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

22nd
Oct 2017
Jon Cluck
@cluckj
Oct 22 2017 00:06
was fixed in the latest NS dev branch, apparently :eyes:
Dana Lewis
@danamlewis
Oct 22 2017 00:26
@garykidd no. Not merged to master. 707 is current open dev PR
I found a Github bug ;) which is why that Pr moved twice
WeedsDIY
@WeedsDIY
Oct 22 2017 00:53
@danamlewis Just wanted to say thanks for jumping in to help us. We often feel utterly lost when attempting to trouble shoot, so we just want to say how much we appreciate any input from experienced ones in the community. Really hoping we'll be able to assist at the same level one day. For this problem, loop failed after restarting dynos (same "Bg too old" error), until we re-ran setup (for the 3rd time). Perhaps a setup entry error? Who knows. Now simply gleeful now that SMB loop is running. Can't wait to watch what it does! Many thanks!
John Benjamin
@Bender1061
Oct 22 2017 00:59
So i'll put in a PR later, but there is a huge difference between jubilinux0.1.1 and jubilinux0.2.0, the Newer version failed like non other when I ran the installing openaps script. I was manually doing most of it. I'm thinking that one does not like the BT-pan stuff. Probably is a way to make it work. but I have other fish to fry. I'm hoping the v0.1.1 works with the offline branch, (Right now I know v0.0 my current rig does not work 100%). but it installed everything no problem, currently running the oref0-setup right now. So I'm thinking we need a warning for the people that install jubilinux on the edison themselves. to run jubilinux0.1.1 instead of 0.2 at least till someone figures out a fix for all the issues with it for Openaps.
Dana Lewis
@danamlewis
Oct 22 2017 01:31
@WeedsDIY hm, maybe a typo in API secret or something? Who knows, glad it's fixed now! You're getting lots of great experience with the core troubleshooting steps (silver lining?)
@Bender1061 I think all of ours are on 0.2.0, haven't seen the kind of issues you've been experiencing. Wondering if there's anything else custom to your setup?
John Benjamin
@Bender1061
Oct 22 2017 01:47
@danamlewis Clean setup, as in I just flashed it with 0.2.0, I'm also having an issue with 0.1.1, and all the issues have to do with the offline branch install.
it kicked me out sometimes after it installed bluez, I'm kicking myself for closing the putty. Cause now, it won't connect to the wifi, but it did not install enough for me to recover. I think I found an issue, but now to try to figure out how to make it connect to my wifi again, cause it won't let me run oref0-setup without a network connection now.
and that was with a completely brand new edison i just bought that i just installed jubilinux 0.1.1 on and only following the steps in the manual. other than when it asked if I wanted to run oref0-setup with master, I stopped it and installed offline on it instead.
John Benjamin
@Bender1061
Oct 22 2017 01:52
good think i have 8 rigs to play with!
jquitos
@jquitos
Oct 22 2017 02:40

So I attempted to re-flash tonight... won't try again until morning... but his is what I got in command prompt window when I tried to re-flash:

Device has DFU interface, but has no DFU functional descriptor
Cannot open DFU device 8087:0a99
Dfu device not found Timeout
Did you plug and reboot your board?
If yes, please try a recovery by calling this script with the --recovery option

John Benjamin
@Bender1061
Oct 22 2017 02:43
@jquitos you do have two usb cords plugged into the same computer running the ./flashall.sh ? that always threw me for a loop i kept forgetting both micro usbs have to be plugged in. not one hooked up to an external power source.
jquitos
@jquitos
Oct 22 2017 02:44
@Bender1061 that's correct...
John Benjamin
@Bender1061
Oct 22 2017 02:47
have you flashed an edison on this computer before? I'm assuming windows.
f
John Benjamin
@Bender1061
Oct 22 2017 02:52
well I hope you figure it out. I'm about to go to bed, it's been a 14 hour day of me sitting in front of the computer trying to work out another issue. so it's been a long day.
Iain
@iainct
Oct 22 2017 03:10
I'm a bit confused my the following:

