmdfreak
@dmdfreak
May 05 2017 03:51
I understand. I'm in the logs too btw ☺️
Thanks @scottleibrand @danamlewis
Martin Haeberli
@mhaeberli
May 05 2017 03:51
how do I get rid of Bluetooth: hci0 command 0x0c14 tx timeout ?
Dana Lewis
@danamlewis
May 05 2017 03:52
@dmdfreak :+1:
Scott Leibrand
@scottleibrand
May 05 2017 03:52
@dmdfreak if you haven't already, take a look at how oref0-pump-loop and oref0-setup work, too
dmdfreak
@dmdfreak
May 05 2017 03:53
Pump-loop I haven't looked at yet. I'll do that now
Martin Haeberli
@mhaeberli
May 05 2017 03:53
for those not in the know, “SMB” appears to stand for “super micro-boluses” - explanation here https://diyps.org/2017/04/30/introducing-oref1-and-super-microboluses-smb-and-what-it-means-compared-to-oref0-the-original-openaps-algorithm/
I wasn’t close to in the know until just now
Scott Leibrand
@scottleibrand
May 05 2017 03:54
yes. not to be confused with Super Mario Brothers. ;-p
dmdfreak
@dmdfreak
May 05 2017 03:54
LoL
Dana Lewis
@danamlewis
May 05 2017 03:55
Say it fast and often enough and it'll roll off your tongue like supercalifragilisticexpialidocious ;)
dmdfreak
@dmdfreak
May 05 2017 03:56
Did you type that or copy paste? LoL
Scott Leibrand
@scottleibrand
May 05 2017 03:57
At first I tried to call it something like superawesomemicrobolus to the tune of that song, but @danamlewis would have nothing of it.
Dana Lewis
@danamlewis
May 05 2017 03:57
And re SMB and other oref1 testing (per @mhaeberli point if you're new to it) - it's in dev and there's some notes in this PR with testing requests and suggestions: openaps/oref0#424
@scottleibrand ;p
@dmdfreak tab complete! (Kidding)
Scott Leibrand
@scottleibrand
May 05 2017 03:58
oh, that's what it was: superlowtempmicrobolus
props to the first person who writes out a full verse of oref1 lyrics to the tune of supercalifragilisticexpialidocious ;-)
Dana Lewis
@danamlewis
May 05 2017 04:00
Bigger props to everyone who actually helps us work through testing it ;)
Scott Leibrand
@scottleibrand
May 05 2017 04:00
heh yeah, and that. ;-)
kenkotch
@kenkotch
May 05 2017 04:03
seems my pump is not connecting- any ideas?: root@brains:~/KKloop# openaps use pump model
Traceback (most recent call last):
File "/usr/local/bin/openaps-use", line 63, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 51, in call
self.run(self.args)
File "/usr/local/bin/openaps-use", line 57, in run
output = app(args, self)
File "/usr/local/lib/python2.7/dist-packages/openaps/uses/init.py", line 92, in call
return self.method.selected(args)(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps/uses/init.py", line 31, in call
return self.method(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps/uses/use.py", line 44, in call
self.before_main(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps/vendors/medtronic.py", line 62, in before_main
self.setup_medtronic( )
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/mmeowlink.py", line 115, in setup_medtronic
setup_func(self)
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/mmeowlink.py", line 75, in setup_medtronic_link
link = builder.build(radio_type, port)
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/link_builder.py", line 16, in build
return SubgRfspyLink(port)
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/subg_rfspy_link.py", line 55, in init
self.open()
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
Scott Leibrand
@scottleibrand
May 05 2017 04:03
@kenkotch that usually just means you failed to stop the pump-loop run by cron, so the radio is still busy and your manual command can't run.
also, a pro-tip someone should PR into the docs somewhere useful: you can usually just paste the last couple lines of a traceback and that provides enough context
along with the command you ran, of course
the stuff in the middle is rarely useful
dmdfreak
@dmdfreak
May 05 2017 04:06
Ok a basic hint question. When I find said flag. It is used like this oref0-setup.sh --flag correct?
Scott Leibrand
@scottleibrand
May 05 2017 04:07
maybe. ;-)
when you run oref0-setup without any arguments, you get:
Usage: oref0-setup.sh <--dir=directory> <--serial=pump_serial_#> [--tty=/dev/ttySOMETHING] [--max_iob=0] [--ns-host=https://mynightscout.azurewebsites.net] [--api-secret=myplaintextsecret] [--cgm=(G4-upload|G4-local-only|shareble|G5|MDT|xdrip)] [--bleserial=SM123456] [--blemac=FE:DC:BA:98:76:54] [--btmac=AB:CD:EF:01:23:45] [--enable='autosens meal dexusb'] [--radio_locale=(WW|US)] [--ww_ti_usb_reset=(yes|no)]
the new flag will look like something in there: it just hasn't been added to the usage help yet
dmdfreak
@dmdfreak
May 05 2017 04:08
Ok gotcha
Scott Leibrand
@scottleibrand
May 05 2017 04:08
you might also look in the docs at how advanced features are usually enabled: it'll be similar to some of those.
oh, and when you figure it out, don't blurt out the answer and ruin it for the rest of the class. ;-)
dmdfreak
@dmdfreak
May 05 2017 04:10
I won't. I def realize now that this is a treasure hunt we all get to do for now. Except you 2 hehe
Sgoldberg89
@Sgoldberg89
May 05 2017 04:10
@scottleibrand what does this mean:
root@edison1:~# mkdir -p ~/src; cd ~/src && git clone git://github.com/openaps/oref0.git || (cd oref0 && git checkout master && git pull)
fatal: destination path 'oref0' already exists and is not an empty directory.
error: Your local changes to the following files would be overwritten by checkout:
bin/oref0_init_pump_comms.py
bin/oref0_subg_ww_radio_parameters.py
Please, commit your changes or stash them before you can switch branches.
Aborting
Scott Leibrand
@scottleibrand
May 05 2017 04:11
oh, it was a much longer treasure hunt to forge all the coins ourselves. ;-)
kenkotch
@kenkotch
May 05 2017 04:11

@scottleibrand thanks. ran sudo service cron stop. now the pump shows up. Forgive me-I'm new here...now what? running tail:

Starting pump-loop at Thu May 4 22:04:16 MDT 2017:
Radio ok, Listening: .No pump comms detected from other rigs
Profile less than 60m old
temp_basal.json less than 5m old
Pumphistory less than 5m old
Settings less than 10m old
Pumphistory-24 less than 20m old
Completed pump-loop at Thu May 4 22:04:52 MDT 2017

Am I reading correctly "No pump comms detected" means it is not talking to my pump?

