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

13th
Jan 2018
Samuchco
@Samuchco
Jan 13 2018 02:26
@tim2000s what parameters/numbers did you put in autotune and autosens?
Deweyoxberg
@Deweyoxberg
Jan 13 2018 06:29
Need help please. Keep getting communication timeouts between rig and pump. They're literally next to each other within 1 inch and still getting timed: out or other comms error - Received an error response Timeout - retrying: 3 of 3 Short of new hardware for both board and explorer block, anything else we can do? It's preventing successful loops.
Dana Lewis
@danamlewis
Jan 13 2018 06:44
Is the edison screwed on well to the explorer board?
jzycom123
@jzycom123
Jan 13 2018 07:15
1月12日央视新闻直播间报道了一则“老糖友”发明“人工胰腺”一则新闻,画面出现一张熟悉的面孔-Dana,这段时间一直在构建她的闭环系统,大学的经历告诉我科技的重要性,转化的重要性,医工合作的典范!继续加油!http://tv.cctv.com/v/v1/VIDEU92bM3cgvE5GBwViEG4w180112.html#
Deweyoxberg
@Deweyoxberg
Jan 13 2018 07:22
@danamlewis : yes.
Ebgineer
@Ebgineer
Jan 13 2018 08:19
@jzycom123 @danamlewis She's famous :smile:
jzycom123
@jzycom123
Jan 13 2018 09:13
是的
kallnap
@kallnap
Jan 13 2018 09:20
Can I safely run autotune across several NS profiles? I switched profiles recently and now I am not sure if running autotune across profiles would make sense. Sorry if this question has been answered before
jzycom123
@jzycom123
Jan 13 2018 14:25
@danamlewis @Ebgineer @scottleibrand Hello everyone, please help me check is not a problem CGM?
root@tangyouyingbeng:~/myopenaps-cgm-loop# openaps use cgm iter_glucose 1
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 45, in call
output = self.main(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps/vendors/dexcom.py", line 418, in main
candidates = itertools.takewhile(self.comparison, self.dexcom.iter_records(self.RECORD_TYPE))
AttributeError: 'NoneType' object has no attribute 'iter_records'
Starting oref0-pump-loop at Sat Jan 13 22:20:32 HKT 2018 with 24 second wait_for_silence:
Waiting up to 4 minutes for new BG: ........................Radio ok. Listening: .No interfering pump comms detected from other rigs (this is a good thing!)
Preflight OK. Profile less than 60m old; Profile valid. RefreshValueError: need more than 0 values to unpack
Retrying invoke_pumphistory_etc
IndexError: bytearray index out of range
Couldn't invoke_pumphistory_etc - continuing
Retrying monitor_pump
IndexError: bytearray index out of range
Retrying invoke_pumphistory_etc
ValueError: need more than 0 values to unpack
Couldn't invoke_pumphistory_etc - continuing
Couldn't monitor_pump - continuing
Couldn't refresh_pumphistory_and_meal
oref0-pump-loop failed. Unsuccessful oref0-pump-loop at Sat Jan 13 22:26:10 HKT 2018
Ebgineer
@Ebgineer
Jan 13 2018 14:29
@jzycom123 there are two things that make it seem as if you are not getting a good connection to the CGM.
One is that you get only two lines returned from lsusb
lsusb
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
There should be a third line corresponding to the CGM
Here's what mine looks like. I am also using G4-Upload
lsusb
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 002: ID 22a3:0047
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
The other part I'm not sure of, but in your error log you get
AttributeError: 'NoneType' object has no attribute 'iter_records'
jzycom123
@jzycom123
Jan 13 2018 14:32
@Ebgineer Is there any problem with my G4? Or my connection problems?
Ebgineer
@Ebgineer
Jan 13 2018 14:32
Possibly means there are no records returned from the CGM
I have had both problems happen. One was a bad CGM and the other a bad cable. The results were similar.
jzycom123
@jzycom123
Jan 13 2018 14:33
CGM can sometimes connect, and sometimes can not connect, I do not know what to do
Ebgineer
@Ebgineer
Jan 13 2018 14:34
Do you get the intermittent connection to the CGM using different cables?
jzycom123
@jzycom123
Jan 13 2018 14:35
No, I always use the same line to connect G4
No, I always use the same line to connect G4
Ebgineer
@Ebgineer
Jan 13 2018 14:36
The easier thing to try is to use different OTG USB cables. If you get the same result with several cables, it is most likely the CGM
jzycom123
@jzycom123
Jan 13 2018 14:37
Yes, I try to change lines
Ebgineer
@Ebgineer
Jan 13 2018 14:41
What do you get when you try this:
grep mmtune /var/log/openaps/pump-loop.log
jzycom123
@jzycom123
Jan 13 2018 14:48
mmtune: "868.486", 5, -62 waiting for 4 second silence before continuing
mmtune: "868.414", 5, -77 waiting for 34 second silence before continuing
mmtune: "868.414", 5, -71 waiting for 22 second silence before continuing
mmtune: "868.390", 5, -71 waiting for 22 second silence before continuing
mmtune: No wait required.