enact/smb-suggested.json: {"insulinReq":0.59,"bg":147,"reservoir":"206.9","temp":"absolute","snoozeBG":279,"predBGs":{"COB":[147,161,173,185,195,204,211,218,224,229,233,235,237,238,239,238,237,234,231,227,222,216,210,204,198,192,187,182,178,173,170,166,162,159,156,154,151,149,147,146,144,143,141,140,139,139,138],"aCOB":[147,146,145,144,142,140,138,136,134,132,130,128,126,124,123,121,120,119,118,117,116,116,115,115,115,115,115,115,116,117,117,118,119,120,121,122,124,125,126,128,129,130,132,133,134,136,137,138],"UAM":[147,161,174,186,196,206,214,222,229,235,240,245,248,251,253,255,256,256,255,254,252,249,246,241,237,231,226,221,217,212,208,205,201,198,195,193,190,188,186,184,183,181,180,179,178,178,177],"IOB":[147,160,170,179,185,189,191,191,189,185,179,171,161,152,143,134,125,117,109,102,95,88,81,75,69,64,58,54,49,45,41,39]},"rate":1.75,"IOB":4.366,"reason":"COB: 62, Dev: 154, BGI: -10.66, ISF: 3.5, Target: 6.4, minPredBG 8.4, minGuardBG 8.1, IOBpredBG 2.2, COBpredBG 7.7, UAMpredBG 9.8; Delta 15 > 10% of BG 147: SMB disabled; temp 1.3<1.75U/hr. ","COB":62,"eventualBG":177,"duration":30,"tick":"+15","deliverAt":"2017-10-22T02:46:26.104Z"}
Chris L
@woodduckie
Oct 22 2017 03:12
Struggling with maxSafeBasal being limited at 1.8 U/hr. I've searched the past discussion & understand that maxSafeBasal will take the lower of the three: 1) max_daily_safety_multiplier (3) highest rate (1.0 U/hr), 2) current_basal_safety_multiplier (4) current rate (1.0 U/hr), or 3) max basal setting in pump (3.0 U/hr). How do I dig deeper to uncover the logic? Under scenarios 1 & 3, maxSafeBasal should be 3.0 U/hr, and scenario 2 would yield 4 U/hr.
jquitos
@jquitos
Oct 22 2017 03:12
@Bender1061 no...it was a pre flashed one that I bought...however the BT stopped working correctly and it was suggested that a reflash would help
Iain
@iainct
Oct 22 2017 03:14
@woodduckie what is the max value in your pump?
Chris L
@woodduckie
Oct 22 2017 03:14
@iainct max pump rate is 3.0 U/hr
Iain
@iainct
Oct 22 2017 03:18
Sorry just saw you clarified that.
Dana Lewis
@danamlewis
Oct 22 2017 03:20
@woodduckie double check your pump isn't on a pattern?
There's a .6 likely somewhere, is it possible your basal just switched?
@mhaeberli same time stamp as tweet. It's the offline branch. No docs or notes exist.
Chris L
@woodduckie
Oct 22 2017 03:27
thanks @iainct. I'll dig around. @danamlewis pump is not on a pattern. they are turned off. certainly would appear there is a .6 somewhere, but has been limited to 1.8 for the past couple of hours. 25 minutes ago basal rate goes to 0.4, so I'd expect 1.6 (0.4 * 4), but I still get 1.8 U/hr...
Dana Lewis
@danamlewis
Oct 22 2017 03:28
Is your rig otherwise running normally? Successfully refreshing history etc? Any errors?
And - have you run Autotune?
If so, check its basal recommendations and see if it has a .6 it is using..
Chris L
@woodduckie
Oct 22 2017 03:37
rig is running normally, yes. no current errors. autotune yes & see a 0.600 there. forgot to reset, as have been running "offline" to gain confidence before going live.
good insight @danamlewis :)
Dana Lewis
@danamlewis
Oct 22 2017 03:38
:+1: so there's your 1.8 = autotune's 0.6x3
Iain
@iainct
Oct 22 2017 03:38
any idea why SMB is disabled under this condition?
"rate":1.75,"IOB":4.366,"reason":"COB: 62, Dev: 154, BGI: -10.66, ISF: 3.5, Target: 6.4, minPredBG 8.4, minGuardBG 8.1, IOBpredBG 2.2, COBpredBG 7.7, UAMpredBG 9.8; Delta 15 > 10% of BG 147: SMB disabled; temp 1.3<1.75U/hr. "
Dana Lewis
@danamlewis
Oct 22 2017 03:38
You can adjust the multipliers to be higher if you are confident that's what you want. But wanted you to understand that number is coming from somewhere logical, first :smile:
@iainct can you repaste w/out formatting? Can't see on gitter app , won't let me swipe right on it.
Iain
@iainct
Oct 22 2017 03:40
"rate":1.75,"IOB":4.366,"reason":"COB: 62, Dev: 154, BGI: -10.66, ISF: 3.5, Target: 6.4, minPredBG 8.4, minGuardBG 8.1, IOBpredBG 2.2, COBpredBG 7.7, UAMpredBG 9.8; Delta 15 > 10% of BG 147: SMB disabled; temp 1.3<1.75U/hr. "
and the complete view
@danamlewis
Dana Lewis
@danamlewis
Oct 22 2017 03:40
SMB safety check.
Iain
@iainct
Oct 22 2017 03:40
enact/smb-suggested.json: {"insulinReq":0.59,"bg":147,"reservoir":"206.9","temp":"absolute","snoozeBG":279,"predBGs":{"COB":[147,161,173,185,195,204,211,218,224,229,233,235,237,238,239,238,237,234,231,227,222,216,210,204,198,192,187,182,178,173,170,166,162,159,156,154,151,149,147,146,144,143,141,140,139,139,138],"aCOB":[147,146,145,144,142,140,138,136,134,132,130,128,126,124,123,121,120,119,118,117,116,116,115,115,115,115,115,115,116,117,117,118,119,120,121,122,124,125,126,128,129,130,132,133,134,136,137,138],"UAM":[147,161,174,186,196,206,214,222,229,235,240,245,248,251,253,255,256,256,255,254,252,249,246,241,237,231,226,221,217,212,208,205,201,198,195,193,190,188,186,184,183,181,180,179,178,178,177],"IOB":[147,160,170,179,185,189,191,191,189,185,179,171,161,152,143,134,125,117,109,102,95,88,81,75,69,64,58,54,49,45,41,39]},"rate":1.75,"IOB":4.366,"reason":"COB: 62, Dev: 154, BGI: -10.66, ISF: 3.5, Target: 6.4, minPredBG 8.4, minGuardBG 8.1, IOBpredBG 2.2, COBpredBG 7.7, UAMpredBG 9.8; Delta 15 > 10% of BG 147: SMB disabled; temp 1.3<1.75U/hr. ","COB":62,"eventualBG":177,"duration":30,"tick":"+15","deliverAt":"2017-10-22T02:46:26.104Z"}
Dana Lewis
@danamlewis
Oct 22 2017 03:41
Ty.
Updating that to be 20% in future, not sure if it's in latest dev or pending in a Pr
Basically, don't want to deliver too much on a jump that could be fluke or calibration jump
Iain
@iainct
Oct 22 2017 03:42
to clarify the jump needs to exceed 10%
of current bg to apply SMB?
Dana Lewis
@danamlewis
Oct 22 2017 03:43
No. >10% jump = won't SMB. Below that it will.
tynbendad
@tynbendad
Oct 22 2017 03:43
Anyone know what to do here? I upgraded to latest dev branch, from about a month old version and direct receiver read isn't working...