dmdfreak
@dmdfreak
May 05 2017 04:11
For sure LoL
Scott Leibrand
@scottleibrand
May 05 2017 04:11
@Sgoldberg89 we need someone to add a note about that to the PR: you just need to do as it suggests and git stash from ~/src/oref0
Sgoldberg89
@Sgoldberg89
May 05 2017 04:11
@scottleibrand full disclosure, I uploaded the dev version 2 days ago, but it was working fun
*fine
Scott Leibrand
@scottleibrand
May 05 2017 04:12
this is an artifact of some permissions changes we made to those scripts (they were originally committed from Windows, which doesn't know about Linux permissions)
Sgoldberg89
@Sgoldberg89
May 05 2017 04:13
so if I were to do what you are suggesting... I would...
Scott Leibrand
@scottleibrand
May 05 2017 04:13
@kenkotch no, it means "No interfering pump comms detected from other rigs" - is that language clearer?
kenkotch
@kenkotch
May 05 2017 04:13
@scottleibrand yes- so good news.
Scott Leibrand
@scottleibrand
May 05 2017 04:14
exactly. and sounds like we have +1 vote for @danamlewis' idea for revised language to put into mmeowlink for that
kenkotch
@kenkotch
May 05 2017 04:15
@scottleibrand do I need to turn cron back on? last papertrail: May 04 22:04:54 localhost Pump: 439116 is already responding. Not sending wakeup messages
Scott Leibrand
@scottleibrand
May 05 2017 04:15
@kenkotch looks like that particular loop run had nothing to do. if you look back further (or wait a bit after re-enabling cron), you should see one that did
kenkotch
@kenkotch
May 05 2017 04:16
@scottleibrand ill re-enable and wait. maybe this is a bad time to be typing. my bg is 62 so insulin should be stopped...I'll treat first
Scott Leibrand
@scottleibrand
May 05 2017 04:17
yeah, take care of that and we can revisit in a few minutes
Martin Haeberli
@mhaeberli
May 05 2017 04:21
how to figure out WHICH git repo I deployed Heroku Nightscout from?
dmdfreak
@dmdfreak
May 05 2017 04:25
I think I have it @scottleibrand
Scott Leibrand
@scottleibrand
May 05 2017 04:29
:-) :+1:
dmdfreak
@dmdfreak
May 05 2017 04:29
Is there a way to know for sure after I run it?
Scott Leibrand
@scottleibrand
May 05 2017 04:30
after you run oref0-setup, you should be able to see the difference pretty quickly in pump-loop.log
dmdfreak
@dmdfreak
May 05 2017 04:30
Ok. Fingers crossed
Scott Leibrand
@scottleibrand
May 05 2017 04:30
Starting supermicrobolus pump-loop at Thu 4 May 21:29:02 MST 2017 with 4 second wait_for_silence:
that kind of thing
dmdfreak
@dmdfreak
May 05 2017 04:39
Yep got it. Sad thing is I had it all day. I just didn't know how to implement it in the command line LoL
Scott Leibrand
@scottleibrand
May 05 2017 04:40
heh. I bet you learned a fair bit of other stuff in the process of figuring that out, too
dmdfreak
@dmdfreak
May 05 2017 04:41
100% correct!
Scott Leibrand
@scottleibrand
May 05 2017 04:41
looking forward to hearing how it works for you.
dmdfreak
@dmdfreak
May 05 2017 04:41
It's a lot more fun after you figure it out than before though LoL
Scott Leibrand
@scottleibrand
May 05 2017 04:42
heh no doubt. that's also especially true when the puzzle you're trying to solve is one where you don't know if there's a solution or not.
dmdfreak
@dmdfreak
May 05 2017 04:42
Def let you know. You guys and gals are the best.
Scott Leibrand
@scottleibrand
May 05 2017 04:42
this has been a really rewarding project to work on for so many reasons
:+1: :smile: thx
dmdfreak
@dmdfreak
May 05 2017 04:44
Man. There's nothing greater than telling someone what the weird little box is with the led lights.
Scott Leibrand
@scottleibrand
May 05 2017 04:45
hehe :-D
dmdfreak
@dmdfreak
May 05 2017 04:47
So many ask. Before finding openaps i thought I was getting by ok. Now I realize just how bad I was doing. You guys are changing the way people think. Not many get to do that. It's amazing.
Scott Leibrand
@scottleibrand
May 05 2017 04:48
I just hope we can make a dent in the way things work for the general public, not just the really engaged subset that's willing and able to build their own DIY closed loop
dmdfreak
@dmdfreak
May 05 2017 04:50
I agree. Having the solution is one thing. Making people realize you have the solution is a whole other thing. But I think you're doing it.
Thanks again for making me earn it. Think I'll go sleep now LoL
Scott Leibrand
@scottleibrand
May 05 2017 04:51
heh. you're welcome. have a good non-beepy night.
kenkotch
@kenkotch
May 05 2017 04:52
ok. I'm done being low. Just realized that on my nitescout the 'pump' line is red. I think that's not good. Still getting: May 04 22:44:55 brains pump-loop.log: {"reason":"BG data is too old (it's probably this), or clock set incorrectly. The last BG data was read at Thu May 04 2017 17:18:21 GMT-0600 (MDT) but your system time currently is Thu May 04 2017 22:44:54 GMT-0600 (MDT)"}
Screenshot 2017-05-04 22.47.58.png
Scott Leibrand
@scottleibrand
May 05 2017 04:53
does NS have up to date CGM data?
and does the rig? it lives in monitor/glucose.json
kenkotch
@kenkotch
May 05 2017 04:56
hmm. rig has gcm data from 5 hrs ago
Scott Leibrand
@scottleibrand
May 05 2017 04:58
ok. can the rig connect to the Internet?
and are you getting BG data from NS or from a connected CGM?
kenkotch
@kenkotch
May 05 2017 04:59
confirming rig connecting to internet. getting gb from NS
Scott Leibrand
@scottleibrand
May 05 2017 04:59
k, you might (or might not) find something in the ns-loop.log
otherwise, you might be best off scrolling way up to this afternoon's conversation for when some other folks were troubleshooting BG data not being updated from NS
kenkotch
@kenkotch
May 05 2017 05:01
Thank you @scottleibrand I'll take a look
Scott Leibrand
@scottleibrand
May 05 2017 05:03
good luck and g'nite. there are always lots of other folks online who can help out, and quite a few of them have been through the same troubleshooting recently.
Martin Haeberli
@mhaeberli
May 05 2017 05:18
@scottleibrand goodnight; thanks - but fyi, ns-loop updates cgm/ns-glucose.json to (just now) 2017-05-04T22:06:57-07:00; curl api still works and gets most recent number, but pump-loop logs error The last BG data was read at Thu May 04 2017 20:11:57 GMT-0700 (PDT)
Scott Leibrand
@scottleibrand
May 05 2017 05:21
is monitor/glucose.json getting updated? might want to investigate why not
or where else it might be getting updated from: you might have stale xdrip or direct microUSB CGM data that is getting copied over when it's old
Martin Haeberli
@mhaeberli
May 05 2017 05:22
yes - monitor/glucose.json is updated .. - no xdrip nor microUSB cable attached / configured / etc
Scott Leibrand
@scottleibrand
May 05 2017 05:23
ok, then this might be something to do with ~/myopenaps-cgm-loop vs. ~/myopenaps
do you have the cgm-loop directory? I don't recall which CGM setup you said you have
Martin Haeberli
@mhaeberli
May 05 2017 05:24
no cgm-loop directory
Scott Leibrand
@scottleibrand
May 05 2017 05:25
not sure then. time to dive deeper into what exactly the aliases are doing, and which data files have which data when
you might have to pause the loop and step through the whole process manually
I've done this before, it's kinda tedious and difficult to keep straight all the different pieces
Martin Haeberli
@mhaeberli
May 05 2017 05:25
i’m not grocking ~/myopenaps-cgm-loop vs. ~/myopenaps
Scott Leibrand
@scottleibrand
May 05 2017 05:25
nevermind cgm-loop: you don't have that setup
that's for g4-raw on master / g4-upload on dev
Martin Haeberli
@mhaeberli
May 05 2017 05:26
fyi - I looked (grep -l pattern -r) for the purported last loaded time (2 + hours ago) and the head of each json was quite recent
grep -l 'T20\:11\:' -r .
./cgm/ns-glucose.json
./cgm/glucose.json
./monitor/glucose.json
Scott Leibrand
@scottleibrand
May 05 2017 05:27
pump-loop is just looking at monitor/glucose.json
Martin Haeberli
@mhaeberli
May 05 2017 05:27
cgm is a G4, but data flows through NS; curl api fetch shows current bg; ns-loop brings current bg
ok - l shows pump-loop is somehow not running since 30 + minutes ago
i did run it by hand in between
so maybe I’ll give up and reboot the Edison hoping it all converges, but no root cause identified / solved
dmdfreak
@dmdfreak
May 05 2017 05:30
So I solve that and the rig stops communicating with the pump again.... Almost got to go to sleep hehe
Martin Haeberli
@mhaeberli
May 05 2017 05:31
l
Sgoldberg89
@Sgoldberg89
May 05 2017 05:34
can I get a tip on how to git stash from ~/src/oref0 from anyone?
Scott Leibrand
@scottleibrand
May 05 2017 05:34
just cd ~/src/oref0 and then git stash
Sgoldberg89
@Sgoldberg89
May 05 2017 05:35
@scottleibrand Shabang! Thank you
@scottleibrand I noticed some new stuff in the crontab, wondering if you ended up incorporating parts of "Thor's Hammer" into the new system
Scott Leibrand
@scottleibrand
May 05 2017 05:37
heh, which thing was "Thor's Hammer"?
Sgoldberg89
@Sgoldberg89
May 05 2017 05:37
The auto reformat thing
Scott Leibrand
@scottleibrand
May 05 2017 05:37
besides the ones he was using to fight with Loki in the sky this afternoon, of course
we have an auto reboot now on radio errors...
and oref0-reset-git is a bit more aggressive...
Sgoldberg89
@Sgoldberg89
May 05 2017 05:38
cool
what does git stash do?
Scott Leibrand
@scottleibrand
May 05 2017 05:40
it saves "your" uncommitted changes to a clipboard type thing
in this case you don't want them, so you never retrieve them from there
Sgoldberg89
@Sgoldberg89
May 05 2017 05:41
so was that added to help prevent loss of info?
Scott Leibrand
@scottleibrand
May 05 2017 05:41
it's something developers use when they're in the middle of working on something and want to pull in new code without committing what they're working on
in this case the need to do that is getting triggered by some weird file permissions thing that I don't fully understand, but know how to fix
(cue up the git xkcd)
Sgoldberg89
@Sgoldberg89
May 05 2017 05:43
One reason I love talking to you, you know I don't understand... almost everything you say yet you say it anyway. And I THINK it's because you want to believe I'm capable of understanding someday... :)
in other news, reruning the setup script didn't fix my problem
running*
Scott Leibrand
@scottleibrand
May 05 2017 05:44
you're perfectly capable: it's just a question of how much googling you're willing to do. ;-)
dmdfreak
@dmdfreak
May 05 2017 05:44
Nod and wave ✋
Scott Leibrand
@scottleibrand
May 05 2017 05:45
I'm probably not up for any more troubleshooting tonight, so I'll leave y'all to it.
Martin Haeberli
@mhaeberli
May 05 2017 05:45
@scottleibrand suppose I want to configure openaps open loop - monitor, autotune, alert, etc.
Sgoldberg89
@Sgoldberg89
May 05 2017 05:45
You know, it's harder than that. I don't even know WHAT to google...
Martin Haeberli
@mhaeberli
May 05 2017 05:45
what to change in oref0-setup
ok night time...
Scott Leibrand
@scottleibrand
May 05 2017 05:45
@mhaeberli I'm working on some pushover stuff in a branch...
Martin Haeberli
@mhaeberli
May 05 2017 05:45
k
Scott Leibrand
@scottleibrand
May 05 2017 05:45
without that you won't get notified of what it's trying to do
Martin Haeberli
@mhaeberli
May 05 2017 05:45
ok, but I can look at logs...
Scott Leibrand
@scottleibrand
May 05 2017 05:45
but you could just cripple the enact aliases if you want...
Martin Haeberli
@mhaeberli
May 05 2017 05:46
just wanted to monitor and gather autotune data without closing tonight
Scott Leibrand
@scottleibrand
May 05 2017 05:46
that wouldn't be an oref0-setup thing, but rather something you'd do manually afterward
Martin Haeberli
@mhaeberli
May 05 2017 05:46
ok ...
Scott Leibrand
@scottleibrand
May 05 2017 05:46
oh, in that case...
Martin Haeberli
@mhaeberli
May 05 2017 05:46
so post hoc delete or edit something
Scott Leibrand
@scottleibrand
May 05 2017 05:46
if all you want running is autotune, you can enable cron and then just comment out the pump-loop one
Martin Haeberli
@mhaeberli
May 05 2017 05:47
got it -edit crontab
Scott Leibrand
@scottleibrand
May 05 2017 05:47
ya
Sgoldberg89
@Sgoldberg89
May 05 2017 05:47
Here's some gobbledygook and maybe you can tell me what's wrong?
mmtune: monitor/mmtune.json raised Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?
pump://JSON/mmtune/monitor/mmtune.json
Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 82, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 51, in call
self.run(self.args)
File "/usr/local/bin/openaps-report", line 75, in run
output = app(args, self)
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/subcommand.py", line 52, in call
return self.method.main(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps/reports/invoke.py", line 40, in main
output = task.method(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps/uses/use.py", line 44, in call
self.before_main(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps/vendors/medtronic.py", line 62, in before_main
self.setup_medtronic( )
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/mmeowlink.py", line 92, in setup_medtronic
setup_medtronic_link(self)
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/mmeowlink.py", line 75, in setup_medtronic_link
link = builder.build(radio_type, port)
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/link_builder.py", line 16, in build
return SubgRfspyLink(port)
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/subg_rfspy_link.py", line 55, in init
self.open()
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/serial_interface.py", line 28, in open
self.check_setup()
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/subg_rfspy_link.py", line 72, in check_setup
self.serial_rf_spy.sync()
File "/usr/local/lib/python2.7/dist-packages/mmeowlink/vendors/serial_rf_spy.py", line 121, in sync
raise CommsException("Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?")
mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?
Usage: grep [OPTION]... PATTERN [FILE]...
Try 'grep --help' for more information.
Scott Leibrand
@scottleibrand
May 05 2017 05:48
see discussion earlier today about what to do about the "Could not get subg_rfspy state or version" error
Sgoldberg89
@Sgoldberg89
May 05 2017 05:48
k
Sgoldberg89
@Sgoldberg89
May 05 2017 05:55
@garykidd I think I am trying to work off your instruction in the docs for the could not get subg_rfspy and am having trouble with the following two lines:
ccprog -p 19,7,36 erase
ccprog -p 19,7,36 write spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
root@edison1:~# ccprog -p 19,7,36 erase
-bash: ccprog: command not found
root@edison1:~# ccprog -p 19,7,36 write spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
-bash: ccprog: command not found
Wondering if there's something else I need to do, would appreciate your help and I
I'll make the PR once I figure it out!
Martin Haeberli
@mhaeberli
May 05 2017 06:03
@Sgoldberg89 look at gitter earlier today
you have to : a) cd ~/src
b) git clone ccprog (i’ll send the url in a minute)
c) cd to ccprog
d) make ccprog
e) cd ~/
f) wget spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex (link in a minute)
g) ccprog/ccprog -p 19,7,36 erase
h) ccprog/ccprog -p 19,7,36 write spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
this channel
git clone https://github.com/ps2/ccprog.git
(after cd ~/src)
cd ccprog
make ccprog
make ccprog
wget https://github.com/EnhancedRadioDevices/subg_rfspy/releases/download/v0.8-explorer/spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
(from within ccprog directory)
then your original
ccprog -p 19,7,36 erase
ccprog -p 19,7,36 write spi1_alt2_EDISON_EXPLORER_US_STDLOC.hex
sorry for the confusing instructions
one or two of us owe troubleshooting a PR
Dana Lewis
@danamlewis
May 05 2017 06:14
:+1:
Chuck Vann
@djtechman
May 05 2017 06:16
how can I wipe out the sqlite database that is used for xdripaps?
even after re-running oref0 using a different target dir, and confirming that xdripaps is getting bg's from xdrip and uploading them to NS.... I still have an error about BG data being too old (from April 7th)
Dana Lewis
@danamlewis
May 05 2017 06:21
@djtechman is your old instance running somehow and thus it's not expecting xdrip for bgs?
Chuck Vann
@djtechman
May 05 2017 06:26
no I removed all the old cron & rebooted...
and I made sure the new run of oref was using a fresh target dir... but this new instance is still hanging onto the last BG from back in april
(the error about April 7th old bg shows up in the openaps logs even though xdrip is only configured to use REST API upload to the bt-pan URL of the aps rig... these new BGs show up in the aps log, and Nightscout is receiving these BGs...)
its so weird
tepidjuice
@tepidjuice
May 05 2017 06:40
@djtechman If you're not on dev you could try using that. I think it has improved handling of xdrip data.
francesc0-cgm
@francesc0-cgm
May 05 2017 06:59
Reading about remainingcarbscap but it is still unclear
andyhandy
@andyhandy
May 05 2017 08:56
Hi All. For some reason my render basal and treatments (e.g. meal boluses) have suddenly stopped displaying anything in nightscout. Openaps is looping perfectly. Any ideas? Thanks
tepidjuice
@tepidjuice
May 05 2017 09:40
@andyhandy Maybe you have future treatments in nightscout?
under settings -> admin tools there should be a button that you can use to delete future entries if you have them
francesc0-cgm
@francesc0-cgm
May 05 2017 10:54
Which is the enable flag to add to oref0 script to enable smb?
Microbolus just seen
elodaille01
@elodaille01
May 05 2017 12:32
Hi..what could justify the error message : 'pump data stale' on NS please ? All was ok..then it stops looping with that message..(battery ok, all ok, perfectly looping just before, 45 minutes ago)..Can't check the logs now because my daughter is at school..Thank you very, very much for your help..
elodaille01
@elodaille01
May 05 2017 12:58
(it's ok again after almost one hour..but ..I would like to have an explanation..or some similar feedback ? thank you +++)
katie disimone
@Kdisimone
May 05 2017 13:59
@elodaille01 that message is the result of the rig not being able to get a fresh read of the pump history. When it can't read the history, the rig can't loop and the pump data goes stale. Perhaps her rig got too far away from her (most likely). Or perhaps there was interference. Sometimes a pump history fetch just fails for a while.
PieterGit
@PieterGit
May 05 2017 14:00
i'm seeeing "reservoir":null in my pump-loop.log. but monitor/reservoir.json and nigthscout reservoir seems fine. is this something to worry about?
katie disimone
@Kdisimone
May 05 2017 14:01
@andyhandy were you recently doing anything in your site settings. Could you have possibly disabled or removed careportal?
PieterGit
@PieterGit
May 05 2017 14:03
@elodaille01 : you should check the pump loop if it's a pump communication problem and the upload.log to see if there is something wrong with the uploading of the pump deatils
elodaille01
@elodaille01
May 05 2017 14:05
ok @Kdisimone and @PieterGit thank you
PieterGit
@PieterGit
May 05 2017 14:20
shouldn't maxCOB in preferences.json be renamed to max_cob to make it consistent with max_iob
andyhandy
@andyhandy
May 05 2017 14:33
@tepidjuice i just checked and the database doesn't contain any future entries
francesc0-cgm
@francesc0-cgm
May 05 2017 14:34
Using smb...great stuff! I have only a question when my son goes to target after having been high and there are cob yet it doesn't start again to loop with temping but it continues with superbolus and zero temping is it normal? Using smb use only microboluses to loop and never basals? Another question: i don't undertand the three settings in preferences.json about smb. Last question: how does remainimgcarbscap work? Tia
andyhandy
@andyhandy
May 05 2017 14:34
@Kdisimone i'm pretty sure i didn't change anything. the careportal plugin is still there and activated
francesc0-cgm
@francesc0-cgm
May 05 2017 14:34
It doesn't affect site and line
Sorry wrong post...smb does't affect site and line with all that zero temping?
PieterGit
@PieterGit
May 05 2017 14:45
i am on a smb test rig now, smb enabled in preferences.json but i didn't see an extra bolus yet. i can recommend apple pi :cake: for testing smb :smile:
kenkotch
@kenkotch
May 05 2017 14:51