mmtune: No wait required.
mmtune: No wait required.
mmtune: No wait required.
mmtune: "868.486", 5, -62 waiting for 4 second silence before continuing
mmtune: "868.462", 5, -54 No wait required.
mmtune: "868.390", 5, -55 No wait required.
mmtune: "868.438", 5, -60 No wait required.
mmtune: "868.582", 3, -80 waiting for 40 second silence before continuing
mmtune: "868.414", 5, -72 waiting for 24 second silence before continuing
mmtune: "868.462", 5, -76 waiting for 32 second silence before continuing
mmtune: No wait required.
root@tangyouyingbeng:~# cd ~/myopenaps-cgm-loop
root@tangyouyingbeng:~/myopenaps-cgm-loop# openaps use cgm iter_glucose 1
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 45, in call
output = self.main(args, app)
File "/usr/local/lib/python2.7/dist-packages/openaps/vendors/dexcom.py", line 418, in main
candidates = itertools.takewhile(self.comparison, self.dexcom.iter_records(self.RECORD_TYPE))
AttributeError: 'NoneType' object has no attribute 'iter_records'
Tim Street
@tim2000s
Jan 13 2018 14:57
@Samuchco - you’d need to decide what you’re comfortable with based on your own observations.
Samuchco
@Samuchco
Jan 13 2018 14:59
@tim2000s Thanks, I wanted to observe your obvervations , LOL
Ebgineer
@Ebgineer
Jan 13 2018 15:00
@jzycom123 Did you get the same result when using different cables?
Tim Street
@tim2000s
Jan 13 2018 15:08
Sorry @Samuchco - has to be a case of do it yourself, for obvious reasons 😉
jzycom123
@jzycom123
Jan 13 2018 15:09
@Ebgineer Yes, still the same result
Ebgineer
@Ebgineer
Jan 13 2018 15:10
I found that I had to reboot each time I switched cables. The USB driver wouldn't recover once it got an error -99.
Steps I used:
  1. Change cables
  2. Reboot rig
  3. Watch for charging indicator on CGM receiver during reboot
  4. Log in to rig
  5. lsusb
  6. If it doesn't return 3 text lines like this
    Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub Bus 001 Device 002: ID 22a3:0047 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
    Then you may have a bad USB port on the CGM.
jzycom123
@jzycom123
Jan 13 2018 15:11
But I found that edison is connected to the pump and that the NS can receive a large dose of pump information
Ebgineer
@Ebgineer
Jan 13 2018 15:12
The rig gets the pump info via radio and sends it to NS. The rig has to get the glucose data from the USB cable. It isn't getting it, so it can't send it to NS.
That most likely means the rig is ok but the USB port on the CGM is not working
Less likely it could be a USB port on the Explorer board is bad.
jzycom123
@jzycom123
Jan 13 2018 15:15
@Ebgineer OK, I'll buy an OTG again
Ebgineer
@Ebgineer
Jan 13 2018 15:15
If the Explorer board had a bad USB port I believe you would not get the two good text lines returned from lsusb
The OTG cables are not too expensive, so buying another cable is easiest. In my case the CGM receiver was bad and I had to buy a new CGM receiver.
jzycom123
@jzycom123
Jan 13 2018 15:16
That is possible there is a problem with the port on the receiver, I also heard that the port is easy to bad
Ebgineer
@Ebgineer
Jan 13 2018 15:16
My Dexcom receiver was very old, so either it never worked or the USB port on the receiver had gone bad.
jzycom123
@jzycom123
Jan 13 2018 15:18
The receiver's job is normal
My Dexcom receiver is working fine
Ebgineer
@Ebgineer
Jan 13 2018 15:19
Normally the receiver doesn't use the USB port except for charging. When we use it to get data it is doing more.
Samuchco
@Samuchco
Jan 13 2018 15:20
@tim2000s PMed you
jzycom123
@jzycom123
Jan 13 2018 15:20
@Ebgineer 非常感谢您!
Ebgineer
@Ebgineer
Jan 13 2018 15:21
我希望这有帮助。我不是这个领域的专家,只是分享我的经验。
jzycom123
@jzycom123
Jan 13 2018 15:22
@Ebgineer ok!!