Sat Oct 21 21:36:02 MDT 2017
Iain
@iainct
Oct 22 2017 03:45
@danamlewis ah, so if my bg climbs by greater than >10% for a bit SMB will be disabled?
i had been wondering why it was never kicking in since upgrading to v0.6.0!
tynbendad
@tynbendad
Oct 22 2017 03:45
155 2017-10-21T20:44:28-06:00
Sat Oct 21 21:35:02 MDT 2017
dexcom_reader.constants.CrcError:
jaylagorio
@jaylagorio
Oct 22 2017 03:45
@danamlewis I hate to second guess you, but didn't it get bumped to 20%?
Dana Lewis
@danamlewis
Oct 22 2017 03:45
@iainct :+1: correct. And that number is changing to 20% in future.
@jaylagorio on my phone, youre welcome to confirm which branch and version it's in :smile:
jaylagorio
@jaylagorio
Oct 22 2017 03:46
Ah - not straight dev?
Dana Lewis
@danamlewis
Oct 22 2017 03:46
@iainct if on dev and haven't updated in a few days, I recommend updating and doing so every few days. Lots of changes
tynbendad
@tynbendad
Oct 22 2017 03:46
r: Could not parse EGVRecord
Iain
@iainct
Oct 22 2017 03:46
will do
jaylagorio
@jaylagorio
Oct 22 2017 03:47
It is in dev, committed 5 days ago
Dana Lewis
@danamlewis
Oct 22 2017 03:47
@jaylagorio dunno. Someone can look at the code or PRs if you care to confirm :smile: I can only track 86 or so of the 254 things we are changing at one time ;)
:+1:
@iainct :+1:
jaylagorio
@jaylagorio
Oct 22 2017 03:47
Yup - openaps/oref0@60e6dd5
What, you don't keep every detail about every change to dev in your head when there's like 1000 moving pieces and it changes almost every day? Jeez, what are you even doing over here!
Dana Lewis
@danamlewis
Oct 22 2017 03:49
🤣
Scott Leibrand
@scottleibrand
Oct 22 2017 04:17
@tynbendad I can't read what you've been pasting: it's just a black box
@danamlewis @Bender1061 we're running 0.1.1 still. 0.2.0 doesn't have any improvements for OpenAPS use, but I don't think it has any changes that would cause the symptoms you're seeing either.
tynbendad
@tynbendad
Oct 22 2017 04:25
Sat Oct 21 22:25:02 MDT 2017
dexcom_reader.constants.CrcError: Could not parse EGVRecord
155 2017-10-21T20:44:28-06:00
(sorry, was trying to do it from phone before)
that is from cgm-loop.log
receiver shows the correct bg, just can't read from it
Scott Leibrand
@scottleibrand
Oct 22 2017 04:31
just merged https://github.com/openaps/oref0/pull/622/files to dev for testing: if anyone has any problems with oref0-setup as a result, please speak up as always.
Dana Lewis
@danamlewis
Oct 22 2017 04:44
Thoughts welcome on changing the approach for SMB, given the edge case found tonight, and given the culture that's been developed in the community around SMB. openaps/oref0#737
LilDucky
@LilDucky
Oct 22 2017 04:44
just updated to dev (0.6.0) and getting a few funnies in the log
CGM data retrieval from pump disrupted, retrying in 5 seconds...
Reattempting to retrieve MDT CGM data
MDT CGM data retrieved
MDT New cgm data reformatted
Old settings refresh    at Module.load (module.js:356:32)
ed. Refresh    at Module.load (module.js:356:32)
ed pumphistoryCould not parse input data:  [SyntaxError: Unexpected end of input]
 and meal.json
Checking pump clock: "2017-10-22T15:42:27+11:00" is within 1m of current time: Sun Oct 22 15:43:07 AEDT 2017
and that pumphistory is less than 1m old.  Temp refresh    at Module.load (module.js:356:32)
ed
Could not parse input data:  [SyntaxError: /root/myopenaps/monitor/iob.json: Unexpected end of input]
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 48, in main
    reporters.Reporter(report, device, task)(output)
  File "/usr/local/lib/python2.7/dist-packages/openaps/reports/reporters/__init__.py", line 33, in __call__
    self.blob = self.serialize(data)
  File "/usr/local/lib/python2.7/dist-packages/openaps/reports/reporters/__init__.py", line 31, in serialize
    return self.method.serialize(render(data), self)
  File "/usr/local/lib/python2.7/dist-packages/openaps/vendors/process.py", line 56, in prerender_json
    return json.loads(data)
  File "/usr/lib/python2.7/json/__init__.py", line 338, in loads
    return _default_decoder.decode(s)
  File "/usr/lib/python2.7/json/decoder.py", line 366, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/lib/python2.7/json/decoder.py", line 384, in raw_decode
    raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded
basal-only pump-loop failed. Listening for 40s silence before mmtuning: .
LilDucky
@LilDucky
Oct 22 2017 04:55
iob.json is empty too.
@scottleibrand @danamlewis any thoughts as this is happening every loop? this was a clean install of 0.6.0 after moving the myopenaps dir and removing the src/oref0 directory
Starting basal-only pump-loop at Sun Oct 22 15:55:02 AEDT 2017 with 28 second wait_for_silence:
MDT CGM configured; not waiting
Radio ok. Listening: .No interfering pump comms detected from other rigs (this is a good thing!)
Preflight OK.
Attempting to retrieve MDT CGM data from pump
MDT CGM data retrieved
MDT New cgm data reformatted
Old settings refresh    at Module.load (module.js:356:32)
ed. Refresh    at Module.load (module.js:356:32)
ed pumphistoryCould not parse input data:  [SyntaxError: Unexpected end of input]
 and meal.json
Checking pump clock: "2017-10-22T15:56:06+11:00" is within 1m of current time: Sun Oct 22 15:56:46 AEDT 2017
and that pumphistory is less than 1m old.  Temp refresh    at Module.load (module.js:356:32)
ed
Could not parse input data:  [SyntaxError: /root/myopenaps/monitor/iob.json: Unexpected end of input]
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 48, in main
    reporters.Reporter(report, device, task)(output)
  File "/usr/local/lib/python2.7/dist-packages/openaps/reports/reporters/__init__.py", line 33, in __call__
    self.blob = self.serialize(data)
  File "/usr/local/lib/python2.7/dist-packages/openaps/reports/reporters/__init__.py", line 31, in serialize
    return self.method.serialize(render(data), self)
  File "/usr/local/lib/python2.7/dist-packages/openaps/vendors/process.py", line 56, in prerender_json
    return json.loads(data)
  File "/usr/lib/python2.7/json/__init__.py", line 338, in loads
    return _default_decoder.decode(s)
  File "/usr/lib/python2.7/json/decoder.py", line 366, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/lib/python2.7/json/decoder.py", line 384, in raw_decode
    raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded
basal-only pump-loop failed. pump_loop_completed more than 15m old; waiting for 40s silence before mmtuning
Radio ok. Listening: .
Scott Leibrand
@scottleibrand
Oct 22 2017 05:01
check your preferences.json is valid
LilDucky
@LilDucky
Oct 22 2017 05:02
found a missing comma on the ned of a line.
Scott Leibrand
@scottleibrand
Oct 22 2017 05:02
I think we need to add a better check for that
LilDucky
@LilDucky
Oct 22 2017 05:03
sounds like a common issue
happened after adding in an SMB line
Dana Lewis
@danamlewis
Oct 22 2017 05:03
I volunteer @scottleibrand
:D
@LilDucky see troubleshooting section under openaps/oref0#707
LilDucky
@LilDucky
Oct 22 2017 05:05
looking loads better
tynbendad
@tynbendad
Oct 22 2017 05:15
does dexcom_reader use a database that might be corrupted on my rig?
thinking i should just re-install
Scott Leibrand
@scottleibrand
Oct 22 2017 05:17
which CGM configuration are you using?
tynbendad
@tynbendad
Oct 22 2017 05:17
g5
rx connected via usb
Scott Leibrand
@scottleibrand
Oct 22 2017 05:18
don't think so. reinstall is probably good regardless though
tynbendad
@tynbendad
Oct 22 2017 05:19
ok
Scott Leibrand
@scottleibrand
Oct 22 2017 05:20
@LilDucky @danamlewis openaps/oref0#738
Is that error message clear enough? Or does it need to be more explicit still?
(And no, I don't mean swearing at people for screwing it up.) ;-)
tynbendad
@tynbendad
Oct 22 2017 05:24
ah, might be a node-js version issue...
npm ERR! Linux 3.10.98-jubilinux-edison
npm ERR! argv "/home/.rootfs/usr/bin/node" "/home/.rootfs/usr/bin/npm" "install"
npm ERR! node v6.11.4
npm ERR! npm  v3.10.10
npm ERR! path /root/src/oref0/node_modules/share2nightscout-bridge/node_modules/request/node_modules/hawk/node_modules/hoek
npm ERR! code ENOENT
npm ERR! errno -2
npm ERR! syscall rename