@scottleibrand @ruess @mhaeberli I'm still not able to get my rig to connect to the net. Tried different wifi's, restarts, ifup, ifdown, last bg (in glucose.json) is from yesterday morning.

Starting oref0-online.
At Fri May 5 08:44:01 MDT 2017 my local IP is: 10.0.0.173

At Fri May 5 08:44:01 MDT 2017, my wifi network name is OurHome, and my public IP is:
Error, cycling networking
Stopping networking (via systemctl): networking.service.
Starting networking (via systemctl): networking.service.
and getting new wlan0 IP

At Fri May 5 08:44:22 MDT 2017, my wifi network name is OurHome, and my public IP is:

At Fri May 5 08:44:23 MDT 2017, my wifi network name is OurHome, and my public IP is: 2601:0281:8000:0f01:92b6:86ff:fe12:c1c6
Finished oref0-online.

and:
root@brains:/# ping -c 4 8.8.8.8
PING 8.8.8.8 (8.8.8.8) 56(84) bytes of data.
64 bytes from 8.8.8.8: icmp_seq=1 ttl=57 time=18.8 ms
64 bytes from 8.8.8.8: icmp_seq=2 ttl=57 time=37.1 ms
64 bytes from 8.8.8.8: icmp_seq=3 ttl=57 time=35.4 ms
64 bytes from 8.8.8.8: icmp_seq=4 ttl=57 time=48.3 ms

--- 8.8.8.8 ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3005ms
rtt min/avg/max/mdev = 18.849/34.930/48.312/10.523 ms

any ideas?

thanks gitter for changing ~ # into an emoji
Martin Haeberli
@mhaeberli
May 05 2017 14:54
@kenkotch I hate to suggest this. Try rebooting the rig. Obviously you have connectivity. I have not yet figured out my root cause either.
Reboot didn't solidly fix it but helped at least a while
PieterGit
@PieterGit
May 05 2017 14:55
@kenkotch can you post the output of route -n and ifconfig wlan0
kenkotch
@kenkotch
May 05 2017 14:56
@mhaeberli rebooted a bunch of times. @PieterGit yep...

route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
0.0.0.0 10.0.0.1 0.0.0.0 UG 0 0 0 wlan0
10.0.0.0 0.0.0.0 255.255.255.0 U 0 0 0 wlan0
192.168.2.0 0.0.0.0 255.255.255.0 U 0 0 0 usb0

ifconfig wlan0
wlan0 Link encap:Ethernet HWaddr 90:b6:86:12:c1:c6
inet addr:10.0.0.173 Bcast:10.0.0.255 Mask:255.255.255.0
inet6 addr: 2601:281:8000:f01:92b6:86ff:fe12:c1c6/64 Scope:Global
inet6 addr: fe80::92b6:86ff:fe12:c1c6/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:7572 errors:0 dropped:0 overruns:0 frame:0
TX packets:18902 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:1300424 (1.2 MiB) TX bytes:3933265 (3.7 MiB)