Luis Betancourt
@betluis
Jan 13 2018 19:01
Hello all... question. I deleted all documents in the mlab for "entries"... now Nightscout is not getting any Glucose points. I run ns-loop in the rig but the entries database remains empty and no Glucose is seen in Night Scout. We use Medtronic sensor
Dana Lewis
@danamlewis
Jan 13 2018 19:02
OpenAPS first-upload?
Luis Betancourt
@betluis
Jan 13 2018 19:02
I remember Scott used to have a command for first entry to nighstcout when using medtronic, not sure if this is still the case and I have to run that but I can't find this in thedocumentation.. any help?
right
Dana Lewis
@danamlewis
Jan 13 2018 19:02
^
Luis Betancourt
@betluis
Jan 13 2018 19:02
that's what I was refering to
I can't find it...
ahh, that's a command in itself
I see, I'll try
Luis Betancourt
@betluis
Jan 13 2018 19:14
@danamlewis thank you!
Dana Lewis
@danamlewis
Jan 13 2018 19:28
:+1:
Also, please consider donating data to the Data Commons if you haven't :smile: for anyone who thinks about deleting data
Thacher
@thacherT1D
Jan 13 2018 19:40
@jburke41 I ran the bluetooth setup and the one thing I remembered is that sometimes it helps to turn the wifi on your phone off when connecting the rig with bluetooth because if you're on a wifi network sometimes it will kick off any devices on your hotspot... other than that the script ran for me
WeedsDIY
@WeedsDIY
Jan 13 2018 20:53
Having problem with rig. after some chats, I believe the problem is the rig memory is full due to log files. Have followed some instructions in documentation and have discovered the /var/log/ directory has consumed 1.1G of the 1.4G of memory. Further investigation shows that there are >5 filenames that are repeated a high number of times, but differing by a .#.gz. I recognize all these files are compressed. There is a single file by the same name, but uncompressed and without a number suffix. For example; messages (file), messages.1 (another file), messages.2-23.gz (several compressed files).
The compressed files tend to be 1/2 - 3/4 Mb in size. The .1 file ~11Mb. What I believe to be the main file messages = 242Mb. The documentation talks about using logrotate to help manage log files. I see that we have logrotate.conf on the rig, so I believe it is there. Probably need to alter the .conf file settings. Before I make changes I want to verify what I am assuming so as to not cause greater problems.
philipgo
@philipgo
Jan 13 2018 20:58
@WeedsDIY The gz files are the archived and compressed log files. To free up memory you can do do rm *.gz in /var/log. After the next update OpenAPS will keep less log files due to a recent change.
WeedsDIY
@WeedsDIY
Jan 13 2018 20:59
It appears to me that the primary log files; auth.log, messages, syslog, daemon, user.log, once they reach some limiting variable or log rotate deadline(?) the file is renamed and compressed. But the compressed files are never removed. Am I correct in assuming the compressed files (.gz) can all be deleted simply from the command line?
I believe the excessively large file sizes seen on the main files (message = 242Mb, for example) could be reduced by rotating files more frequently than weekly. Is this correct? I believe I know why after months of not having this problem, the main files are suddenly growing so much larger than historically typical and am working on that separately.
philipgo
@philipgo
Jan 13 2018 21:02
@WeedsDIY See openaps/oref0#901
WeedsDIY
@WeedsDIY
Jan 13 2018 21:48
@drnoname82 Thanks, that permitted loop to resume. How much free space is normal? Still have 90% full after rm *.gz. Is this a problem?
kallnap
@kallnap
Jan 13 2018 22:14
@WeedsDIY sudo apt-get clean usually clears a small amount of space as well
philipgo
@philipgo
Jan 13 2018 22:17
@WeedsDIY I have 55% in use. You can either install the latest dev or manually change your logrote configuration as done in the PR.
WeedsDIY
@WeedsDIY
Jan 13 2018 22:23
@drnoname82 @kallnap Thanks, will try those options. At least I know how much free space would be more "normal" now. Thanks so much for the feedback!
Giuseppe
@giupo
Jan 13 2018 22:36
Was wondering if any of you has ever thought about building an insulin pump, what kind of hurdles has found, are there some docs... just wondering...
Dana Lewis
@danamlewis
Jan 13 2018 22:37
several have thought about it. but because it is a bad idea (the necessary safety engineering to make sure it doses exactly what it says it doses), it's not something the community supports.
Giuseppe
@giupo
Jan 13 2018 22:38
Hi @danamlewis !! So basically the main problem is being sure to dose the correct volume of insulin... mhhmm ..