npm ERR! enoent ENOENT: no such file or directory, rename '/root/src/oref0/node_modules/share2nightscout-bridge/node_modules/request/node_modules/hawk/node_modules/hoek' -> '/root/src/oref0/node_modules/coveralls/node_modules/hoek'
npm ERR! enoent ENOENT: no such file or directory, rename '/root/src/oref0/node_modules/share2nightscout-bridge/node_modules/request/node_modules/hawk/node_modules/hoek' -> '/root/src/oref0/node_modules/coveralls/node_modules/hoek'
npm ERR! enoent This is most likely not a problem with npm itself
npm ERR! enoent and is related to npm not being able to find a file.
npm ERR! enoent
Scott Leibrand
@scottleibrand
Oct 22 2017 05:25
why do you have share2nightscout-bridge in ~/src/oref0/ ?
I would rm -rf ~/src/oref0 and re-clone it
tynbendad
@tynbendad
Oct 22 2017 05:25
ok
it may be leftover from a long time ago
Scott Leibrand
@scottleibrand
Oct 22 2017 05:26
yeah, probably. I think I may be the only one still using that package, since it's been built into NS forever. :)
tynbendad
@tynbendad
Oct 22 2017 05:40
that worked, thanks!
i'm getting much faster at reinstalls ;)
Scott Leibrand
@scottleibrand
Oct 22 2017 05:41
:+1: ;-)
LilDucky
@LilDucky
Oct 22 2017 06:47
@scottleibrand @danamlewis that looks good.
LilDucky
@LilDucky
Oct 22 2017 08:23
Has had a meal and entered carbs. Loop logs show microbolus of 0.8 but appears to be running basal only loop not super micro bolus loop.
Dave Acklam
@dcacklam
Oct 22 2017 09:00
Having trouble with a 712 pump & radio comms...
getting -99 for every mmtune.
Did the steps in the setup guide (created extra files, etc)
Pump has a fresh battery...
"mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?"
On an explorer board/edison setup
Velibor Maric
@vebaba
Oct 22 2017 09:03
0.6.0-dev2. I found SMB a bit inefficient (bcs of no basal) if I don't have activity and I want to have it disabled in that period of day. I was thinking of enableSMB_with_bolus set to False and regularly bolusing (to have Basal also kicked in). Is this correct approach? @danamlewis
Velibor Maric
@vebaba
Oct 22 2017 09:13
@dcacklam If that was a new Rig, I have had similar output and the reason was wrong serial # entered in OAPS
Dave Acklam
@dcacklam
Oct 22 2017 09:19
it's new
And the S/N on the back of my pump is 014512-D062
What should I be putting in?
is it just the 014512 part ?
Velibor Maric
@vebaba
Oct 22 2017 09:31
umm, I have 722 and SN part is serial number
Dave Acklam
@dcacklam
Oct 22 2017 09:31
I guess what I'm asking is, the instructions in the setup say 'numbers only'.
Velibor Maric
@vebaba
Oct 22 2017 09:32
So, if this was your pump, the serial would be: 541134
Dave Acklam
@dcacklam
Oct 22 2017 09:32
Ahh, ok...
So mine says 014512-D062 and I should just feed it 014512
(into openaps)
Velibor Maric
@vebaba
Oct 22 2017 09:33
Well 014512-D062 doesn't look like serial number
Dave Acklam
@dcacklam
Oct 22 2017 09:34
712 is a much older pump, so that is probably the diff...
Velibor Maric
@vebaba
Oct 22 2017 09:34
press esc on pump and scroll at the bottom
Dave Acklam
@dcacklam
Oct 22 2017 09:35
says 014512
Velibor Maric
@vebaba
Oct 22 2017 09:35
there should be S/N# line
then, it was correct, sorry
Dave Acklam
@dcacklam
Oct 22 2017 09:36
Yes
At one point I had the full S/N in there, not the one from the menu, so I think I will clear it out and try from scratch now that I know the right nums to use
Velibor Maric
@vebaba
Oct 22 2017 09:37
try that and maybe some of the 'senior' troubleshooter will get online by then :)
kallnap
@kallnap
Oct 22 2017 10:15
just a stupid question for setting up a new rig, is jubilinux 0.2.0 the current way to go?
dmdfreak
@dmdfreak
Oct 22 2017 11:17
Something weird happened about 3 hours ago. NS stopped updating openaps pill, pump pill, and prediction lines. Everything else is updating. Also the pump is looping. But it still feels like it's throwing the whole system off as my numbers and higher than I would expect. I rebooted the azure site but it didn't appear to have had an effect. Also I updated NS yesterday.
Iain
@iainct
Oct 22 2017 11:41
@kallnap for sure it is the most up to date, and should work better than previous version, avoid the non stable version
kallnap
@kallnap
Oct 22 2017 11:41
@iainct thanks, will do :-)
jquitos
@jquitos
Oct 22 2017 12:22
Can someone help with reflashing? I get this in command prompt when trying to
Device has DFU interface, but has no DFU functional descriptor
Cannot open DFU device 8087:0a99
Dfu device not found Timeout
Did you plug and reboot your board?
If yes, please try a recovery by calling this script with the --recovery option
Velibor Maric
@vebaba
Oct 22 2017 16:43
@jquitos you can't flash jubilnux and you get that after reboot?
live4sw
@live4sw
Oct 22 2017 16:53
I think this message got accidentally deleted (but if not sorry for the duplicate), but is there an emerging consensus on DIA for humalog/novolog? I used to see people talking regularly about 3 or 4 hours working well with OpenAPS, but increasingly the blogs (diabettech, seemycgm) seem to have converged on 5 hours. Just asking since I started with 3, moved to 4 a little while back, and am now thinking I should potentially shift to 5 as I've been having some unexplained lows and suspect stacking.
dmdfreak
@dmdfreak
Oct 22 2017 17:30
Anyone else having issues with the updated NS?
Scott Leibrand
@scottleibrand
Oct 22 2017 17:35
@vebaba enableSMB_with_bolus has now been removed and doesn't do anything in the latest dev. even when it did, setting it to false wouldn't disable SMB with boluses, it just wouldn't consider boluses to be a reason to enable it. If you had enableSMB_with_COB, for example, SMB would still get enabled with COB, with or without a bolus.
@kallnap jubilinux 0.2.0 should work fine, but certainly won't work any better than 0.1.1: none of the changes in 0.2.0 are OpenAPS-focused. @Bender1061 has reported some issues with 0.2.0 trying to test the offline branch, but we haven't reproduced those yet or determined conclusively whether they're related to any changes in jubilinux 0.2.0 or not
@live4sw DIA means different things for different insulin curves. with bilinear, 3-4h works best. with exponential curves (with the peak time fixed independent of DIA), nothing shorter than 5h makes sense (to the point where OpenAPS will set your DIA to 5h if you set anything less), and some people have found better behavior with 6-7h (mostly around dealing with the residual tail from the insulin for very large meals)
if you haven't already, I would switch to the appropriate exponential curve, and then consider extending DIA past 5h if you are seeing lows 4-6h after large meals and suspect stacking.
Velibor Maric
@vebaba
Oct 22 2017 17:40
@scottleibrand Thanks for answer. Is there any way cover some meals with SMB and other meals as regular ones (bolus via pump and basal)?
btw. I have to reconsider my findings regard 'SMB being ineffective'. I have had some issue with either infusion set spot, infusion set itself, cable or even insulin.
Scott Leibrand
@scottleibrand
Oct 22 2017 17:40
easiest way to enable and disable SMB on a per-meal basis is with temp targets
a temp target of 101 or higher will disable SMB (regardless of any enableSMB settings). If you have enableSMB_with_temptarget set to true, a temp target of 99 or less will activate SMB.
jquitos
@jquitos
Oct 22 2017 17:43
@vebaba I cannot reflash the rig that I have. It has jubilinux on it already however was trying to reflash because BT is not working correctly. I followed instructions in the docs and on the command prompt in windows that's what I get when I flashall.bat after I reboot the rig
Velibor Maric
@vebaba
Oct 22 2017 17:44
@jquitos It might be that last step isn't passing through (flashing 1.6GB image)
jquitos
@jquitos
Oct 22 2017 17:46
yeah...so that's pretty much where I'm at. Because the rig reboots and just continues to run...
ok... found some other things in docs... let me retry
Velibor Maric
@vebaba
Oct 22 2017 17:48
OK. You need to hassle with Page file size
@scottleibrand OK. Just to confirm: I set temp target above 101, use bolus wizard on my 722, pump gives me insulin, OAPS will leave my basal (no zero temping)?
live4sw
@live4sw
Oct 22 2017 17:53
@scottleibrand The option of using exponential is just starting with 0.6.0 though, right? If we're on the most current master (0.5.5), am I right that bilinear is the only option?
kallnap
@kallnap
Oct 22 2017 18:04
@scottleibrand thanks for furhter clarifying. I have not followed the development of offline-branch closely is there any documentation/discussion, ypu could point me to. I typically do not have reliable online connection. this is why i usually try to run NS on my Edison. Ever since this broke I was essentially using none of the nice features apart from autosense (no eating soon, no smb, ... you name it) consequently I would be happy to learn more about the offline branch and be encouraged to test
Dana Lewis
@danamlewis
Oct 22 2017 18:05
@live4sw correct re master having bilinear only
@vebaba we can't say "yes, it will never low temp". It depends on what your bg is and is predicted to be. But yes for that scenario, with a temp target of 101 and SMB with temp target set to true, it will SMB when needed.
jquitos
@jquitos
Oct 22 2017 18:06

ok... still having problems reflashing and trying to follow "troubleshooting doc". Still getting