PieterGit
@PieterGit
May 05 2017 15:01
@kenkotch you seem to have a ipv6 only address. try curl --compressed -4 -s -m 15 icanhazip.com. I don't think oref0-online is prepared for that
kenkotch
@kenkotch
May 05 2017 15:02
I have no idea what ipv6 means:
root@brains:~# curl --compressed -4 -s -m 15 icanhazip.com
67.190.87.57
PieterGit
@PieterGit
May 05 2017 15:02
ok, that's a ipv4 address
@kenkotch there are two types of ip addresses. ipv4 address use 4 numbers from 0-255, ipv6 adressess are much longer
kenkotch
@kenkotch
May 05 2017 15:04
thanks. google just told me. should I have ipv6 instead, additionally?
PieterGit
@PieterGit
May 05 2017 15:12
no, the oref0-online script is ipv4 based. theoretically you could have a ipv6 adress only. that doesn't work nice with oref0-online (yet). but that's not relevant for your problem
so you can connect to the internet, DNS nameresolving works, you have a ipv4 adress. Can you tell me more about what connectivity problem you have?
dmdfreak
@dmdfreak
May 05 2017 15:16
Does Nightscout need to be updated to see the extra prediction line and the SMB boluses? It will those just work. I'm seeing SMB action in the logs but not on my azure site.
kenkotch
@kenkotch
May 05 2017 15:18
thanks. monitor/glucose.json is not getting new data and on NS the 'pump' field is red. I am all of the sudden getting openAPS prediction line
Screenshot 2017-05-05 09.16.34.png
dmdfreak
@dmdfreak
May 05 2017 15:18
I'm on Nightscout 0.9.2
kenkotch
@kenkotch
May 05 2017 15:21
@dmdfreak 0.9.2 is the latest version. My predict line comes in and out- though that may be related to the other issue im having
Jason Wittmer
@jbwittmer
May 05 2017 15:33

Looking for some help trying to get raw data to display on Nightscout. Using g4-upload branch from dev. Nightscout 0.9.2 with raw "always" enabled. Have checked MongoDB entries (including one from both Share and from the CGM attached to my Pi rig) and will paste following. Am I correct that the "unfiltered" entry represents the raw data (and therefore should be available for Nightscout to display) or am I incorrect about the meaning of the "filtered" and "unfiltered" entries and it's not actually uploading raw data. Thanks in advance for the help. {
"_id": {
"$oid": "590be2fc4b9b4c5a006e77bb"
},
"trend_arrow": "FLAT",
"display_time": "2017-05-04T21:24:37-05:00",
"system_time": "2017-01-22T20:17:31-06:00",
"direction": "Flat",
"sgv": 105,
"dateString": "2017-05-04T21:24:37-05:00",
"device": "openaps://Raspi3/cgm",
"unfiltered": 125792,
"rssi": 185,
"date": 1493951077000,
"filtered": 122608,
"type": "sgv",
"glucose": 105
}

{
"_id": {
"$oid": "590be34a4b9b4c5a006e77c2"
},
"sgv": 100,
"date": 1493951166000,
"dateString": "2017-05-05T02:26:06.000Z",
"trend": 4,
"direction": "Flat",
"device": "share2",
"type": "sgv"
}

Scott Leibrand
@scottleibrand
May 05 2017 15:45
The only thing NS needs updating for is the UAM prediction line. See the PR for the link to the NS branch for that if you're testing UAM already. Microboluses will show up as small boluses with an unmodified NS.
dmdfreak
@dmdfreak
May 05 2017 15:51
Ok. Time to learn again LoL
Scott Leibrand
@scottleibrand
May 05 2017 15:53
@francesc0-cgm as long as SMB mode is still active and there's enough deviations from carb activity that naive_eventualBG (excluding deviations) is well below target, it will keep the zero temp running and do microboluses as needed. Depending on the enable flags in preferences, SMB mode will be deactivated after DIA hours from the last manual bolus, after COB completely decays, and/or after temp targets end. Also, if IOB decays and then BG rises, such that eventualBG goes above target, it will set a high temp while still SMBing, and then set another zero temp once it gets enough IOB going.
The remainingCarbsCap limits the extent to which oref1 "trusts" that entered carbs will be absorbed at some point in the next 4h, even if no absorption deviations are detected right now. If you sometimes enter carbs and don't eat them, you'll want to leave that at zero. If you want SMB to start bolusing for carbs as soon as you enter them, you can raise remainingCarbsCap to an appropriate limit for your situation.
@PieterGit consistency is not my strong suit. :smile: PRs welcome if it's a priority for you.
I see reservoir null often when it's not actively SMBing: that isn't a problem.
Did I miss anything else overnight? 🙂
Jason Wittmer
@jbwittmer
May 05 2017 15:59
@scottleibrand sorry to bother, could you peak at that question about filtered/unfiltered CGM data and point me in the right direction? Trying to figure out if it's a problem with my rig uploading raw data or the nightscout page not taking the data and displaying it properly. Thanks.
kenkotch
@kenkotch
May 05 2017 16:01
morning @scottleibrand nothing new here. my rig is still not connecting to the net (I think). I've tried everything I can think of and everything others have suggested (above)
should I just cut my losses and reflash?
Scott Leibrand
@scottleibrand
May 05 2017 16:01
@jbwittmer I missed that one. Can you repeat?
Jason Wittmer
@jbwittmer
May 05 2017 16:02

Looking for some help trying to get raw data to display on Nightscout. Using g4-upload branch from dev. Nightscout 0.9.2 with raw "always" enabled. Have checked MongoDB entries (including one from both Share and from the CGM attached to my Pi rig) and will paste following. Am I correct that the "unfiltered" entry represents the raw data (and therefore should be available for Nightscout to display) or am I incorrect about the meaning of the "filtered" and "unfiltered" entries and it's not actually uploading raw data. Thanks in advance for the help. {
"_id": {
"$oid": "590be2fc4b9b4c5a006e77bb"
},
"trend_arrow": "FLAT",
"display_time": "2017-05-04T21:24:37-05:00",
"system_time": "2017-01-22T20:17:31-06:00",
"direction": "Flat",
"sgv": 105,
"dateString": "2017-05-04T21:24:37-05:00",
"device": "openaps://Raspi3/cgm",
"unfiltered": 125792,
"rssi": 185,
"date": 1493951077000,
"filtered": 122608,
"type": "sgv",
"glucose": 105
}

{
"_id": {
"$oid": "590be34a4b9b4c5a006e77c2"
},
"sgv": 100,
"date": 1493951166000,
"dateString": "2017-05-05T02:26:06.000Z",
"trend": 4,
"direction": "Flat",
"device": "share2",
"type": "sgv"
}

Scott Leibrand
@scottleibrand
May 05 2017 16:02
@kenkotch you can do a lot of troubleshooting with ping and traceroute and similar if you want...
Yes, unfiltered is used to calculate raw. But that has to happen on the rig, not in NS. Looks like yours isn't yet doing that. Do you have valid calibration records?
Jason Wittmer
@jbwittmer
May 05 2017 16:05
No I wasn't aware of how to get calibration records onto the rig. Do I need to setup a separate cron job or can you point me toward where to RTD? Thanks.
kenkotch
@kenkotch
May 05 2017 16:05
thanks. I'll go learn...
Scott Leibrand
@scottleibrand
May 05 2017 16:07
@jbwittmer it should happen automatically. Check the data files in the ~/myopenaps-cgm-loop directory.
And if you want, try the manual drill-down troubleshooting process there.
Jason Wittmer
@jbwittmer
May 05 2017 16:08
OK. I'll do that later. At work now and rig at home. I may need some additional help but am trying to figure it out myself. At least I know what I'm missing and a little more about how the process works. Thanks.
Scott Leibrand
@scottleibrand
May 05 2017 16:08
It's a separate openaps instance from ~/myopenaps
kenkotch
@kenkotch
May 05 2017 16:27

sorry to bother you again @scottleibrand I'm working on getting a handle on traceroute. Tried

root@brains:~# traceroute kknitescout.herokuapp.com
traceroute to kknitescout.herokuapp.com (54.225.175.205), 30 hops max, 60 byte packets
1 10.0.0.1 (10.0.0.1) 2.987 ms 5.961 ms 5.558 ms
2 96.120.12.253 (96.120.12.253) 24.177 ms 26.654 ms 27.763 ms
3 ae-201-sur02.boulder.co.denver.comcast.net (68.86.105.13) 26.682 ms 26.006 ms 26.085 ms
4 ae-10-sur03.boulder.co.denver.comcast.net (162.151.51.33) 25.882 ms 28.985 ms 28.826 ms
5 ae-29-ar01.denver.co.denver.comcast.net (162.151.50.249) 33.010 ms 31.791 ms 31.786 ms
6 be-33652-cr01.1601milehigh.co.ibone.comcast.net (68.86.92.121) 33.749 ms 37.297 ms 39.917 ms
7 be-11719-cr02.denver.co.ibone.comcast.net (68.86.86.77) 39.683 ms 19.927 ms 21.956 ms
8 be-11724-cr02.dallas.tx.ibone.comcast.net (68.86.84.230) 40.526 ms 40.350 ms 40.166 ms
9 be-12495-pe03.1950stemmons.tx.ibone.comcast.net (68.86.85.194) 39.973 ms 40.653 ms 42.731 ms
10 66.208.233.34 (66.208.233.34) 42.984 ms 43.479 ms 66.208.232.158 (66.208.232.158) 42.117 ms
11
12