Dana Lewis
@danamlewis
Jan 13 2018 22:38
also, beware of anyone who tries to ask you to buy a DIY pump, too. even "for testing" purposes. nothing like that is connnected to, or supported by, the community. our reference design explicitly is for already-approved devices that we can rely upon.
Giuseppe
@giupo
Jan 13 2018 22:38
Not that I want to do it, just a proof of concept to play with
Ebgineer
@Ebgineer
Jan 13 2018 22:39
Its a ton of engineering to redo too. Lots of design/build/analyze cycles.
Giuseppe
@giupo
Jan 13 2018 22:39
absolutely, nono, I just would like to playaround with things ... so basically it's a waste of time?
Dana Lewis
@danamlewis
Jan 13 2018 22:39
I would say so, yes
Ebgineer
@Ebgineer
Jan 13 2018 22:40
There's a lot of other great things to work on though :smile:
Dana Lewis
@danamlewis
Jan 13 2018 22:40
lots of reverse engineering work avialable for the Combo, or the Omnipod, though, if you wanna help make more pump types available for DIY looping :)
^^ :+1:
Giuseppe
@giupo
Jan 13 2018 22:41
I was just not satisfied with the RF coms, I hate them :D
Dana Lewis
@danamlewis
Jan 13 2018 22:41
you may want to help with a rileylink driver for AndroidAPS, for example, to make AAPS an option for medtronic pumps (in addition to the DANA pumps, and soon to be Combo pumps as well)
those have bluetooth built in, so you might be interested in checking those out
Ebgineer
@Ebgineer
Jan 13 2018 22:42
I haven't heard anything further on 670G communications recently. Any word after the initial success report?
Dana Lewis
@danamlewis
Jan 13 2018 22:42
nope
Giuseppe
@giupo
Jan 13 2018 22:42
same here.
Ebgineer
@Ebgineer
Jan 13 2018 22:43
I did hear that JDRF put out a statement supporting DIY closed loop
Giuseppe
@giupo
Jan 13 2018 22:44
Before OpenAPS I rarely used a CGM sensor, I (dumbly) considered it a burden, now I can't live without
kallnap
@kallnap
Jan 13 2018 22:45
@guiseppe I was not even considering a pump before OpenAPS
Giuseppe
@giupo
Jan 13 2018 22:45
Anyway, I have a some devices in my house, (termovalves) that use the same freq 868mhz, I suspect that they create a lot of noise for my rig.
that's why I was wondering how cut down those RF building a single device ... but as you said, it's a no go.
Ebgineer
@Ebgineer
Jan 13 2018 22:47
I'd speculate that there will be some merging of devices in the future.
The CGM functions and closed loop can in theory be run in the pump, eliminating some of the communication failure modes.
The development cycles for pump manufacturers are pretty lengthy though so I'm not holding my breath.
Giuseppe
@giupo
Jan 13 2018 22:48
anyway, for my useless pump, I could use a linear stepper motor , right? does the oref define an interface/protocol per adding a custom device? (maybe I could use this stuff for my engineering thesis...)
Dana Lewis
@danamlewis
Jan 13 2018 22:48
@Ebgineer there's downsides to one company holding the full stack, though
Ebgineer
@Ebgineer
Jan 13 2018 22:49
@danamlewis Agreed, I was saddened to see Medtronic gaining an even larger market share with two of the companies exiting the industry
Competition is good for patients
Giuseppe
@giupo
Jan 13 2018 22:50
my main concern is also that my pump is getting old, and they (the national health care system in Italy) will change it with one that is not compatible with openaps
( and I can't afford a used one, too :( )
kallnap
@kallnap
Jan 13 2018 22:51
@giupo so you have only access to Medtronic pumps in intaly?
Ebgineer
@Ebgineer
Jan 13 2018 22:51
Hopefully the advocacy efforts for pump companies to support open standards will be successful.
Giuseppe
@giupo
Jan 13 2018 22:51
no, there are others, but my doctor in on the Medtronic train
in -> is.
kallnap
@kallnap
Jan 13 2018 22:52
@giupo and there are no other doctors?
jzycom123
@jzycom123
Jan 13 2018 22:53
大家好,好热闹!
Giuseppe
@giupo
Jan 13 2018 22:53
oh yes. But this one accepted to give me a 365/24/7 cover of enlite sensors.
kallnap
@kallnap
Jan 13 2018 22:53
I know of people in Germany that switched docs, because te docs were not supportive to DIY
Giuseppe
@giupo
Jan 13 2018 22:53
others won't .
So I wanna be friend with him :)
He does. When he saw my first rig wanted to present it to a conference, but being a sociopatic software developer I declined.
:D
but I'm pretty sure that they don't have other Med 554 available... I saw 640s around.. and I don't want it.
open standards.. I wish.