Device has DFU interface, but has no DFU functional descriptor
Cannot open DFU device 8087:0a99
Dfu device not found Timeout
Did you plug and reboot your board?
If yes, please try a recovery by calling this script with the --recovery option

Scott Leibrand
@scottleibrand
Oct 22 2017 18:12
@vebaba a temp target >= 101 mg/dL will disable SMB. if OpenAPS thinks you gave too much insulin for the carbs, it might still zero temp to prevent a low.
Dana Lewis
@danamlewis
Oct 22 2017 18:13
^ I meant 99. Sorry.
Scott Leibrand
@scottleibrand
Oct 22 2017 18:14
@live4sw I wouldn't mess with your DIA while using bilinear curves. If you want to correct stacking caused by poor insulin activity curves, I would get a second rig updated to 0.6.0-dev and start testing out the exponential curves and get comfortable with that before making it your production rig.
@kallnap I would read the PR: openaps/oref0#716
Velibor Maric
@vebaba
Oct 22 2017 18:16
@scottleibrand @danamlewis OK, if safety kicks in, that's whole another pair of the shoes :) (I hope that proverb is correct) :) Thanks for clarifying
Eric
@ecc1
Oct 22 2017 18:21
@vebaba that proverb doesn't quite translate (what language is it from?) I'd guess the equivalent would be "another kettle of fish"
Dana Lewis
@danamlewis
Oct 22 2017 18:22
@vebaba :+1: :shoe: :fish: :smile:
Velibor Maric
@vebaba
Oct 22 2017 18:23
@jquitos If you take a look at right terminal screen and compare it with yours, do you see that your flashing process flashed rootfs (the biggest one)?
http://openaps.readthedocs.io/en/latest/_images/mid_flash.png
I have no idea where I get it from. Totally not from my language but I will use 'another kettle of fish' :)
jquitos
@jquitos
Oct 22 2017 18:25
@vebaba i don't get any of that as the flash process does not start at all
in command prompt I type in flashall.bat and the result is as above
in the putty terminal screen after I reboot it just reboots the rig
stop autoboot, type run do_flashand flash process will start
you have to start it on putty terminal, too
jquitos
@jquitos
Oct 22 2017 18:28
@vebaba the funny things is that I tried that... interrupted the boot... run do_flash... and nothing... I'll try it again though
Velibor Maric
@vebaba
Oct 22 2017 18:29
you have to type flashall.bat on PC
jquitos
@jquitos
Oct 22 2017 18:30

Here is where I'm at

Hit any key to stop autoboot: 0
boot >
boot >
boot >
boot > run do_flash
Saving Environment to MMC...
Writing to MMC(0)... done
GADGET DRIVER: usb_dnl_dfu

Velibor Maric
@vebaba
Oct 22 2017 18:30
Also, you put dfu-util.exe and libusb-1.0.dll in jubilinux isntalation directory
?
jquitos
@jquitos
Oct 22 2017 18:31
Device has DFU interface, but has no DFU functional descriptor
Cannot open DFU device 8087:0a99
Dfu device not found Timeout
Did you plug and reboot your board?
If yes, please try a recovery by calling this script with the --recovery option
again
I did put the 2 files in the directory as it stated
Velibor Maric
@vebaba
Oct 22 2017 18:32
so, flashing process from PC won't start?
jquitos
@jquitos
Oct 22 2017 18:33
i guess not... that's the response I get all the time
and it's just hangining in the putty terminal
Velibor Maric
@vebaba
Oct 22 2017 18:36
OK, you open command prompt on your PC, navigate to installing folder of jubilinux on your PC. There, you have those two dfu files. You type in command line flashall.batand in same time you run `run do_flash' on Putty terminal window connected to your Edison and that's all you got?
jquitos
@jquitos
Oct 22 2017 18:37
correct... will retry in a different order I guess and see if it does anything...
@vebaba I rebooted the rig... ran flashall.bat... ran run do_flash.... same thing happened
Eric
@ecc1
Oct 22 2017 18:46
It's been a while since I've flashed an Edison, but I remember that when it said "plug in and reboot", I literally had to reboot by holding down the pushbutton until the light went out, then pressing it again to get the quick flash (and boot).
jquitos
@jquitos
Oct 22 2017 18:50
Thanks @ecc1 ... just tried that... same thing lol
Velibor Maric
@vebaba
Oct 22 2017 18:51
you installed drivers on PC?
jquitos
@jquitos
Oct 22 2017 18:56
yes... i will re check that as well
PieterGit
@PieterGit
Oct 22 2017 19:06
@scottleibrand @sulkaharo : what is the current status of upgrade to node8 for oref0?
The branch https://github.com/openaps/oref0/tree/nodejs-8 and the unit test improvements of openaps/oref0#585 seem not te be updated recently. is there a difficult problem that needs be fixed, or just other priorities and lack of hours in a day
Scott Leibrand
@scottleibrand
Oct 22 2017 19:07
The latter. Node 8 mostly works.
Just a matter of getting all the install stuff ironed out I think.
PieterGit
@PieterGit
Oct 22 2017 19:08
do you still plan to do it in the 0.6.0 release version?
Scott Leibrand
@scottleibrand
Oct 22 2017 19:08
Dunno. We haven't started release planning yet.
Will need it for the Explorer HAT though.
PieterGit
@PieterGit
Oct 22 2017 19:11
is there any way to participate in the Explorer HAT project. when is it expected to create a first batch of Explorer HAT's?
jquitos
@jquitos
Oct 22 2017 19:14
@vebaba i reinstalled drivers just to be safe...rebooted computer... and will try again i guess
Velibor Maric
@vebaba
Oct 22 2017 19:15
Good luck!
PieterGit
@PieterGit
Oct 22 2017 19:16
grrr. spontaneous reboot during bluez compilation
jquitos
@jquitos
Oct 22 2017 19:16
@vebaba FINALLY!!!!! @#$@#$%%^&$^@%$
Velibor Maric
@vebaba
Oct 22 2017 19:17
Dana would say: I like simple solutions :D
Dana Lewis
@danamlewis
Oct 22 2017 19:17
:confetti_ball:
;)
jquitos
@jquitos
Oct 22 2017 19:18
i dunno if it was the reboot that helped...reinstall...trying different usb ports... all i know is it's finally flashing Good LAwdy!
Dana Lewis
@danamlewis
Oct 22 2017 19:19
:+1:
It was the cursing. The final stage of tech troubleshooting ;) that sometimes magically Just Works.
jaylagorio
@jaylagorio
Oct 22 2017 19:20
^ Like a law of physics this is
Dana Lewis
@danamlewis
Oct 22 2017 19:21
@tynbendad are you still having issues with plugged in receiver? And was it a G5?
jquitos
@jquitos
Oct 22 2017 19:21
kinda like hitting it with a hammer? lol
Dana Lewis
@danamlewis
Oct 22 2017 19:21
I think @jaylagorio (?) had done some upload stuff that got merged in, so that might be helping/hurting - but either way, he might be a good brain to pick if you are still stuck.
jaylagorio
@jaylagorio
Oct 22 2017 19:22
I did! oref0-upload-entries in the /usr/local/bin/ directory, it only gets added to your cron if you use g5-upload in your setup script
Scott Leibrand
@scottleibrand
Oct 22 2017 19:23
@PieterGit are you on the openaps-hardware-dev google group?
Dana Lewis
@danamlewis
Oct 22 2017 19:23
By the way - huge :clap: :sparkles: :heart: :+1: to everyone who's been helping with other people's troubleshooting; testing and giving feedback on new code; creating new code and PRing it in; PRing the docs - huge awesome volume of awesomeness from the community the past few weeks. Much appreciated on all fronts!
jquitos
@jquitos
Oct 22 2017 19:25
I can only provide problems because I'm not tech savvy... LOL... but I def appreciate the help!
jquitos
@jquitos
Oct 22 2017 19:40
ok...so i'm a little confused when setting up the wifi portion in putty... is this what the file is supposed to look like? I deleted the wpa and psk lines

interfaces(5) file used by ifup(8) and ifdown(8)

auto lo
iface lo inet loopback

auto usb0
iface usb0 inet static
address 192.168.2.15
netmask 255.255.255.0

auto wlan0
iface wlan0 inet dhcp
wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf

# For WPA
# For WEP
#wireless-essid Emutex
#wireless-mode Managed
#wireless-key s:password

And the following 4 lines are for when using hostapd...

auto wlan0

iface wlan0 inet static

address 192.168.42.1

netmask 255.255.255.0

-- INSERT --

or is it only supposed to look like this :

auto lo
iface lo inet loopback

auto usb0
iface usb0 inet static
address 192.168.2.15
netmask 255.255.255.0

auto wlan0
iface wlan0 inet dhcp
wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf

Velibor Maric
@vebaba
Oct 22 2017 19:42
2nd one
jquitos
@jquitos
Oct 22 2017 19:43
ok...so delete all the other stuff out
Velibor Maric
@vebaba
Oct 22 2017 19:43
did you installed jubilinux? I installed ubilinux at first
jquitos
@jquitos
Oct 22 2017 19:43
yes... jubilinux-v0.2.0
Dana Lewis
@danamlewis
Oct 22 2017 19:51
Hehe. Forgot to turn my hotspot on. Realized it 10 min on the road, looked and it spun up a hotspot quite nicely :) for me to log on and see what it's doing. This will be soooo nice. Not just for planes :smile:
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:01
Hi everyone, happy Sunday!
Screen Shot 2017-10-22 at 3.58.01 PM.png
I am noticing a trend of what seems like overaggressive SMB tht results in a low later on..
Scott Leibrand
@scottleibrand
Oct 22 2017 20:02
@jquitos everything you learn is something you can help the next newbie with. You don't have to start out "tech savvy" to pay it forward.
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:02
Autosense aready has me up at a ISF of 150
I am using dev, and have eSMB at 60 minutes
jquitos
@jquitos
Oct 22 2017 20:03
@scottleibrand i understand Scott... definitely learning as much as I can and trying to retain as well... this is worse than Spanish II! LOL
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:03
These overly aggressive corrections tend to happen over an hour after the initial carb entry
Scott Leibrand
@scottleibrand
Oct 22 2017 20:03
What does the OpenAPS pill or pump-loop.log day at the time of the SMB?
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:04
Screen Shot 2017-10-22 at 4.04.31 PM.png
Dana Lewis
@danamlewis
Oct 22 2017 20:05
Pam, when is the last time you updated dev?
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:06
maybe 2 days ago?
maybe 3?
Scott Leibrand
@scottleibrand
Oct 22 2017 20:07
do you think you still had 9 COB at that point?
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:07
it was almost 2 hours post carb entry.. so no!
Scott Leibrand
@scottleibrand
Oct 22 2017 20:09
ok, then your carbs may have been overestimated, or your CR too low
Velibor Maric
@vebaba
Oct 22 2017 20:09
OK, I have had an issue today with my infusion set or insulin. To make long story short: is there anything that I can do to exclude that data from Autotune and Autosense calculations?
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:10
I started with 10 g around 12:30, are you sure it is the cr?
seems like only 1 g decay over almost 2 hours seems a bit low
Velibor Maric
@vebaba
Oct 22 2017 20:12
@PopcornPam_twitter Sorry to interupt. On NS site, 'profile editor'. There is 'Carbs activity / absorption rate: [g/hour]'
Dana Lewis
@danamlewis
Oct 22 2017 20:13
@vebaba nope, that's not used in OpenAPS
You can ignore that in NS
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:13
I am wondering if something wonky is going on with how it is calculating carb absorbtion
Velibor Maric
@vebaba
Oct 22 2017 20:14
But where do you enter CR? @danamlewis
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:14
cr= carb ratio.. in your pump
Scott Leibrand
@scottleibrand
Oct 22 2017 20:14
@PopcornPam_twitter you had 36g of stacked carbs
Velibor Maric
@vebaba
Oct 22 2017 20:15
@PopcornPam_twitter Ah, OK... I mixed two terms...
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:16
but 20 of them were from around noon, I thought they would be mostly gone 2 hours later
currently I have it set at 17, and autotune is recommeding 14
Scott Leibrand
@scottleibrand
Oct 22 2017 20:17
do you have min_5m_carbimpact defined in your preferences.json, or are you using the default?
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:17
default
Scott Leibrand
@scottleibrand
Oct 22 2017 20:17
also, if you want to send me your NS URL I can do some other digging
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:18
worth noting that this was after 4 hours of teaching circus (aka pretty active)
The exercise temp was no longer going, but I do run into issues with post exercise
it almost seems like it needs a way to change the carb absorbtion for the few hours post exercise
However, admittedly, only when it has been an extended exercise time, versus just an hour long workout
Scott Leibrand
@scottleibrand
Oct 22 2017 20:22
so your reservoir readings in your pump pill go from 3.9U to 0.6U, so 3.3U delivered from noon to 4pm, vs. a normal basal of 0.85U/h, which over 4h would deliver 3.4U
in other words, OpenAPS gave you 0.1U for those 36g of carbs
so yes, this is definitely a post-exercise sensitivity thing. we actually have some new tools to help with that
John Benjamin
@Bender1061
Oct 22 2017 20:30
@scottleibrand my rig is now messed up with connecting to wifi, I posted the details in the Offline discussion. I'm wondering if you've added something more to the rig while testing, that did not get installed in the branch.
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:34
I have the exercise feature turned on.. but I am not sure about how it addresses once the exercise mode is turned off
Scott Leibrand
@scottleibrand
Oct 22 2017 20:34
@Bender1061 not that I know of. I installed a new rig from the branch, and it's working fine (Dana forgot to hotpot on the drive to brunch, and it spun up its OpenAPS wifi hotspot normally)
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:34
But is it time to update my dev regardless?
Scott Leibrand
@scottleibrand
Oct 22 2017 20:35
@PopcornPam_twitter was autosens detecting sensitivity at the time?
@PopcornPam_twitter the new improvements that would help in that case are in the activity-SMB branch if you want to test that
openaps/oref0#734
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:37
mmy pump is ISF at 110, autosense took it to 149
I wonder if I need to increase the autosense min and max
Scott Leibrand
@scottleibrand
Oct 22 2017 20:38
k. with the improvements in that branch, it would respond more strongly to an ISF increase that large
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:38
I also will often have a hard time coming down from fatty meals
John Benjamin
@Bender1061
Oct 22 2017 20:38
@scottleibrand I wonder what you guys are doing different. Cause clean installed and even clean flashed install and I'm not doing anything stranger than normal. but something in that branch is breaking the DHCP drivers
Scott Leibrand
@scottleibrand
Oct 22 2017 20:39
have you tried with jubilinux 0.1.1?
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:40
is this an add activity-smb as a line in preferences, and set to true?
Scott Leibrand
@scottleibrand
Oct 22 2017 20:40
no
you'd have to cd ~/src/oref0 && git fetch && git checkout activity-smb && git pull to install that branch
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:41
ok, I'll give it a shot!
John Benjamin
@Bender1061
Oct 22 2017 20:41
@scottleibrand I've tried 0.0 0.1.1 and 0.2
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:43
error: pathspec 'activity-smb' did not match any file(s) known to git.
Dana Lewis
@danamlewis
Oct 22 2017 20:44
Try a git pull first
I think that's what Brian did, then the other line, and that got it
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:44
just "git pull"?
Dana Lewis
@danamlewis
Oct 22 2017 20:44
Yes
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:46
git stash, and git pull successful, but getting the same error
root@myopenaps2:~/src/oref0# cd ~/src/oref0 && git fetch && git checkout activity-smb && git pull
error: pathspec 'activity-smb' did not match any file(s) known to git.
Scott Leibrand
@scottleibrand
Oct 22 2017 20:48
sorry, case-sensitive
PopcornPam
@PopcornPam_twitter
Oct 22 2017 20:48
ah, ya, just figured that out
Scott Leibrand
@scottleibrand
Oct 22 2017 20:48
cd ~/src/oref0 && git fetch && git checkout activity-SMB && git pull
Dave Acklam
@dcacklam
Oct 22 2017 20:55
does anyone have a working source for the latest jubilunux zipfile?
The original site doesn't have any downloads that work & I need to reflash my edison
John Benjamin
@Bender1061
Oct 22 2017 21:00
@scottleibrand I think what I'm going to do is reflash my edison. And then install openaps, then I'm going to manually install any packages into the rig that would have been added by the offline branch and then after every update reboot and see if I can find what it is that breaking my rigs.
Velibor Maric
@vebaba
Oct 22 2017 21:05
@dcacklam 0.2.0? Direct link still works http://www.jubilinux.org/dist/jubilinux-v0.2.0.zip
Dave Acklam
@dcacklam
Oct 22 2017 21:07
Thanks
I am still fighting with the 0, -99 issue from last night
About to give up on the edison & try a Pi/Carelink setup (Since I have a spare Pi around & the pump came with a carelink stick)
Dana Lewis
@danamlewis
Oct 22 2017 21:10
@dcacklam did you quadruple check pump sn in setup? Are you getting any other errors? Has this rig/pump successfully looped before?
kallnap
@kallnap
Oct 22 2017 21:17
@scottleibrand thank you for the link to openaps/oref0#716, so far this looks like a bug fix for Iphones to me. I get the impression that the more interesting feature (at least for me) of being able to cut out NS from the loop equation and still use all features of OpenAPs is still work in progress for a later stage. Unfortunately I cannot help with any REST or basic Webserver running on the Edison, this is why I try to get NS running on my Edison so I can stay completely offline and use the latest bells and whistles from your algorithm.
philipgo
@philipgo
Oct 22 2017 21:25
How often do others need to use ccprog to reflash the radio chip on the Explorer Boards? We have 2 Explorer Boards, both less than 3 months old, and both had to be reflashed twice already. Is that normal or is that an indication that something is wrong with them? Do we know what causes the radio chip to lose its firmware?
LilDucky
@LilDucky
Oct 22 2017 21:33
@drnoname82 I have had to reflash 2 or 3 times. My initial gut feel is that the trend appears to be if the rig overheats then the CC1110 is much more likely to dump the flash.
My rig case is quite compact and well enclosed to protect the edison from water splashes etc (T1D is my daughter's BF that works in the water supply industry)
I also preload the ccprog and hex files
philipgo
@philipgo
Oct 22 2017 21:36
Hm, ours never get extremely hot, at least not that I have noticed. Yep, I also started to preload ccprog and the files
LilDucky
@LilDucky
Oct 22 2017 21:38
@scottleibrand @danamlewis Is it worthwhile to preload ccprog and the hex file as well as a short script so that a reflash is possible on the fly without internet access?
Dave Acklam
@dcacklam
Oct 22 2017 21:47
@danamlewis
This is a brand new setup. 712 pump & explorer board.
I just started setting it up last night, and I could not get it to communicate with the pump.
I initially had the S/N wrong, but that was fixed last night. I've tried both master & dev branch, and still no comms
0,-99 from mmtune every time it runs
Velibor Maric
@vebaba
Oct 22 2017 21:54
@dcacklam And you got correctly US and WW step during setup?
tynbendad
@tynbendad
Oct 22 2017 21:57
@danamlewis no problem anymore - Scott helped me last night, i just had some old share directory in src/oref0 that i needed to obliterate ;)
then the install would finish properly
but yes, g5
Dave Acklam
@dcacklam
Oct 22 2017 21:59
@vebaba Yes, I have an NA pump
MMT-712NA
Velibor Maric
@vebaba
Oct 22 2017 22:06
Does it work with Carelink stick? That Edison board must be coming with some warranty. Maybe to try some other pump if you have someone to borrow it from. Maybe someone is near you and willing to help...
Dave Acklam
@dcacklam
Oct 22 2017 22:09
I haven't been able to get the edison/jubi build to work with the carelink
Was going to try it on a raspberry pi
Dave Acklam
@dcacklam
Oct 22 2017 22:22
Would also appriciate some help with the edison if anyone has any ideas
LilDucky
@LilDucky
Oct 22 2017 22:25
@dcacklam carelink on a Pi works OK when I tested it a couple of weeks ago. Was on a Pi B. I took the edison scripts and removed a few bits that were not needed (mainly the edison user and password aging bits)
I have used the edison with the explorer and also an old Ciseco ERF and setting them up is a doddle.
Marco
@CaptainBalou
Oct 22 2017 22:29
I am living in Germany. We have summer and winter time zone. What happens in OpenAPS-rig with Medtronic setup at time change? I think I’ll get lost in the middle of the night if I don’t do things manually isn’t it?
jquitos
@jquitos
Oct 22 2017 22:35