13
14
* 54.239.104.151 (54.239.104.151) 76.128 ms
connect: Network is unreachable

ping returns:
root@brains:~# ping kknitescout.herokuapp.com
PING us-east-1-a.route.herokuapp.com (54.235.102.46) 56(84) bytes of data.
ping: sendmsg: Network is unreachable
ping: sendmsg: Network is unreachable
ping: sendmsg: Network is unreachable
ping: sendmsg: Network is unreachable
ping: sendmsg: Network is unreachable
ping: sendmsg: Network is unreachable
^C
--- us-east-1-a.route.herokuapp.com ping statistics ---
56 packets transmitted, 0 received, 100% packet loss, time 55017ms

seems to work fine when I traceroute or ping to google

dmdfreak
@dmdfreak
May 05 2017 16:28
@scottleibrand Is there anything to do after I click fork in GitHub? When I went to azure and updated, it didn't look like it was the right one. It says #2376
PieterGit
@PieterGit
May 05 2017 16:36
why doesn't openaps increase the temp basal in this case?
image.png
Martin Haeberli
@mhaeberli
May 05 2017 16:56
@kenkotch I have sometimes had trouble pinging heroku when I can still ping google, for example. But even then a curl -v against an nightscout api works
katie disimone
@Kdisimone
May 05 2017 16:57
@dmdfreak it's been awhile since I was on azure, but there's a deployment area. You need to stop that deploy and redeploy but choose the floating-carbs branch of cgm-remote-monitor.
dmdfreak
@dmdfreak
May 05 2017 16:58
@
dmdfreak
@dmdfreak
May 05 2017 17:06
@Kdisimone well I did all that twice but I still cant tell the difference. I had purple lines before and I still have purple lines. LOL
katie disimone
@Kdisimone
May 05 2017 17:08
I am new to the UAM lines as well...but I would recommend reading over Dana's post about oref1. There was some info in there about when those new lines would appear vs not appear.
dmdfreak
@dmdfreak
May 05 2017 17:09
i can do that. thanks!
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 17:14
I have been having issues communicating with NTP servers when my rig is connecting to my phone's wifi, my work's wifi and, my phones BT. I am getting internet on my rig and have tried loads of things...
you can see all troubleshooting steps by looking here https://goo.gl/3odk3S
any further assistance would be appreciated
francesc0-cgm
@francesc0-cgm
May 05 2017 17:19
Thank you @scottleibrand for your answers. Did you meet any failure on set or line with prolonged zero temping of smb?
Dana Lewis
@danamlewis
May 05 2017 17:23
@francesc0-cgm no, haven't had any set failures
fjungman
@fjungman
May 05 2017 17:25
@diabetic_dude_twitter what is the error code when it fails? Enter systemctl status systemd-timesyncd.service
garykidd
@garykidd
May 05 2017 17:31
one thing that is pretty neat about using MDT sensor....you're rig can offline and you remain looping....no receiver...no BT connectivity.....can see you're BGs without a phone....too bad the accuracy isnt there---but still working OK for looping.
Scott Leibrand
@scottleibrand
May 05 2017 17:33
@PieterGit if the eventualBG is above target, but you're falling faster than required to eventually get you below target, we go into wait-and-see mode, to determine whether eventualBG ends up dropping below target from continued negative deviations, or whether the BG flattens out and we need to high temp.
@francesc0-cgm SMB is no more prone to extended periods of zero insulin delivery than AMA was. I haven't seen any evidence of that extended zero temping causing infusion set or cannula blockage in anyone using either system. there was some concern it might, but we never ended up implementing any of the proposed workarounds to avoid the situation, and it hasn't really been a problem.
@dmdfreak you have to be using the floating-carbs branch of NS to get UAM prediction lines displayed, until someone merges that PR to NS dev.
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 17:38
@fjungman Unit systemd-timesyncd.service entered failed state.
no code
fjungman
@fjungman
May 05 2017 17:38
@scottleibrand @PieterGit I really like that feature. When the sensor starts getting "noisy" near the end of its life, that feature tends to err on the high-side, due to all of the rapid up and down movements. Without that "wait and see" code, it would probably go low as much as it goes high. I see it as a safety that really works. I was watching it last night, as my kid slept, and it stayed higher than I would have liked, but it was better than the alternative...
@diabetic_dude_twitter enter systemctl status systemd-modules-load and get the PID for the failed module.
Then, journalctl _PID=xxxx where xxxx is the PID
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 17:47
@fjungman 106
dmdfreak
@dmdfreak
May 05 2017 17:47
@scottleibrand I called myself doing that, but I'm not sure I accomplished it. Is there a way to tell? I was already getting the purple lines. My choices are AR2 and OpenAPS. Does it add one to that list?
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 17:50
root@in10aps:~# journalctl _PID=106
-- Logs begin at Fri 1999-12-31 17:00:26 MST, end at Fri 1999-12-31 17:03:38 MST
Dec 31 17:00:27 in10aps systemd-modules-load[106]: Inserted module 'bcm_bt_lpm'
Dec 31 17:00:27 in10aps systemd-modules-load[106]: Inserted module 'bcm4334x'
Dec 31 17:00:27 in10aps systemd-modules-load[106]: Inserted module 'g_multi'
Dec 31 17:00:27 in10aps systemd-modules-load[106]: Failed to find module 'uvcvid
Dec 31 17:00:27 in10aps systemd-modules-load[106]: Failed to find module 'ftdi_s
lines 1-6/6 (END)
Scott Leibrand
@scottleibrand
May 05 2017 17:51
@dmdfreak when UAM is active and you have COB, you should see both UAMpredBGs and COBpredBGs in the pump-loop.log and the OpenAPS pill, and you should have four purple lines. If you ever see UAMpredBGs in the logs/pill and only three purple lines, that probably means you're still running the old NS code
dmdfreak
@dmdfreak
May 05 2017 17:52
@scottleibrand also on a side note, by the time I received your private message I was already sure I won the treasure hunt LOL
ok. Then I havent figured out how to update it correctly then. I have everything but the fourth line
dmdfreak
@dmdfreak
May 05 2017 18:31
Will the micro boluses show up with the old NS @scottleibrand ?
Dana Lewis
@danamlewis
May 05 2017 18:34
@cameronrenwick :) thanks for your comment on the blog! FYI - if you’d like to test it while it’s in dev (keeping in mind we don’t have all the documentation written, so it requires a little bit more reading of in-channel conversations to see what info has already been described), there’s some notes in this PR about different things that we’d like testing feedback on: https://github.com/openaps/oref0/pull/424#issuecomment-298287623
(ditto for anyone else who has expressed interest in testing oref1-related features before they get all the way to master) ^
Scott Leibrand
@scottleibrand
May 05 2017 18:34
@dmdfreak Yes, they're just regular small boluses.
Dana Lewis
@danamlewis
May 05 2017 18:35
cameronrenwick
@cameronrenwick
May 05 2017 18:36
just starting to do that as we speak @danamlewis ! SMB really excites me (if one can get excited about T1DM...) so I'm certainly going to read up over the next few days...
Dana Lewis
@danamlewis
May 05 2017 18:36
:)
dmdfreak
@dmdfreak
May 05 2017 18:38
I haven't figured out how to update my NS to the correct branch so I wasn't sure if your picture was what mine would look like it not. @danamlewis hehe
Martin Haeberli
@mhaeberli
May 05 2017 18:40
reminders re preferred NS branch? master or dev? (and, foolishly, at the moment I think we’re deployed from the public cgm-remote-monitor repository, not our own fork...
Dana Lewis
@danamlewis
May 05 2017 18:40
@mhaeberli for standard use, master is fine
Martin Haeberli
@mhaeberli
May 05 2017 18:41
:+1: any easy way to switch a deployed heroku from dev to master? (I have come to spin up n heroku test sites for different test rigs…)
fjungman
@fjungman
May 05 2017 18:45
@diabetic_dude_twitter was there another PID in the list? That one is not the right one...
PieterGit
@PieterGit
May 05 2017 18:45
@scottleibrand you mean with falling faster that the the glucose level is falling. You see that by the "Min. Delta 6.0 < Exp. Delta -3.4 part"? Can you explain what the Min Delta means? I can't figure it out from the determine-basal.js
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 18:45
yes working on that now 8772
root@in10aps:~# journalctl _PID=8772
-- Logs begin at Fri 1999-12-31 17:00:26 MST, end at Fri 1999-12-31 17:58:35 MST
lines 1-1/1 (END)
root@in10aps:~# systemctl status systemd-timesyncd.service -l
● systemd-timesyncd.service - Network Time Synchronization
   Loaded: loaded (/lib/systemd/system/systemd-timesyncd.service; enabled)
   Active: failed (Result: start-limit) since Fri 1999-12-31 17:19:46 MST; 16s ago
     Docs: man:systemd-timesyncd.service(8)
  Process: 8772 ExecStart=/lib/systemd/systemd-timesyncd (code=exited, status=226/NAMESPACE)
 Main PID: 8772 (code=exited, status=226/NAMESPACE)

Dec 31 17:19:46 in10aps systemd[1]: Failed to start Network Time Synchronization.
Dec 31 17:19:46 in10aps systemd[1]: Unit systemd-timesyncd.service entered failed state.
Dec 31 17:19:46 in10aps systemd[1]: systemd-timesyncd.service has no holdoff time, scheduling restart.
Dec 31 17:19:46 in10aps systemd[1]: Stopping Network Time Synchronization...
Dec 31 17:19:46 in10aps systemd[1]: Starting Network Time Synchronization...
Dec 31 17:19:46 in10aps systemd[1]: systemd-timesyncd.service start request repeated too quickly, refusing to start.
Dec 31 17:19:46 in10aps systemd[1]: Failed to start Network Time Synchronization.
Dec 31 17:19:46 in10aps systemd[1]: Unit systemd-timesyncd.service entered failed state.
root@in10aps:~#
PieterGit
@PieterGit
May 05 2017 18:51
Is it possible to use nightscout uploading with openaps and AUTH_DEFAULT_ROLES=denied . Should I set the API_SECRET to the main password, or is it also possible to authenticate the rig with a token (http://www.nightscout.info/wiki/welcome/website-features/0-9-features/authentication-roles )
Scott Leibrand
@scottleibrand
May 05 2017 18:58
Min delta is how much BG is falling over the last ~5m, the last ~15m, or the last ~30m (the minimum of those three)
PieterGit
@PieterGit
May 05 2017 18:59
aha, will PR that for the glossary
Scott Leibrand
@scottleibrand
May 05 2017 18:59
@PieterGit we need someone to update oref0-setup to support NS auth properly. It'll probably require refactoring all the OpenAPS to NS upload methods to use the nightscout script instead of direct curls
expected delta is better documented, I think. that's the BGI, adjusted for how much BG would need to rise/fall linearly over DIA to get eventualBG to target.
PieterGit
@PieterGit
May 05 2017 19:00
i just managed by hand, by adding the token to /usr/local/bin/ns-get and adding some roles
Scott Leibrand
@scottleibrand
May 05 2017 19:00
:+1:
PieterGit
@PieterGit
May 05 2017 19:05
which program is responsible for parsing ns.ini? could we use args = ns NIGHTSCOUT_URL TOKEN=yourtokenname-yourtokenhash in ns.ini and then use the token in tools such as ns-get (instead of the API_SECRET)
francesc0-cgm
@francesc0-cgm
May 05 2017 19:09
@scottleibrand using smb i have hours of zero temping...maybe due to low amount of insulin of my son?
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 19:12
@fjungman did I miss something doing that ?? journalctl _PID=8772 did not return much at all
David
@davidkeddydb
May 05 2017 19:13
If I have a newer explorer board, I should be able to plug it in to charge on the uart and be able to put the G4 Dex Receiver on OTG with a cable and it should be able to work off-line right?
PieterGit
@PieterGit
May 05 2017 19:21
yes
but if you want to use temp targets it's probably best to use bt or wifi tethering to your phone, to connect to your nightscout instance
David
@davidkeddydb
May 05 2017 19:25
I have issues sometimes overnight with my Share going down and then no temp basal happening. Maybe I plugged into my old explorer board rather than my newer one. I built to many rigs!
Dana Lewis
@danamlewis
May 05 2017 19:25
@davidkeddydb Do you routinely plug your receiver in overnight? I would do that regardless of what kind of rig you use, so you have it for offline looping if share or wifi goes down
David
@davidkeddydb
May 05 2017 19:27
@danamlewis Yes, I think I may have plugged into the original Explorer board than the latest one. I should have looked at the date stamp on the board, but I may have overlooked it. Try again tonight. It is something with my phone or something and share. It drops it every once in a while and I have to restart it.
fjungman
@fjungman
May 05 2017 19:34
@diabetic_dude_twitter Yeah, I was expecting it would return the error code, but it looks like there isn't one. Have you tried starting the service manually? Does it have the same issue?
systemctl start systemd-timesyncd.service
cameronrenwick
@cameronrenwick
May 05 2017 19:35
@danamlewis so just fyi, I've got an unused 712 pump that I think I'm going to set up with a new explorer/edison vs working on my current 522 rig. It's working really well and I don't want to mess with it so my SMB right now will be confined to reading and prep work
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 19:36
root@in10aps:~# systemctl start systemd-timesyncd.service
Job for systemd-timesyncd.service failed. See 'systemctl status systemd-timesyncd.service' and 'journalctl -xn' for details.
and I have all of this stuff on that webpage I sent out
Dana Lewis
@danamlewis
May 05 2017 19:37
@cameronrenwick ahhh if you do an x12 can I put in a request?
cameronrenwick
@cameronrenwick
May 05 2017 19:38
sure ting!
Dana Lewis
@danamlewis
May 05 2017 19:38
the x12 tweak instructions were written over a year and a half ago, and no one’s touched them since then. so if you do 712 PLEAAAAASSSEEEEEE (i’m resorting to begging, since I’ve posted this something like 6 times in the last two weaks) update the docs with more clear instructions on what you have to do (maybe all in one place) to get the x12 to work? :)
everyone gets in here, asks, I say that…then they go away happily looping but don’t update or make note of what else they had to do. docs stay the same. someone else comes along. rinse, repeat…now I resort to begging ;)
@Kdisimone did you end up updating your NS? (see earlier convo’s on this to display all the purple lines when SMBing)
cameronrenwick
@cameronrenwick
May 05 2017 19:40
for sure - will do. My big problem is time as it take waay too long to get parts up from the US to me. (Free trade...lol). Anyway, I will work on that for sure once I've got rig #2 going.
Dana Lewis
@danamlewis
May 05 2017 19:41
:+1: thanks!
elodaille01
@elodaille01
May 05 2017 19:44
Good evening everybody. .could you tell me (again) what I have to do when the rig goes off because there is no battery, so that OpenAPS loops again naturally once plugged ?
cameronrenwick
@cameronrenwick
May 05 2017 19:45
once back and up and running I usually do killall -g and then openaps first-upload
elodaille01
@elodaille01
May 05 2017 19:48
When the rig is 'on' again, I run 'killall -g' then 'openaps first -upload' (with a 'space' between 'first' and '-'?)
Isn't it ?
fjungman
@fjungman
May 05 2017 19:50
@diabetic_dude_twitter I had read your webpage prior to asking the first questions, but we ended up at the same place! :)
I see this could relate to your issue: https://bbs.archlinux.org/viewtopic.php?id=189194
cameronrenwick
@cameronrenwick
May 05 2017 19:52
@elodaille01 no there is no space
first-upload (all one word)
katie disimone
@Kdisimone
May 05 2017 19:56
@danamlewis yup, I have NS on floating carbs.
Dana Lewis
@danamlewis
May 05 2017 19:56
:+1:
Lynne Butts
@flutelynne_twitter
May 05 2017 20:00
Can’t wait til we get the release for SMB!
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 20:00
@fjungman yeah I saw that page the other day but did not try the command because when i traced the paths and come up empty
dmdfreak
@dmdfreak
May 05 2017 20:00
@Kdisimone could I beg you to tell me what I'm missing? I'm doing something wrong between hitting fork and telling azure to update. I feel like I'm missing a step.
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 20:00
root@in10aps:~# cd ..
root@in10aps:/# cd usr
root@in10aps:/usr# cd lib
root@in10aps:/usr/lib# cd systemd/
root@in10aps:/usr/lib/systemd# cd system
-bash: cd: system: No such file or directory
root@in10aps:/usr/lib/systemd# ls
catalog  network  ntp-units.d  user  user-generators
root@in10aps:/usr/lib/systemd#
@fjungman and this is more of a jubilinux issue not an openAPS issue right? so I would need to reflash my edison so start over? Or is there an easier way?
Dana Lewis
@danamlewis
May 05 2017 20:03
@flutelynne_twitter :) needs a lot of testing before we get to a release to master!
Lynne Butts
@flutelynne_twitter
May 05 2017 20:04
Understandable.
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 20:04
@danamlewis are you testing now with FIASP of humalog?
Lynne Butts
@flutelynne_twitter
May 05 2017 20:04
Exciting, though!
Dana Lewis
@danamlewis
May 05 2017 20:04
@flutelynne_twitter yea :)
@diabetic_dude_twitter I haven’t tried FIASP
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 20:04
good me either!
katie disimone
@Kdisimone
May 05 2017 20:05
@dmdfreak looks like you don't have the floating carbs branch yet. What I suggest is switch to the dev branch on your cgm-remote-monitor in GitHub. Then add a branch called floating-carbs. (It will create it from dev branch). Then do a compare using this link (below) to create a pull request and merge it into your new floating-carbs branch.
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 20:05
I know it will work better for all of us but still cant get it yet
Dana Lewis
@danamlewis
May 05 2017 20:05
@diabetic_dude_twitter not sure that it’ll work for everyone.
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 20:06
@danamlewis it meaning FIASP or oref1 ?
Dana Lewis
@danamlewis
May 05 2017 20:06
like all things, worth trying, but it’s not cinnamon ;)
FIASP
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 20:06
yup!
agreed
Dana Lewis
@danamlewis
May 05 2017 20:06
technically oref1, too
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 20:06
lol
Dana Lewis
@danamlewis
May 05 2017 20:06
but I was referring to FIASP originally
Diabetic_Dude
@diabetic_dude_twitter
May 05 2017 20:06
too true but ... I will try them both and stick with what works
Dana Lewis
@danamlewis
May 05 2017 20:07
:+1:
francesc0-cgm
@francesc0-cgm
May 05 2017 20:08
@dmd
dmdfreak
@dmdfreak
May 05 2017 20:08
@Kdisimone thanks! I'll go try that :)
francesc0-cgm
@francesc0-cgm
May 05 2017 20:08
Sorry
@dmdfreak ^^
@dmdfreak let me know if you need help on creating the new branch. I'm on a mobile right now, so I can't screen shot.
PieterGit
@PieterGit
May 05 2017 20:13
@scottleibrand : would the following scheme for preferences.json be ok?
  • all lower caps
  • no init caps but underscores for separating words
  • feature toggles start with enable_
  • fractions end with _fraction (?)
fjungman
@fjungman
May 05 2017 20:13
@diabetic_dude_twitter Considering the amount of time this is taking, and there are hints that it could be a missing permission or part of ntpd still left behind on your system, I would just reflash from the start. It doesn't take that long, and I haven't had any problems with the time server, and I've installed from scratch five times last month.
PieterGit
@PieterGit
May 05 2017 20:14
examples:
  • max_cob instead of maxCOB
  • enable_uam instead of enableUAM
  • enable_rewind_resets_autosens instead of rewind_resets_autosens
  • enable_att instead of adv_target_adjustments
garykidd
@garykidd
May 05 2017 20:16
anyo ne no why I am getting could not get a bored instance when attempting to paste a pic in here...hadn't happened before...
No responsive uploaders
dmdfreak
@dmdfreak
May 05 2017 20:18
@Kdisimone I'm a little lost. everything I've done in github in the past had exact instructions on what to do. So I'm feeling very green
katie disimone
@Kdisimone
May 05 2017 20:18
Ok. Hold on one sec
dmdfreak
@dmdfreak
May 05 2017 20:19
the treasure hunt yesterday was easier than this :smile:
katie disimone
@Kdisimone
May 05 2017 20:21
dmdfreak_cgm-remote-monitor_at_dev.jpg
Go to your repo and then select dev branch from that little dropdown
Dana Lewis
@danamlewis
May 05 2017 20:22
@PieterGit att is the kind of abbreviation I'd like to stay away from - I think we do need words fully spelled out since we are inventing features and are trying to make them as clear as possible (and appreciate your other thoughts around clarity on these).
katie disimone
@Kdisimone
May 05 2017 20:22
then, after you’ve selected it, type in the word floating-carbs into that line that has the greyed out words “filter branches/tags"
It will create a new branch after you press return
then do the compare link and create a pull request and merge
PieterGit
@PieterGit
May 05 2017 20:24
@danamlewis : i'm in favor of using explicit is better than implicit, but the same would apply to UAM en SMB
the ratio/fractions stuff is also quite inconsistent on current dev properties.json. Isn't "remainingCarbsCap" also a ratio?
the easiest is max_cob instead of maxCOB :smile:
Scott Leibrand
@scottleibrand
May 05 2017 20:27
@francesc0-cgm usually along with the hours of zero temping, you'll see a few scattered microboluses when BG upticks. that's what I mean by not actually having hours of zero insulin delivery
PieterGit
@PieterGit
May 05 2017 20:27
i think the renaming and making it consistent must be done before merging it to master, because otherwise the docs will also have the inconsistent names
garykidd
@garykidd
May 05 2017 20:29
@scottleibrand was interested in hearing your view on why SMBs would be better than just rasing temp basal limit/s...
Lynne Butts
@flutelynne_twitter
May 05 2017 20:29
@Kdisimone I am trying to update along with your instructions above, and I get “Some check haven’t completed yet."
dmdfreak
@dmdfreak
May 05 2017 20:30
I'm still missing something
Scott Leibrand
@scottleibrand
May 05 2017 20:30
@PieterGit I am fine with underscores, lowercase, and enable_ and _fraction on anything that we're newly adding to dev. I agree with @danamlewis that something like enable_adv_target_adjustments would be clearer. I do not want to change any preference names for anything already in master.
and no, remainingCarbsCap has units of grams: the _fraction version only made it as far as a comment, because I don't think anyone will actually need to change that from 2/3 (and we have way too many preferences items already)
so remaining_carbs_cap would be fine there.
katie disimone
@Kdisimone
May 05 2017 20:33
@dmdfreak I've made the pull request for you. You just need to go into your repo and approve (merge) it. That should help
@flutelynne_twitter what's your GitHub name?
Lynne Butts
@flutelynne_twitter
May 05 2017 20:35
LynneRB
dmdfreak
@dmdfreak
May 05 2017 20:35
ok I think I managed to do THAT part right. Thank you sooo much. Now lets see if I can do the azure part correctly now.
katie disimone
@Kdisimone
May 05 2017 20:35
@dmdfreak looks like you got it! Nice. 👍🏻
Scott Leibrand
@scottleibrand
May 05 2017 20:36
@garykidd for example, let's say you need 1.5U of additional insulin on top of a regular basal of 1U/hr. if you set a temp basal at 4U/hr, you could deliver that over 30m. but with SMB, you could deliver that more quickly, as microboluses of 0.5U, 0.4U, 0.3U, 0.2U, and 0.1U, so that most of the extra insulin is delivered in the first 10-15m, rather than taking 30m
dmdfreak
@dmdfreak
May 05 2017 20:36
Not nice. All I had to do was push the button and even then I was worried I was doing it wrong LOL
Scott Leibrand
@scottleibrand
May 05 2017 20:37
@garykidd 0.5U per 5m would be equivalent to a temp basal rate of 6U/hr: if you set that for 30 minutes with the intention of canceling it after 15m, you would risk a severe hypo if your rig lost connectivity and it ran to completion
dmdfreak
@dmdfreak
May 05 2017 20:37
Nice of you to do it for me though. I was not getting anywhere quickly
Lynne Butts
@flutelynne_twitter
May 05 2017 20:37
@dmdfreak At least you found the treasure. I am just getting things ready.
Scott Leibrand
@scottleibrand
May 05 2017 20:37
@garykidd by contrast, with SMB's zero temps, you have walk-away hypo safety because it will stop delivering SMBs and run a long enough zero temp to cancel out the IOB even if carb absorption stops
katie disimone
@Kdisimone
May 05 2017 20:38
@flutelynne_twitter I left you a PR to approve (merge) too. That should work...
garykidd
@garykidd
May 05 2017 20:39
so the MB is limited by how much insulin can be offset by subsequent zero temps?
dmdfreak
@dmdfreak
May 05 2017 20:39
lol @flutelynne_twitter it feels good once you find it though lol
Lynne Butts
@flutelynne_twitter
May 05 2017 20:39
@Kdisimone Thank you! At least that part is complete!
katie disimone
@Kdisimone
May 05 2017 20:40
👍🏻👍🏻
Scott Leibrand
@scottleibrand
May 05 2017 20:40
@garykidd no, the SMB size is limited to 30 minutes' worth of basal, and the total SMB IOB is limited by max_iob
katie disimone
@Kdisimone
May 05 2017 20:41
If you're on heroku @flutelynne_twitter then you just need to login there and click on the deploy tab...choose the new floating carbs branch from the list (it's the bottom of the deploy screen)
Lynne Butts
@flutelynne_twitter
May 05 2017 20:41
@Kdisimone Already done!
katie disimone
@Kdisimone
May 05 2017 20:41
👍🏻👍🏻
Scott Leibrand
@scottleibrand
May 05 2017 20:41
@garykidd we limit ourselves to setting 120-minute long zero temps: if your max_iob is high enough, it's possible that 2 hours is not quite enough to offset all the IOB. but in reality, if you eat a big meal with SMB and no activity mode, and carb absorption stops completely because you start running , you're going to go low unless you eat sugar regardless
garykidd
@garykidd
May 05 2017 20:42
thinking
Lynne Butts
@flutelynne_twitter
May 05 2017 20:43
@Kdisimone I redid dev setup last night on an RPi that I run the usual dev on, but did not see the SMB in the preferences.json part. I will wait until it is closer or I can find the treasure.
PieterGit
@PieterGit
May 05 2017 20:43