another question while going through set up... :

enable log compression by removing the hash on the #compress line, to reduce the probability of running out of disk space

Do I delete just that line or everything below:

see "man logrotate" for details

rotate log files weekly

daily

keep 4 weeks worth of backlogs

rotate 4

create new (empty) log files after rotating old ones

create

uncomment this if you want your log files compressed

compress

packages drop log rotation information into this directory

include /etc/logrotate.d

no packages own wtmp, or btmp -- we'll rotate them here

/var/log/wtmp {
missingok
monthly
create 0664 root utmp
rotate 1
}

jquitos
@jquitos
Oct 22 2017 22:46
never mind... got it LOL. After reflashing just go through normal oref0 setup?
Dana Lewis
@danamlewis
Oct 22 2017 22:48
@jquitos yup.
Scott Leibrand
@scottleibrand
Oct 22 2017 23:18
@Bender1061 sounds like a plan. thanks for your testing on this: I hope you don't feel like this is something you "have to" do: you can always re-flash and go back to dev and let someone else work on reproducing and fixing the issue.
katie disimone
@Kdisimone
Oct 22 2017 23:20

Have a replacement board from Hamshield that I just started to use…alas, not working. So, Just to start fresh and clean, I reflashed edison, installed a new loop, and got the subs_rfspy error in pump loop. So, I stopped cron, killall -g oref0-pump-loop, cd myopenaps and did openaps mmtune. Still same error. I did the ccprog step from the logs…no dice either.