@danamlewis @scottleibrand
Ok, recap:
rewind_resets_autosens => enable_rewind_resets_autosens
adv_target_adjustments => enable_adv_target_adjustments
maxCOB => max_cob
remainingCarbsCap => REMOVE

What about SMB and UAM? Isn't microbolus more clear than SMB?
You wrote that enableUAM won't be very effective without enableSMB_with_bolus, wouldn't it be good to create a
enable_microbolus_unannounced_meals. What would be a good suggestion for enableSMB_with_bolus

francesc0-cgm
@francesc0-cgm
May 05 2017 20:43
@Kdisimone is it normal that papertrail doesn't show recent log when you are looking to pump-log locally via terminal? While when the rig goes on bt tethering papertrail shows logs
garykidd
@garykidd
May 05 2017 20:44
And no separate max_smb_basal_iob limit?
just 30 mins of basal and not greater than max_iob
?
katie disimone
@Kdisimone
May 05 2017 20:44
You have to edit your runagain script to add microbolus to the enable part. Where meals and autosens are. Sorry. Not logged into rig to get a better copy paste for you. Then run your runagain script
@flutelynne_twitter
Dana Lewis
@danamlewis
May 05 2017 20:45
@Kdisimone prefer people to read to find enable line
Vs copy paste
katie disimone
@Kdisimone
May 05 2017 20:46
Read where? You mean send her to PR? I was just going to show sample...not my whole runagain line
Scott Leibrand
@scottleibrand
May 05 2017 20:47
@PieterGit no, remainingCarbsCap => remaining_carbs_cap
elodaille01
@elodaille01
May 05 2017 20:47
@scottleibrand Wouldn't there be a way for openaps to start looping once the rig has been turned on ?? (my rig went out for empty battery cause ..)..@cameronrenwick , thank you..but I imagine that the rebooting of openaps should be done on its own ?
dmdfreak
@dmdfreak
May 05 2017 20:48
@Kdisimone we have to look through the code to get the flag. It was worth it in the end :smile:
katie disimone
@Kdisimone
May 05 2017 20:48

@flutelynne_twitter scott covered what needs to be added in the PR, see the first bullet of one of his comments on the PR.

openaps/oref0#424

Dana Lewis
@danamlewis
May 05 2017 20:49
@Kdisimone code to see what the flag is called
@dmdfreak :+1:
Scott Leibrand
@scottleibrand
May 05 2017 20:49
@elodaille01 I don't know why you have that problem: no one else seems to.
dmdfreak
@dmdfreak
May 05 2017 20:50
And I have 4 purple lines! Thank you @Kdisimone !
Dana Lewis
@danamlewis
May 05 2017 20:52
:clap:
katie disimone
@Kdisimone
May 05 2017 20:52
👍🏻
garykidd
@garykidd
May 05 2017 20:52
has @bewest been back here since new gig?
PieterGit
@PieterGit
May 05 2017 20:53
ok, remainingCarbsCap => remaining_carbs_cap
. I was misled by the comment in https://github.com/openaps/oref0/commit/143879ffcdbc58feaef3594a7603a5c58a683eab#diff-ca0087dd37432c09b616efd9197f60a0R30
isn't that a oref1 feature?
elodaille01
@elodaille01
May 05 2017 20:53
thank you @scottleibrand ..Whatever, the way @cameronrenwick told me to follow doesn't work for me (killall -g , then openaps first-upload)..any idea to fix it manually please ? thank you ;)
PieterGit
@PieterGit
May 05 2017 20:54
does SMB work with carbs entered on the pump? I see quite some COB:0 or COB: undefined
dmdfreak
@dmdfreak
May 05 2017 20:54
My goal is to one day be able to actually answer a question on here....correctly. Just one...
PieterGit
@PieterGit
May 05 2017 20:55
@dmdfreak : we'll an easy one to start: do you like openaps and this community so far? :smile:
Dana Lewis
@danamlewis
May 05 2017 20:55
@dmdfreak there's going to be people who need coached through the treasure hunt as you call it; plus enabling purple lines. So keep your eyes peeled and I bet that day will be today or tomorrow ;)
Scott Leibrand
@scottleibrand
May 05 2017 20:55
@PieterGit yeah, remainingCarbsFraction is not a thing. If anyone comes up with a use case where we need that, we can uncomment it. otherwise, it'll be removed before we merge to master.
dmdfreak
@dmdfreak
May 05 2017 20:56
HAHA! @PieterGit not sure it counts but I'll take it! YES
Scott Leibrand
@scottleibrand
May 05 2017 20:56
@PieterGit yes, pump bolus wizard is a great way to enter carbs. not sure why you're seeing 0 or undefined: that'll require some investigation
dmdfreak
@dmdfreak
May 05 2017 20:57
I'll keep my eye open @danamlewis :+1:
Dana Lewis
@danamlewis
May 05 2017 20:57
@scottleibrand can't remember if we tested SMB without AMA ...
That might yield undefined ?
PieterGit
@PieterGit
May 05 2017 20:57
AMA is enabled
Dana Lewis
@danamlewis
May 05 2017 20:57
K
PieterGit
@PieterGit
May 05 2017 21:00
found a COB issue because , because i use http://www.nightscout.info/wiki/welcome/website-features/0-9-features/authentication-roles
cat monitor/carbhistory.json
{"status":401,"message":"Unauthorized","description":"Invalid/Missing"}
Lynne Butts
@flutelynne_twitter
May 05 2017 21:02
@Kdisimone Thanks. I will look after dinner, and appreciate the “not copy paste” idea, too. If not, I will wait.
Scott Leibrand
@scottleibrand
May 05 2017 21:03
@PieterGit ah, yeah, that'll do it. are you up for refactoring NS uploads & downloads (as configured by oref0-setup) to support auth roles? might be a good idea to have an issue for it too.
katie disimone
@Kdisimone
May 05 2017 21:03
@dmdfreak here you can start here. What's the treasure you are referring to? What did I miss? 😜
PieterGit
@PieterGit
May 05 2017 21:04
yeah, i'll add an issue, seems quite easy so maybe even a pull request quite soon
dmdfreak
@dmdfreak
May 05 2017 21:08
Treasure hunt = search for the secret flag to turn on SMB
Alin Artiom Kenibasov
@ktomy
May 05 2017 21:09
Hello everyone. I have 3 OpenAPS instances and on 2 of them I experience an interesting issue. In NS web page, like each minute or two, "x min ago" disapears and "unknown" aprars, then, after another one minute or two it comes back. and it "flickers" in this way all the time. All 3 instances have last dev version. I'll also update NS to have last version now. But, is this a known issue or should I research? Thank you.
katie disimone
@Kdisimone
May 05 2017 21:09
Ahhh...I missed that somehow. 😜
Dana Lewis
@danamlewis
May 05 2017 21:11
@ktomy check for future treatments in NS
katie disimone
@Kdisimone
May 05 2017 21:12
How long have you had NS running? If those messages are in the pump and OpenAPS pills, that can be a problem with the DEVICESTATUS_ADVANCED line in your site configuration. Check that the line has been added and that the variable is set to true. Make sure there are no typos in it.
Alin Artiom Kenibasov
@ktomy
May 05 2017 21:36
@danamlewis indeed, NS detects "future" treatments, but they are from last 4-5 hours. I have seen that in my profile timezone is the right one (GMT+03:00). Do you know where from does it take timezone?
PieterGit
@PieterGit
May 05 2017 21:59
blob
@danamlewis @scottleibrand : can you explain why I can't get SMB to bolus? COB in reason still 0. The 0.7U bolus was a manual bolus on the pump
PieterGit
@PieterGit
May 05 2017 22:05
what does minCOBPredBG: 999 mean?
francesc0-cgm
@francesc0-cgm
May 05 2017 22:09
@seems not activated...
@PieterGit
Did you set microbolus in enable section of oref0-setup?
PieterGit
@PieterGit
May 05 2017 22:12
@francesc0-cgm : thanks. I only found the setttings in preferences.json
francesc0-cgm
@francesc0-cgm
May 05 2017 22:14
Np... nano your oref0-setup and in enable section write "microbolus"
Then use run again and it will be on
Then write true in preferences.json for each one you want to use
PieterGit
@PieterGit
May 05 2017 22:16
yep, i did it manual for now, because i have some open changes for a patch for better nightscout authentication
francesc0-cgm
@francesc0-cgm
May 05 2017 22:19
:+1:
@scottleibrand will UAM work also without setting anything on NS or using eating soon features? For example...i give to my friend some uncovered carbs without doing anything...
My friend is my son...sorry but here in italy it is midnight...
PieterGit
@PieterGit
May 05 2017 22:45
@scottleibrand @danamlewis @francesc0-cgm : thanks for the help tonight, will work on a pullrequest for support for token authentication with nightscout and reinstall with microbolus/smb mode
cameronrenwick
@cameronrenwick
May 05 2017 22:52
kind of off topic but every time I load a new battery into my 522 I get a 'A21' error (warning??) anyway, I have to fully reset the time/date/year...although other things like carb/ISF are maintained. Any body know what the A21 means?
Scott Leibrand
@scottleibrand
May 05 2017 22:59
@cameronrenwick sounds like the (capacitor?) that keeps the real-time-clock running is no longer functioning well on that pump
@francesc0-cgm there are config flags to enable SMB after a bolus, from carb entry, or via a temp target. if you do none of those, it will just continue the old oref0 behavior of high-temping based on the observed deviation.
we don't have an option that would cause it to SMB 24x7 any time a high temp would be justified. at the current stage of testing, we felt it would be far safer to only run SMBs when triggered by user action of some sort
plus, the constant vibrations/beeps every time it microboluses would get annoying when you're trying to sleep. ;-)
francesc0-cgm
@francesc0-cgm
May 05 2017 23:19
@scottleibrand i was talking about UAM
SMB is already full working for me and i find it a great feature
Scott Leibrand
@scottleibrand
May 05 2017 23:19
var enableUAM=(profile.enableUAM && (bolusiob > 0.1 || enableSMB));
francesc0-cgm
@francesc0-cgm
May 05 2017 23:20
Would like to know if i will switch on UAM too i have to follow a rule to get it working or i can give my son carbs without telling anything to NS
Is this the string to enable UAM?
Could i do manually on preferences.json?
Scott Leibrand
@scottleibrand
May 05 2017 23:21
that line of code is the logic in determine-basal.js that tells oref0: enableUAM if the preference is turned on, and either you've bolused in the last DIA hours, or SMB is enabled.
francesc0-cgm
@francesc0-cgm
May 05 2017 23:21
Yes but reading docs i red out you have to set an eating soon 2 hours before
Scott Leibrand
@scottleibrand
May 05 2017 23:22
what is the desired behavior for you?
francesc0-cgm
@francesc0-cgm
May 05 2017 23:22
Now we can avoid that?
Scott Leibrand
@scottleibrand
May 05 2017 23:22
or, which SMB/UAM options do you have enabled in preferences?
francesc0-cgm
@francesc0-cgm
May 05 2017 23:22
Now i have only SMB for all the 3 choices
Scott Leibrand
@scottleibrand
May 05 2017 23:23
k
francesc0-cgm
@francesc0-cgm
May 05 2017 23:23
When i will enable UAM, to test it i can just feed him?
And looking what happens?
Scott Leibrand
@scottleibrand
May 05 2017 23:23
so if you enableUAM as well, then any time SMB is active, UAM will be active as well.
francesc0-cgm
@francesc0-cgm
May 05 2017 23:23
Ok that's fine :D
Scott Leibrand
@scottleibrand
May 05 2017 23:23
but you have to do one of those three things to activate both: bolus, enter carbs, or enter a low temp target
francesc0-cgm
@francesc0-cgm
May 05 2017 23:23
Got it!
Thanks
dmdfreak
@dmdfreak
May 05 2017 23:42
@scottleibrand it says starting supermicrobolus l pump-loop
Scott Leibrand
@scottleibrand
May 05 2017 23:45
ok, then you've got the oref0-setup piece correct
what else do you have in pump-loop.log? is there anything that's failing?
dmdfreak
@dmdfreak
May 05 2017 23:46
Ok cool. That makes me feel better. Let me look
Not that I can tell
Scott Leibrand
@scottleibrand
May 05 2017 23:50
what does enact/smb-suggested.json say?
dmdfreak
@dmdfreak
May 05 2017 23:50
Less than 1m old
Scott Leibrand
@scottleibrand
May 05 2017 23:52
no, the actual contents. it shows in the pump-loop.log as something like:
enact/smb-suggested.json: {"insulinReq":0.1,"bg":117,"reservoir":"74.7","temp":"absolute","snoozeBG":105,"predBGs":{"IOB":[117,117,117,118,118,118,117,117,117,116,116,115,114,114,113,112,111,111,110,109,109,108,108,107,107,107,106,106,106,105,105,105,105,105,105,104,104,104,104,104,104,105]},"minPredBG":104,"IOB":0.234,"reason":"COB: 0, Dev: 2, BGI: -0.09, ISF: 61, Target: 98, IOBpredBG 105; Eventual BG 105 >= 98, temp 1.125 >~ req 1.15U/hr","COB":0,"eventualBG":105,"tick":"+1","deliverAt":"2017-05-01T13:34:13.981Z"}
dmdfreak
@dmdfreak
May 05 2017 23:52
Yeah I realized I looked at the wrong thing. One sec
enact/smb-suggested.json: {"insulinReq":0.1,"bg":80,"reservoir":"62.6","temp":"absolute","snoozeBG":92,"predBGs":{"UAM":[80,81,81,82,83,83,84,85,86,87,88,89,90,91,93,94,95,96,97,98,99,100,101,102,103,103,104,105,105,106,106,107,107,108,108,109,109,109,110],"IOB":[80,82,83,84,85,86,87,88,88,89,89,89,89,89,89,89,89,90]},"minPredBG":90,"IOB":0.076,"reason":"COB: 0, Dev: 15, BGI: -0.88, ISF: 42, Target: 86, IOBpredBG 90, UAMpredBG 110; Eventual BG 110 >= 89, insulinReq 0.1; . temp 1.3 >~ req 1.25U/hr","COB":0,"eventualBG":110,"tick":"+2","deliverAt":"2017-05-05T23:48:11.890Z"}
Scott Leibrand
@scottleibrand
May 05 2017 23:57
that looks good to me. you don't need an SMB right now, as insulinReq is only 0.1. but it's calculating UAM predBGs and looks ready to SMB if it needs to
dmdfreak
@dmdfreak
May 05 2017 23:58
At lunch I under bolused and under counted my carbs to see if it would SMB but never did. Eventually I had to correct bolus to bring it back down so I thought maybe I had something wrong
Scott Leibrand
@scottleibrand
May 05 2017 23:59
would be worth looking at the reason field from that time period. you can scroll back in NS, or look at pump-loop.log