During the installation of the new loop, i did get this message at the end, before cron start question

Screen Shot 2017-10-22 at 2.50.44 PM.png
and then after all those troubleshooting steps, here’s what I get
Scott Leibrand
@scottleibrand
Oct 22 2017 23:20
@drnoname82 @LilDucky heat is as good a theory as any, but really we don't know. there seem to be some people who need to do it a fair bit, and others that have never had to. could be hardware, or something to do with a usage corner case, or who knows. I do like the idea of making it easier to install and run ccprog, if any of you want to work on automating that.
katie disimone
@Kdisimone
Oct 22 2017 23:21
Screen Shot 2017-10-22 at 3.03.45 PM.png
Scott Leibrand
@scottleibrand
Oct 22 2017 23:22
@Kdisimone what does your ccprog output look like?
katie disimone
@Kdisimone
Oct 22 2017 23:22
I compared the files listed in the first screenshot error message, with a working rig…no issues with the files themselves. They match.
Scott Leibrand
@scottleibrand
Oct 22 2017 23:23
I would also try unseating and reseating the Edison from/to the Explorer Block.
katie disimone
@Kdisimone
Oct 22 2017 23:23
@scottleibrand you mean when I flashed the radio chip?
Scott Leibrand
@scottleibrand
Oct 22 2017 23:23
yeah
and also, what are the cc1110 LEDs doing?
katie disimone
@Kdisimone
Oct 22 2017 23:24
LEDs are the normal two reds (it is plugged into charge)
Screen Shot 2017-10-22 at 4.23.38 PM.png
Scott Leibrand
@scottleibrand
Oct 22 2017 23:24
do the cc1110 LEDs flash briefly when you mmtune?
or after running the ccprog write?
you can also try reset_spi_serial directly (that's what mmtune does first)
katie disimone
@Kdisimone
Oct 22 2017 23:25
yes, they flash with mmtune briefly
root@hondo:~/myopenaps# bash reset_spi_serial.py
/usr/local/bin/reset_spi_serial.py: line 2: import: command not found
/usr/local/bin/reset_spi_serial.py: reset_spi_serial.py: line 3: syntax error near unexpected token `('
/usr/local/bin/reset_spi_serial.py: reset_spi_serial.py: line 3: `s = spi_serial.SpiSerial()’
Scott Leibrand
@scottleibrand
Oct 22 2017 23:26
ok. I would try removing your crontab, completely powering down the rig and turning it back on, re-doing the ccprog erase then write, and then try another mmtune.
that's a python command not a bash script
so just reset_spi_serial.py
katie disimone
@Kdisimone
Oct 22 2017 23:27
root@hondo:~/myopenaps# reset_spi_serial
-bash: reset_spi_serial: command not found
I tried that originally
Scott Leibrand
@scottleibrand
Oct 22 2017 23:27
k one sec
try it with the .py but without the bash: reset_spi_serial.py
I'm so used to tab completing I don't notice the stuff at the end
so for me it's really reset_spi<tab>
katie disimone
@Kdisimone
Oct 22 2017 23:28
no returned message…went back to prompt
LilDucky
@LilDucky
Oct 22 2017 23:28
@scottleibrand I am not really a good enough shell script person to do it. I can confirm that heat is definitely a factor - my initial temp case was totally sealed and it dumped the flash twice in rapid succession until I changed the case.
katie disimone
@Kdisimone
Oct 22 2017 23:29
root@hondo:~/myopenaps# reset_spi_serial.py
root@hondo:~/myopenaps#
Scott Leibrand
@scottleibrand
Oct 22 2017 23:29
ok, that's normal. it should just blink the LEDs when you do that
katie disimone
@Kdisimone
Oct 22 2017 23:29
it does
Scott Leibrand
@scottleibrand
Oct 22 2017 23:29
@LilDucky shell scripting doesn't require much skill, mostly just knowing what you want to do
if you can type out a series of commands that you can run interactively to accomplish what you want, then you can put those into one of the bash scripts pretty much as-is
then it's just a matter of finding the right time to do those things, if you want it done in something like oref0-setup
@Kdisimone ok, then try removing your crontab, completely powering down the rig and turning it back on, re-doing the ccprog erase then write, and then try another mmtune.
if re-doing those steps doesn't fix it, we'll probably be swapping hardware to see which piece the problem follows
katie disimone
@Kdisimone
Oct 22 2017 23:33
error remains
Scott Leibrand
@scottleibrand
Oct 22 2017 23:35
ok, then I'd swap that Explorer Block to a different Edison and see whether the problem follows the Explorer Block or stays with the Edison
katie disimone
@Kdisimone
Oct 22 2017 23:37
ohhh…unexpected…followed the edison. But, let me try different edison on that old board
two lights staying on thoug
hmmmm
Scott Leibrand
@scottleibrand
Oct 22 2017 23:39
if the two red+green lights stay on means it doesn't have working cc1110 firmware and needs another ccprog
(if they stay on after an mmtune that is)
katie disimone
@Kdisimone
Oct 22 2017 23:40
no green
Scott Leibrand
@scottleibrand
Oct 22 2017 23:40
they changed the color of one of the two cc1110 LEDs at some point, so depends on how old the Explorer Block is
but the two LEDs right next to each other are the ones of interest, whichever color they are
katie disimone
@Kdisimone
Oct 22 2017 23:41
This is weird…the rig that has been problematic (the edison is named Hondo)…hondo on the explorer board #2 is giving two solid red lights down low (not the power/loop lights). But when I did cd ccprog; make ccprog it says the directory doesn’t exist
Doesn’t that directory travel with the edison?
Scott Leibrand
@scottleibrand
Oct 22 2017 23:41
yes, but hondo may never had had to do a ccprog
katie disimone
@Kdisimone
Oct 22 2017 23:41
it did though…didn’t we do that on the previous steps?
Scott Leibrand
@scottleibrand
Oct 22 2017 23:42
the edisons have names, the EBs don't
which Edison were you doing the flashing from?
katie disimone
@Kdisimone
Oct 22 2017 23:42
Hondo
LilDucky
@LilDucky
Oct 22 2017 23:42
Have a really close look under a magnifier at the edison connector and its mate on the explorer - I have seen one damaged and also had some fluff in one that stopped it seating correctly.
Scott Leibrand
@scottleibrand
Oct 22 2017 23:42
try a locate ccprog to see if you put it somewhere odd
it doesn't much matter where, as long as you can find it :)
katie disimone
@Kdisimone
Oct 22 2017 23:43
nada
Scott Leibrand
@scottleibrand
Oct 22 2017 23:43
updatedb and try again
you can also check history to see if that's the rig you think it was
(to make sure that's the one you were running the ccprog commands on)
katie disimone
@Kdisimone
Oct 22 2017 23:45
found it now
reprogrammed, lights off, trying mmtune now
works
Scott Leibrand
@scottleibrand
Oct 22 2017 23:46
:+1:
so is there another hardware combo that's still not working?
katie disimone
@Kdisimone
Oct 22 2017 23:46
I’m trying Pongo (working edison on the suspect explorer board now). Pongo was working on another EB just a few min ago...
Hey, Pongo is working on suspect EB.
Ok…let’s try Hondo on original EB…the original fail point.
Scott Leibrand
@scottleibrand
Oct 22 2017 23:48
could also be something to do with the connection like @LilDucky mentioned
katie disimone
@Kdisimone
Oct 22 2017 23:51
no go
both little lights…let me flash
working.
what the what?! so strange. but, thank you for walking me through!
now to rebuild cron
Dana Lewis
@danamlewis
Oct 22 2017 23:58
:+1: