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

2nd
Mar 2017
peterfulvi
@peterfulvi
Mar 02 2017 00:18
image.png
Great day i had a cold and autotune picked upon it and adjusted my ISF down to 27, and it kept me all day.
zekeat09
@zekeat09
Mar 02 2017 00:34
@scottleibrand I enter carbs through pump. The NS behave weird. At 0:00, it show no carbs. Then suddenly shown 21g carbs in COB, then disappear in a few minutes. No carbs was enter from pump for 5 hrs.
Openaps-02.png
Got around to making a case for the Explorer board, edison and 2000 mAh lipo...
...it beats the giant tic-tac container I was using. :smile:
fjungman
@fjungman
Mar 02 2017 00:36
:+1:
@ecc1 Thanks! I look into that. I think my bnep0 was created, but I've never seen it.
tazitoo
@tazitoo
Mar 02 2017 00:50
@danamlewis - think I created a PR to add the links to the CAD and STL into the docs
Scorponok
@Scorponok
Mar 02 2017 01:30
@fjungman maybe too late, but try just rebooting to fix your "can't change local name" issue? It didn't work for me initially, but then I rebooted and it was fine
John Benjamin
@Bender1061
Mar 02 2017 01:36
@tazitoo have you got it printed yet yourself? I've just gotten a 3D printer, but I'm waiting to others to say they love their case before I print any of them yet. I like how you formatted it though someone else has plans, but it's not really laid on on the pad for easy printing. I'm no CAD expert (yet) so it's nice seeing an easy to print file.
and how is heat in that box, not much room for air
fjungman
@fjungman
Mar 02 2017 01:56
@Scorponok I'll try that soon and see. Maybe it just need that...
katie disimone
@Kdisimone
Mar 02 2017 02:26

@logichammer I’m a virtual noob on this. Managed to get the syslog from rig into papertrail pretty easily. But, I see that I need that remote_syslog2 to get the other logs coming in from rig (like NS and pump). and therein ends my abilities currently. I am unsure how to even start their first line of command to download the latest version of remote_syslog2. By git? should i create a new directory for it? If it’s supposed to be one of the tar…which?

or if anyone else has setup their papertrail to take in multiple logs from rig…can you help me out? Thanks :+1:

fjungman
@fjungman
Mar 02 2017 02:36
I just did that an hour ago! Are you running Edison or RPi?
katie disimone
@Kdisimone
Mar 02 2017 02:39
edison
fjungman
@fjungman
Mar 02 2017 02:40
Some things I learned...
I was using systemd not init.d, so use the remote_syslog.systemd.service file when doing the auto start part.
For Edison, you need the i386 version. RPi is ARM.
The tutorial is here:
katie disimone
@Kdisimone
Mar 02 2017 02:41
oh jees, so embarassing to say, but I can’t get past step one on the tutorial.
download.
I hate not being proficient in this stuff.
fjungman
@fjungman
Mar 02 2017 02:42
from any directory you like, run:
katie disimone
@Kdisimone
Mar 02 2017 02:42
i imagine it is a git command…but
Then issue the commands show in step 1
katie disimone
@Kdisimone
Mar 02 2017 02:44
:+1: doing it now
fjungman
@fjungman
Mar 02 2017 02:44
Yeah, I had to look up "wget" myself... :)
katie disimone
@Kdisimone
Mar 02 2017 02:44
I had tried to create a papertrail directory. managed to download it…but clearly not the way that wget must work
yes, already looks much better
fjungman
@fjungman
Mar 02 2017 02:46
Step 2 should be easy, but for the "path to this app’s log file, and any other log file(s) to watch." you just want to delete all the files listed in the example file they give you and just have the file /var/log/openaps/pump-loop.log listed there. Then make the changes they mention using the info papertrail gave you when you signed up
Step 3 is a no-brainer...
katie disimone
@Kdisimone
Mar 02 2017 02:46
Yup, I had skipped to step 2 earlier because that part I could understand…looks like i’ve made it through without errors…off to check logs
:smile: YOU ROCK
fjungman
@fjungman
Mar 02 2017 02:47
Step 4 you need to use the tip I gave earlier. use the remote_syslog.systemd.service file when doing the auto start part.
katie disimone
@Kdisimone
Mar 02 2017 02:47
It’s going
fjungman
@fjungman
Mar 02 2017 02:47
Yeah!
Don't forget step 4
katie disimone
@Kdisimone
Mar 02 2017 02:48
doing step 4 now. hopefully
jcorbett80
@jcorbett80
Mar 02 2017 02:53
setup openaps on Sy using xdrip and wifi tether. All of a sudden this evening it always has a failure trying to replace temp basal so it never starts running now. I tried bash to quickly rerun oref0 and got the following message
katie disimone
@Kdisimone
Mar 02 2017 02:54
@fjungman Do I have to be in root to do step 4?
root@edison1:~/remote_syslog# cp examples/remote_syslog.systemd.service /etc/init.d/remote_syslog
cp: cannot stat ‘examples/remote_syslog.systemd.service’: No such file or directory
I mean should i cd first?
jcorbett80
@jcorbett80
Mar 02 2017 02:55
the last error was from me
fjungman
@fjungman
Mar 02 2017 02:57
@Kdisimone you should be putting the file at: /etc/systemd/system/remote_syslog.service
katie disimone
@Kdisimone
Mar 02 2017 02:57
ahhh
fjungman
@fjungman
Mar 02 2017 02:58
Note that it is renamed in the process, as it is stored on github as remote_syslog.systemd.service but you store it as remote_syslog.service
katie disimone
@Kdisimone
Mar 02 2017 02:58
I like that I got part a of the string…but completely missed part b.
duh
Ok, having a problem with the examples/ path…it’s still not liking that first part
feel like I’m missing a front part for that.
fjungman
@fjungman
Mar 02 2017 03:03
@jcorbett80 I can't see your "Capture.MHT" files. Are the time and dates set properly on all your devices?
jcorbett80
@jcorbett80
Mar 02 2017 03:04
@fjungman They were but I will check again
katie disimone
@Kdisimone
Mar 02 2017 03:04
Can I just create that file? rather than copying?
fjungman
@fjungman
Mar 02 2017 03:04
@Kdisimone you didn't wget the example, did you? I did that, and it is an html file... You have to copy and paste...
Yes, create.
katie disimone
@Kdisimone
Mar 02 2017 03:05
No, I didn’t wget it. (note to self…need to remember wget)
katie disimone
@Kdisimone
Mar 02 2017 03:14
Ok, I think I have step 4 done. I suppose the test is to reboot the rig?
fjungman
@fjungman
Mar 02 2017 03:14
Yep!
zekeat09
@zekeat09
Mar 02 2017 03:16
@scottleibrand To avoid openaps pull carbs from NS, can i remove plugin COB or careportal ? Will this solve the problem ? I found openaps hv suddenly COB increase. I suspect is double carbs from NS + pump.
Openaps-03.png
Openaps-04.png
Openaps-05.png
@05:53 carbs is 31g bolus through pump.
@09:48 carbs 30g bolus through pump
katie disimone
@Kdisimone
Mar 02 2017 03:19
@fjungman BRILLIANT. working great. I owe you! Thank you so much
fjungman
@fjungman
Mar 02 2017 03:20
No problem! Plenty of people have carried me to this point! Just pay it forward...
katie disimone
@Kdisimone
Mar 02 2017 03:21
Typing it up so I can…thanks!
fjungman
@fjungman
Mar 02 2017 03:22
:+1:
katie disimone
@Kdisimone
Mar 02 2017 03:38
hmm, maybe i spoke to soon

The other logs aren’t showing up with the reboot. Does this command line look ok for step 4?

sudo ln -s /etc/systemd/system/remote_syslog.service /etc/rc3.d/S30remote_syslog

fjungman
@fjungman
Mar 02 2017 03:41
It should have been systemctl enable remote_syslog.service and probably "sudo" if you aren't root.
I'm not too sure where you got that line from... :smile:
katie disimone
@Kdisimone
Mar 02 2017 03:43
From the read me…I got an error for the other line sudo update-rc.d remote_syslog defaults
papertrail_remote_syslog2__To_install__see__Releases__tab__Self-contained_daemon_for_reading_local_files_and_emitting_remote_syslog__without_using_local_syslogd__.jpg
fjungman
@fjungman
Mar 02 2017 03:44
Ahh, you need to be two sections down in the instructions...
blob
systemd, not init.d
katie disimone
@Kdisimone
Mar 02 2017 03:47
I don’t even see that page. Can you post link?
nevermind…found it
fjungman
@fjungman
Mar 02 2017 03:48
:+1:
katie disimone
@Kdisimone
Mar 02 2017 03:52
@fjungman sounds like you didn’t need either of the lines from that screen shot I had then?
(and it is working at reboot this time. so really…thank you x10. :smile: )
fjungman
@fjungman
Mar 02 2017 03:54
Yeah, I made the same mistakes, but did not succeed in preventing you from doing the same. Glad it is up and running!!!
katie disimone
@Kdisimone
Mar 02 2017 03:58
LMAO…give me a chance. i can obstruct with the best of them.
diabeticgonewild
@diabeticgonewild
Mar 02 2017 04:03
Windows 10 kind of sucks. Had to initiate the administrator account via command line, then had to log into it, to flash to Explorer Board. A few months ago, it didn't require the admin account on Win10.
fjungman
@fjungman
Mar 02 2017 04:08
@diabeticgonewild Really? I didn't notice that, but then maybe I'm the admin by default... Why would they require that?
Posted this in dexdrip gitter, but I'll post it here, also, since no one goes there anymore:
I want to connect the Wixel directly to the Edison serial port. I have level-shifters installed on the RiledUp board just for this. They connect UART2 Rx and Tx. I have a couple spare GPIOs if we need an interrupt or other control signal. Has anyone tried to take part of xdrip and compile it for linux? How to do calibrations, through the pump? We eliminate the graphics and bluetooth, so a big piece of the code can be removed. The Wixel code should be pretty close to the same...
diabeticgonewild
@diabeticgonewild
Mar 02 2017 04:10
@fjungman I have no idea. I tried flashing a couple of times today and it didn't work. Supposedly I am an "administrator" but it still didn't work. I had to actually enable the "hidden" administrator Windows 10 account: http://www.ghacks.net/2014/11/12/how-to-enable-the-hidden-windows-10-administrator-account/
fjungman
@fjungman
Mar 02 2017 04:11
Bizarre. Good you figured it out. Might want to add that to the docs...
diabeticgonewild
@diabeticgonewild
Mar 02 2017 04:11
After enabling the hidden Windows 10 admin account and logging into it, the flashing actually worked. For some reason, they have the "administrator function" all jacked up within Windows 10 Pro if you "do it by hand/GUI" versus command line
Yeah, I need to. I have been sick the past 2 weeks so haven't been able to PR. It really sucks.
fjungman
@fjungman
Mar 02 2017 04:12
Sorry to hear that.
diabeticgonewild
@diabeticgonewild
Mar 02 2017 04:12
It's OK. It is what it is.
fjungman
@fjungman
Mar 02 2017 04:27
@/all posting this again, as it got lost in an exchange:
I want to connect the Wixel directly to the Edison serial port. I have level-shifters installed on the RiledUp board just for this. They connect UART2 Rx and Tx. I have a couple spare GPIOs if we need an interrupt or other control signal. Has anyone tried to take part of xdrip and compile it for linux? How to do calibrations, through the pump? We eliminate the graphics and bluetooth, so a big piece of the code can be removed. The Wixel code should be pretty close to the same...
RedCyclist1
@RedCyclist1
Mar 02 2017 04:30
I just got the loop running today and haven't had a chance to see how it will behave. I was going to take it off for the night but not sure how to shut it down without impacting my set up. Where can I find the steps in the documentation or how do I stop it until I can observe it for a full 24 hours
diabeticgonewild
@diabeticgonewild
Mar 02 2017 04:33
@fjungman You don't need to do that. Check out @oschumac repo https://github.com/oschumac/xDripAPS
katie disimone
@Kdisimone
Mar 02 2017 04:33
@RedCyclist1 You can just power it down. Hold down the power button until the red light turns off.
(assuming you are using edison)
RedCyclist1
@RedCyclist1
Mar 02 2017 04:34
@Kdisimone ok thanks Katie
Scott Leibrand
@scottleibrand
Mar 02 2017 04:52
@zekeat09 it's still not clear to me if you're manually entering carbs through the NS care portal, and/or using the Loop iOS app (with NS uploads enabled). If so, you don't want to do those things. You probably should disable AMA until you're able to troubleshoot it properly. You might need to test out the dev branch of oref0.
Ah, just saw your earlier message. So sounds like it's not care portal that's the problem. Are you running Loop?
fjungman
@fjungman
Mar 02 2017 04:58
@diabeticgonewild Thanks, but that is not what I'm asking. This is looping with a Wixel and no phone. CGM data read into the Edison board without a phone or Dexcom receiver.
diabeticgonewild
@diabeticgonewild
Mar 02 2017 04:59
I think @oschumac mentioned a way for calibrating without the phone. I missed when he tagged me in a message though awhile back.
When he discussed this.
does anyone know what this error means?
Resetting spi_serial
Traceback (most recent call last):
  File "/usr/local/bin/reset_spi_serial.py", line 3, in <module>
    s = spi_serial.SpiSerial()
  File "/usr/local/lib/python2.7/dist-packages/spi_serial/spi_serial.py", line 11, in __init__
    self.dev = m.spiFromDesc("spi-raw-5-1")
AttributeError: 'module' object has no attribute 'spiFromDesc'
Attempting to communicate with pump:
openaps: no process found
fjungman
@fjungman
Mar 02 2017 05:04
@Kdisimone Hmm. I didn't do the steps shown here. Otherwise it looks ok, but it seems so short, but that is probably because you didn't put in all the missteps we made... :)
blob
katie disimone
@Kdisimone
Mar 02 2017 05:08
I felt like it was a teaser step. I was so excited how easy that first step was…and then whomp whomp…only syslog. Maybe starting farther in is better? I’ll have some people guinea pig this hopefully. :+1:
fjungman
@fjungman
Mar 02 2017 05:11
:+1:
zekeat09
@zekeat09
Mar 02 2017 05:12
@scottleibrand i am not running Loop iOS app. It just openaps edison rig with Medtronic sensor. I found the NS COB is not correctly display. As my last picture, at 11:07, the COB shown 33, while i only bolus 30g through pump at 9:48. For picture 11:06, it shown the edison read COB 33g. I wonder where it gets this value, from nightscout ? If nightacout, why it is 33g, not 20+ (after 1.15hr bolus).
Scott Leibrand
@scottleibrand
Mar 02 2017 05:14
Please check out the oref0 issues page for details on all the troubleshooting that has been done on these issues to date. If you are experiencing a similar issue, there are details in there about what troubleshooting steps to take to collect information we need to understand whether your issue is the same as one of those, or potentially a new issue.
Also, please ignore any NS COB pill results from NS's internal calculations. Only the COB in the OpenAPS pill will be accurate.
Dana Lewis
@danamlewis
Mar 02 2017 05:16
@zekeat09 also please confirm if you're reporting cob from multiple places - and be specific. The NS cob pill is separate from reading cob in the OpenAPS pill, for example, in NS.
:+1:
Stephen Chandler
@stephenc95
Mar 02 2017 05:21
Does anyone know if the special instructions for the 512, "512 users - make sure that you have created your static json files as outlined in https://openaps.readthedocs.io/en/dev/docs/walkthrough/phase-0/hardware/pump.html for raw-pump/settings.json, raw-pump/bg-targets-raw.json, and raw-pump/selected-basal-profile.json. You will also have to remove the calls for them from the your get-settings alias." also applies to the 712. Or is it just the 512?
gregtcm
@gregtcm
Mar 02 2017 05:23
@Kdisimone I'd be happy to guinea pig it for you ... once I get my explore board! :smile: Thanks for getting that all documented.
katie disimone
@Kdisimone
Mar 02 2017 05:25
@gregtcm awesome! Thanks!!
diabeticgonewild
@diabeticgonewild
Mar 02 2017 05:25
Weird. I entered python reset_spi_serial.py and then re-ran the install script for oref0 and now it's working without problems. So there's something up with mraa and the install script. Yes, I was running the install script from the root account on my Explorer board
Dana Lewis
@danamlewis
Mar 02 2017 05:25
Yes @stephenc95 should also apply to 712. Should really say x12, i.e. Either 512 or712
diabeticgonewild
@diabeticgonewild
Mar 02 2017 05:33
OpenAPS is setup for me on the Explorer board with minimal problems. Just had to reset a script and flash from the hidden administrator account on Windows. Those were the only unacknowledged contigencies.
Paul Dickens
@thebookins
Mar 02 2017 05:43
So quick question - our t1d boy goes on school camp next week - and we found out today that the cabins don't have power. Will one of those USB phone chargers keep the iPhone and edison going overnight? Any recommendations on what to buy? The charger itself can recharge in the camp kitchen during the day.
diabeticgonewild
@diabeticgonewild
Mar 02 2017 05:44
I would get a couple of 10,000 mAH USB power banks.
@thebookins Get a couple of these. Just verify that they are fully charged prior to leaving. They should hold about 10-20x the battery capacity of the battery pack on the Edison. https://www.amazon.com/Anker-PowerCore-Ultra-Compact-High-speed-Technology/dp/B0194WDVHI/ref=sr_1_1?ie=UTF8&qid=1488433549&sr=8-1&keywords=10000+mah+power+bank
Stephen Chandler
@stephenc95
Mar 02 2017 05:51
'[FAILED] Failed to start Load Kernel Modules.' Has anyone gotten this while flashing the edison from windows? It gives that error while its rebooting towards the end of the flash, and now it is stuck at ' Ready to receive application '. It is connected to a charged battery, so I don't know what's wrong.
Paul Dickens
@thebookins
Mar 02 2017 05:53
thanks @diabeticgonewild . Are these power banks one device only? I.e. I need one for the iPhone and one for the edison? Should we be looking at current rating? I remember in the early days with OpenAPS on raspberry pi we needed particular power banks with enough peak current.
diabeticgonewild
@diabeticgonewild
Mar 02 2017 05:59
You will probably want 4-5 of those to get you through the week if you are using both a phone and an edison. There is 1 USB charge port per battery
Dana Lewis
@danamlewis
Mar 02 2017 06:01
Several different kinds of batteries. Many of the common ones have 2 USB ports, so you could easily keep the Edison charged alongside the phone.
Paul Dickens
@thebookins
Mar 02 2017 06:01
Actually we have found out that we can recharge the power supply itself using a staff area, so 1 should see us through! The one I have seen actually has 2 outputs so should be sweet.
Dana Lewis
@danamlewis
Mar 02 2017 06:02
:+1:
Paul Dickens
@thebookins
Mar 02 2017 06:02
Thanks @danamlewis @diabeticgonewild appreciate your help.
fjungman
@fjungman
Mar 02 2017 06:12
FYI, @thebookins , there are many. I bought three 20,000mAhr ones at Frys a few weeks ago (on sale for $25 each) and they had two ports, but if you looked closely, one was marked 1A and one was marked 2.4A. So, make sure the Pi goes on the 2.4A port. Also, they take about 12 hours to recharge with a good (2.1A) wall charger.
Paul Dickens
@thebookins
Mar 02 2017 06:15
The one we're considering has 1 A and 2.4 A @fjungman . Good to know about the 12 hr recharge.
Tim Street
@tim2000s
Mar 02 2017 06:22
I have a couple of 20Ah packs for those types of occasion. One pack got me through 5 days away recharging an iPhone and an xDrip last year.
Scott Leibrand
@scottleibrand
Mar 02 2017 06:28
It's definitely nice to have two big batteries, so you can charge one while your have the other one out and about.
andyhandy
@andyhandy
Mar 02 2017 09:29
any suggestions how to troubleshoot this on edison with explorer board?
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.
Listening: .................................................................................................
never happened on edison before
Tim Street
@tim2000s
Mar 02 2017 09:45
Have you rebooted?
andyhandy
@andyhandy
Mar 02 2017 09:48
yes and reinstalling oref0
changed battery on pump too
Stephen Chandler
@stephenc95
Mar 02 2017 09:58
What do most people use to flash their edisons? Is one way easier than another? Because I cannot for the life of me get this edison to work when flashing with windows
andyhandy
@andyhandy
Mar 02 2017 10:34
where are the instructions for reflashing the cc111x chip for the explorer board? thanks
Tim Street
@tim2000s
Mar 02 2017 10:53
@stephenc95 I've used mac and Linux successfully multiple times. Linux has proven the most robust for me.
@andyhandy Try the mmeowlink channel. I think I've seen discussion there.
andyhandy
@andyhandy
Mar 02 2017 11:18
@tim2000s any ideas?
root@myedison:~/ccprog# ./ccprog -p 19,7,36 erase
Using pins: DC=19, DD=7, RESET=36
Erasing chip.
NOP failed!
Could not put device into debug mode.
Chip erase failed.
Tim Street
@tim2000s
Mar 02 2017 11:20
@andyhandy afraid not. It's not something I've tried.
andyhandy
@andyhandy
Mar 02 2017 11:55
@tim2000s finally figured it out. turns out a nut had fallen off and the edison had become loose from the explorer board. thanks for your help.
Stephen Chandler
@stephenc95
Mar 02 2017 11:58
Is there any way to erase everything from the edison? It seems to be stuck in a boot loop and wont flash now
Tim Street
@tim2000s
Mar 02 2017 12:15
Are you reporting it and hitting enter before it gets to that point?
*rebooting
So that you just end up with >
Then when you try to flash it start the flashall process, then type 'run do_flash' at the >
Stephen Chandler
@stephenc95
Mar 02 2017 12:19
I'll give it a try
Stephen Chandler
@stephenc95
Mar 02 2017 12:25
Ok so I did that, and this is what happened. Using U-boot target: edison-blankrndis
Now waiting for dfu device 8087:0a99
Please plug and reboot the board
Dfu device found
Flashing IFWI
Download [=========================] 100% 4194304 bytes
Download done.
Download [=========================] 100% 4194304 bytes
Download done.
Flashing U-Boot
Download [=========================] 100% 245760 bytes
Download done.
Flashing U-Boot Environment
Download [=========================] 100% 65536 bytes
Download done.
Flashing U-Boot Environment Backup
Download [=========================] 100% 65536 bytes
Download done.
Rebooting to apply partiton changes
Dfu device found
Flashing boot partition (kernel)
Download [=========================] 100% 5980160 bytes
Download done.
Flashing rootfs, (it can take up to 5 minutes... Please be patient)
Rebooting
U-boot & Kernel System Flash Success...
Your board needs to reboot to complete the flashing procedure, please do not unplug it for 2 minutes.
It just completely skipped over flashing rootfs
Everything that was flashed before rootfs went really, really fast too
Stephen Chandler
@stephenc95
Mar 02 2017 12:42
Ok updated the page file size for drives in windows. That was weird because it never gave me that problem before, plus I have 16gb of ram, which is much more than 6gb
Stephen Chandler
@stephenc95
Mar 02 2017 12:56
Welcome to rescue mode! Type "systemctl default" or ^D to enter default mode.
Type "journalctl -xb" to view system logs. Type "systemctl reboot" to reboot.
Give root password for maintenance
(or type Control-D to continue): [ 18.960967] dwc3-device dwc3-device.1: request f5dafd80 was not queued to ep6in-bulk
[ 18.961060] dwc3-device dwc3-device.1: request f5dafcc0 was not queued to ep6in-bulk
[ 19.538615] dwc3-device dwc3-device.1: request f5daf360 was not queued to ep6in-bulk
[ 19.538707] dwc3-device dwc3-device.1: request f5daf420 was not queued to ep6in-bulk
[ 19.921087] dwc3-device dwc3-device.1: request f5dafea0 was not queued to ep6in-bulk
[ 19.921181] dwc3-device dwc3-device.1: request f5dafde0 was not queued to ep6in-bulk
[ 20.112432] dwc3-device dwc3-device.1: request f5daf120 was not queued to ep6in-bulk
[ 20.112525] dwc3-device dwc3-device.1: request f5daf060 was not queued to ep6in-bulk
[ 20.303283] dwc3-device dwc3-device.1: request f5daf9c0 was not queued to ep6in-bulk
[ 20.874986] dwc3-device dwc3-device.1: request f5e6d000 was not queued to ep6in-bulk
[ 20.875078] dwc3-device dwc3-device.1: request f5e6d780 was not queued to ep6in-bulk
[ 21.065331] dwc3-device dwc3-device.1: request f5daf720 was not queued to ep6in-bulk
[ 21.065425] dwc3-device dwc3-device.1: request f5daf660 was not queued to ep6in-bulk
[ 21.449662] dwc3-device dwc3-device.1: request f5daf300 was not queued to ep6in-bulk
[ 21.449756] dwc3-device dwc3-device.1: request f5daf900 was not queued to ep6in-bulk
[ 21.640837] dwc3-device dwc3-device.1: request f5e6d000 was not queued to ep6in-bulk
[ 21.640930] dwc3-device dwc3-device.1: request f5e6d780 was not queued to ep6in-bulk
[ 21.834313] dwc3-device dwc3-device.1: request f5daf9c0 was not queued to ep6in-bulk
[ 21.834405] dwc3-device dwc3-device.1: request f5dafde0 was not queued to ep6in-bulk

PSH KERNEL VERSION: b0182929
WR: 20104000


SCU IPC: 0x800000d0 0xfffce92c

PSH miaHOB version: TNG.B0.VVBD.0000000c

microkernel built 10:14:01 Jan 13 2015

* PSH loader *
PCM page cache size = 192 KB
Cache Constraint = 0 Pages
Arming IPC driver ..
Adding page store pool ..
PagestoreAddr(IMR Start Address) = 0x04899000
pageStoreSize(IMR Size) = 0x00080000

Ready to receive application
now I'm stuck at this

Anybody know what could possibly be wrong?
Stephen Chandler
@stephenc95
Mar 02 2017 13:16

systemd-modules-load.service - Load Kernel Modules
Loaded: loaded (/lib/systemd/system/systemd-modules-load.service; static)
Active: failed (Result: exit-code) since Sat 2000-01-01 00:00:26 UTC; 38s ago
Docs: man:systemd-modules-load.service(8)
man:modules-load.d(5)
Process: 102 ExecStart=/lib/systemd/systemd-modules-load (code=exited, status=1/FAILURE)
Main PID: 102 (code=exited, status=1/FAILURE)

Jan 01 00:00:25 ubilinux systemd-modules-load[102]: Inserted module 'bcm_bt_lpm'
Jan 01 00:00:26 ubilinux systemd-modules-load[102]: Inserted module 'bcm4334x'
Jan 01 00:00:26 ubilinux systemd[1]: systemd-modules-load.service: main pro...RE
Jan 01 00:00:26 ubilinux systemd[1]: Failed to start Load Kernel Modules.
Jan 01 00:00:26 ubilinux systemd[1]: Unit systemd-modules-load.service ente...e.
Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable.
Hint: Some lines were ellipsized, use -l to show in full.
also getting this error when it tries to boot

Stephen Chandler
@stephenc95
Mar 02 2017 13:22
So I get the failed to start load kernel modules error, and the edison keeps booting into rescue mode. I have tried the rescue mode fix. I tried a different pc, different usb cables, battery, redownloaded jubilinux, reinstalled the edison drivers, and nothing seems to work. Any ideas? I'm about ready to give up
dan lightner
@danimaniac
Mar 02 2017 13:54
@thebookins I just purchased an iFrogz 13,000mAh battery pack at Sam's. Clearanced for $15. any portable charger with sufficient capacity and charge rate should work.
ejoe132
@ejoe132
Mar 02 2017 14:11
I was working with Scott the other night after I had a file use all the memory on my edison, we got it removed but it deleted everything off my edison, now I cant even get it to do anything when i try to serial in. Any tips at this point?
dan lightner
@danimaniac
Mar 02 2017 14:40
@tim2000s I will be using my linux box to flash my Edison once my Explorer Board arrives. I assume it's pretty straightforward. I would generally follow the Rasp Pi instructions without the prerequisite steps, right? There are no clear setup instructions for using a linux box...I will attempt to PR some instructions after I successfully perform my setup.
Tim Street
@tim2000s
Mar 02 2017 16:29
yes, that's pretty much correct.
Eric
@ecc1
Mar 02 2017 16:46
@fjungman This is different enough from what you're asking for that it might not help, but I've connected a CC2500 (the same radio as the Wixel, but no MCU) directly to an Edison over SPI, and have written user-code in Linux (in Go) to read and decode the transmissions from G4 sensors. Code is in https://github.com/ecc1/cc2500
Eric
@ecc1
Mar 02 2017 17:14
@ejoe132 If you really deleted everything, you'll have to reflash.
kenstack
@kenstack
Mar 02 2017 17:27
@fjungman after restarting BT but before setting hostname, try sudo hciconfig hic0 up
hci0 not hic0 @fjungman sorry typo
@ecc1 are you using your own calibrations or seperately reading cals from the receiver? Been debating what the best plan here is to do a truly offline rig - calibrations can be tricky
Eric
@ecc1
Mar 02 2017 18:05
I don't have that CC2500 in my production rig yet, precisely because of the calibration issue. So still doing offline via BLE to the G4 Share receiver.
Russell Gardner
@RussGG
Mar 02 2017 18:59
I’m taking my son skiing for the weekend and want to be set up for the wifi at the condo. There isn’t a PW for the network. Do I leave psk=“” or not even put that line in?
katie disimone
@Kdisimone
Mar 02 2017 19:12
@RussGG openaps/docs#671
And update your oref0 if it is outdated so that it will automatically disconnect from captive portals in that situation. Redirects to a login page on open networks will result in poor rig results.
fjungman
@fjungman
Mar 02 2017 19:26
@ecc1 Thanks, I'll look at that code tonight! I think the functions that xdrip performs need to be enumerated and the needed ones ported over.
@kenstack Thanks, I'll try that tonight!
Scott Leibrand
@scottleibrand
Mar 02 2017 19:34
@RussGG you can also add a single open network (instead of allowing all of them) if you prefer, by specifying the network name and the OPEN line (but not necessarily the really low priority)
fjungman
@fjungman
Mar 02 2017 19:39
Interesting problem. Any ideas? The loop is calculating temps and thinks it is setting them, but they are not getting set. He is at his internship now, so I cannot work with it, but I see the logs and he said no temps are being set. It worked at home. Pump comms look ok. Is there something that could stop the temps from being set?
ar 02 11:11:13 ubilinux pump-loop.log: Starting pump-loop at Thu Mar 2 11:11:13 PST 2017:
Mar 02 11:11:46 ubilinux pump-loop.log: Radio ok, Listening: .No pump comms detected from other rigs
Mar 02 11:11:47 ubilinux pump-loop.log: Profile less than 60m old
Mar 02 11:11:47 ubilinux pump-loop.log: glucose.json newer than temp_basal.json
Mar 02 11:11:54 ubilinux pump-loop.log: Temp refreshed
Mar 02 11:11:55 ubilinux pump-loop.log: Temp duration 27m or more
Mar 02 11:12:16 ubilinux pump-loop.log: glucose.json newer than pumphistory: Refreshed pumphistory
Mar 02 11:12:21 ubilinux pump-loop.log: {"carbs":0,"boluses":0,"mealCOB":0}
Mar 02 11:12:21 ubilinux pump-loop.log: {"ratio":1}
Mar 02 11:12:21 ubilinux pump-loop.log: {"iob":0.787,"activity":-0.002,"bolussnooze":0,"basaliob":0.787,"netbasalinsulin":0.55,"hightempinsulin":3.8,"time":"2017-03-02T19:10:33.000Z"}
Mar 02 11:12:21 ubilinux pump-loop.log: {"delta":2,"glucose":201,"short_avgdelta":4.5,"long_avgdelta":7.84}
Mar 02 11:12:21 ubilinux pump-loop.log: {"duration":27,"rate":6,"temp":"absolute"}
Mar 02 11:12:21 ubilinux pump-loop.log: Adjusting target_bg from 82.5 to 83
Mar 02 11:12:21 ubilinux pump-loop.log: Carb Impact: 1.8 mg/dL per 5m; CI Duration: 0 hours
Mar 02 11:12:21 ubilinux pump-loop.log: Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: 0 hours
Mar 02 11:12:22 ubilinux pump-loop.log: determine-basal://text/shell/enact/suggested.json
Mar 02 11:12:22 ubilinux pump-loop.log: reporting enact/suggested.json
Mar 02 11:12:22 ubilinux pump-loop.log: {"temp":"absolute","bg":201,"tick":"+2","eventualBG":197,"snoozeBG":197,"predBGs":{"IOB":[201,203,204,206,207,208,208,208,208,208,208,207,206,205,204,203,201,200,199,197,196,194,193,191,190,188,187,186,185,184,183,182,181,180,180,179,178,178,177,177,177,176,176,175,175,175,175,175,174]},"COB":0,"IOB":0.787,"reason":"COB: 0, Dev: 11, BGI: 0.19, ISF: 19, Target: 83; Eventual BG 197 >= 85, adj. req. rate: 15.3 to maxSafeBasal: 6, temp 6 >~ req 6U/hr"}
Mar 02 11:12:23 ubilinux pump-loop.log: {"bg":199,"temp":"absolute","snoozeBG":228,"recieved":true,"predBGs":{"IOB":[199,205,211,216,221,225,228,230,232,234,234,234,234,234,233,232,232,231,231,230,230,229,228,228,227,227,227,226,226,226,226,225]},"rate":6,"reason":"COB: 0, Dev: 40, BGI: 0.29, ISF: 19, Target: 83; Eventual BG 228 >= 85, max_iob 7, adj. req. rate: 16.1 to maxSafeBasal: 6, no temp, setting 6U/hr","COB":0,"eventualBG":228,"timestamp":"2017-03-02T11:08:15.030285","duration":30,"tick":"+7","IOB":0.581}
Mar 02 11:12:23 ubilinux pump-loop.log: Settings less than 10m old
Mar 02 11:12:24 ubilinux pump-loop.log: Pumphistory-24 less than 20m old
Mar 02 11:12:24 ubilinux pump-loop.log: Completed pump-loop at Thu Mar 2 11:12:23 PST 2017
Mar 02 11:15:18 ubilinux pump-loop.log: Starting pump-loop at Thu Mar 2 11:15:18 PST 2017:
Mar 02 11:15:52 ubilinux pump-loop.log: Radio ok, Listening: .No pump comms detected from other rigs
Mar 02 11:15:53 ubilinux pump-loop.log: Profile less than 60m old
Mar 02 11:15:53 ubilinux pump-loop.log: glucose.json newer than temp_basal.json
Mar 02 11:16:09 ubilinux pump-loop.log: Temp refreshError, retrying
Mar 02 11:16:14 ubilinux pump-loop.log: Starting pump-loop at Thu Mar 2 11:16:14 PST 2017:
Mar 02 11:16:48 ubilinux pump-loop.log: Radio ok, Listening: .No pump comms detected from other rigs
Mar 02 11:16:49 ubilinux pump-loop.log: Profile less than 60m old
Mar 02 11:16:49 ubilinux pump-loop.log: glucose.json newer than temp_basal.json
Mar 02 11:16:56 ubilinux pump-loop.log: Temp refreshed
Mar 02 11:17:02 ubilinux pump-loop.log: {"carbs":0,"boluses":0,"mealCOB":0}
Mar 02 11:17:02 ubilinux pump-loop.log: {"ratio":1}
Mar 02 11:17:03 ubilinux pump-loop.log: {"iob":0.99,"activity":-0.0008,"bolussnooze":0,"basaliob":0.99,"netbasalinsulin":0.5,"hightempinsulin":3.75,"time":"2017-03-02T19:15:22.000Z"}
Mar 02 11:17:03 ubilinux pump-loop.log: {"delta":6,"glucose":207,"short_avgdelta":5,"long_avgdelta":7.53}
Mar 02 11:17:03 ubilinux pump-loop.log: {"duration":22,"rate":6,"temp":"absolute"}
Mar 02 11:17:03 ubilinux pump
diabeticgonewild
@diabeticgonewild
Mar 02 2017 19:39
@stephenc95 Have you tried following these videos?
katie disimone
@Kdisimone
Mar 02 2017 19:42
@RussGG the OPEN line that Scott is referring to is the key_mgmt=NONE in that pull request btw
(Noting it for others in case they are reading the topic later)
@fjungman pump won't set temps if the pump is in an unfinished prime screen...but if he's able to see the pump screen, that would've probably been already addressed.
fjungman
@fjungman
Mar 02 2017 19:46
@Kdisimone Thanks! Knowing that kid... :worried:
Nope, he just bolused, but it is still not setting temps...
What gets me is it keeps saying "temp refreshed". Does it read back the temp to know it is set? If so, how is it getting erased?
Matt Pressnall
@logichammer
Mar 02 2017 20:10
@Kdisimone, I'll see about typing up some notes / putting this into a PR...more to come!
Scott Leibrand
@scottleibrand
Mar 02 2017 20:20
@fjungman I'll wager money that temps are being set, and he just isn't reading the pump display correctly. You have "recieved":true," and {"duration":22,"rate":6,"temp":"absolute"}, indicating the 6U high temp was successfully set and had been running for 8 minutes at that point.
fjungman
@fjungman
Mar 02 2017 20:22
Hmm. Why does it say "adj. req. rate: 16.1 to maxSafeBasal: 6, no temp, setting 6U/hr"? The "no temp" is worrying me, and his BG was spiking (of course, the kid doesn't know what the word "bolus" means...)
Scott Leibrand
@scottleibrand
Mar 02 2017 20:22
also, "basaliob":0.99, and "hightempinsulin":3.75,, which further confirms (via pumphistory) that high temps are being set
fjungman
@fjungman
Mar 02 2017 20:23
So, it pulls those from the pump? Ok, that gives me some confidence here. Still don't understand "no temp", when it was showing temps properly yesterday...
Scott Leibrand
@scottleibrand
Mar 02 2017 20:25
it looks like there was no temp set as of 11:08, but it set one at 11:09 or so
what does NS show?
so could be he was correct, and just happened to look at the pump right before it set a temp
fjungman
@fjungman
Mar 02 2017 20:26
NS shows the temps.
Scott Leibrand
@scottleibrand
Mar 02 2017 20:26
those also come from pumphistory
fjungman
@fjungman
Mar 02 2017 20:27
Could be. So, pumphistory is always read from the pump, not what it "tried" to send to the pump?
Scott Leibrand
@scottleibrand
Mar 02 2017 20:27
there would have to be something really weird going on for all of that info to be self-consistent but wrong. (like maybe you have two pumps and it's setting the temps on the test pump)
fjungman
@fjungman
Mar 02 2017 20:27
I've been working on his rigs for a year, but I still don't understand the "engine" behind oref0. Where all the data is being read from vs. being calculated from. In my mind, I don't know what is calculated and stored, vs what is read and used.
Scott Leibrand
@scottleibrand
Mar 02 2017 20:27
correct, pumphistory reflects what was actually enacted and run, as reported later by the pump
fjungman
@fjungman
Mar 02 2017 20:28
That is important for me to remember. I guess I just need to debug the missing temps in the report.
Scott Leibrand
@scottleibrand
Mar 02 2017 20:28
you might want to work through the old manual setup sections in the docs to see how the various reports fit together
and/or start drilling down into your setup with openaps alias show and openaps report show commands
fjungman
@fjungman
Mar 02 2017 20:29
I did last summer. Weeks of headaches and printing each report on a tiny slip of paper, then numbering them and listing which ones they called. It is just too much to keep straight in my head. I guess that is from being a hardware guy... :)
It was like trying to build a 100-piece jigsaw puzzle, in my head.
Scott Leibrand
@scottleibrand
Mar 02 2017 20:30
well, most of the pieces don't matter most of the time, so if you want you can focus just on the ones that do. :)
fjungman
@fjungman
Mar 02 2017 20:31
If you wrote the code, it probably is much easier to keep straight. I will dig through it and learn where the no temp is happening, tonight.
Scott Leibrand
@scottleibrand
Mar 02 2017 20:31
fjungman
@fjungman
Mar 02 2017 20:32
Oh. I never dared stare into the Sun like that! I'll take a look at that now! I hope my looking doesn't break something (I have that ability, apparently...)
Scott Leibrand
@scottleibrand
Mar 02 2017 20:32
that main() loop does the same thing as the aliases in your setup, but it's a lot easier to read. each function is defined down below. you can ignore the smb stuff
fjungman
@fjungman
Mar 02 2017 20:33
what is "smb"?
not something you want to be running at this point. I'm only referencing it because I refactored the regular non-SMB stuff out into a more readable shell script in order to do the SMB work
fjungman
@fjungman
Mar 02 2017 20:35
Something I have been meaning to ask you: Why listen for 30s before running mmtune? Seems the longer you wait, the more likely you are to talk over another device...
Scott Leibrand
@scottleibrand
Mar 02 2017 20:35
wait-for-silence restarts the clock every time it hears another rig talking
the idea there is that if you have to mmtune, you probably have a weak signal, and if there's another rig with a stronger signal already talking to the pump, you don't want to talk over it
fjungman
@fjungman
Mar 02 2017 20:36
Right, but is there some magic behind 30s of silence?
Scott Leibrand
@scottleibrand
Mar 02 2017 20:37
not really. I randomized it in oref0-pump-loop instead of using a fixed 30s
fjungman
@fjungman
Mar 02 2017 20:39
Oh, cool. I just thought that with a running loop, there is a maximum length of silence during the operation, and you would just go 1 second past that and then know you are clear. But, maybe there is variance in the code execution, so longer times would be needed. I'm just looking at the power consumption angle...
Scott Leibrand
@scottleibrand
Mar 02 2017 20:40
the biggest thing with waiting for silence is (trying to avoid) synchronizing different rigs all waiting for exactly the same amount of silence, and then all starting to talk at once
fjungman
@fjungman
Mar 02 2017 20:42
Yeah, could be a real problem. I like the randomization, and for that to work, 30s-ish time frames make a lot of sense. Ok, I'll leave you alone. I love the code of oref0-pump-loop, it is SO MUCH EASIER to read than all my previous loops!!!
Scott Leibrand
@scottleibrand
Mar 02 2017 20:42
:+1:
Dana Lewis
@danamlewis
Mar 02 2017 21:18
(Seattle area OpenAPS-ers, and anyone else who wants to visit ;) Next meetup scheduled for Monday, 3/6: http://whoozin.com/KCQ-XRU-D6YE)
Matt Pressnall
@logichammer
Mar 02 2017 21:27
IN!
Dana Lewis
@danamlewis
Mar 02 2017 21:31
:)
fjungman
@fjungman
Mar 02 2017 22:04
Is there a way to make the log more verbose for debugging? I have an issue that I want to debug while he is out of the house, but there is not enough detail in the log. I need to see why it is not reading the BD data from Nightscout, even though the data in NS is up to date, and the logs are going to Papertrail, so it has Internet access.
Elliott
@esteward
Mar 02 2017 22:06
For those who use offline mode by plugging their CGM into their Edison. Is there a particular cable that works reliably that you would recommend?
Dana Lewis
@danamlewis
Mar 02 2017 22:11
@tim2000s @ceben80 or anyone else who’s been Hot Buttoning, can you comment on or thumbs up this PR and let me know if it’s good to merge as-is? https://github.com/openaps/docs/pull/667/files
Scott Leibrand
@scottleibrand
Mar 02 2017 22:16
@fjungman are you looking at ns-loop.log (vs. pump-loop.log)?
fjungman
@fjungman
Mar 02 2017 22:20
No, I wasn't aware of that one. I'll have that sent to papertrail, also.
The one I used to use last summer was loop.log. I assumed pump-loop.log was the replacement.
John Benjamin
@Bender1061
Mar 02 2017 22:28
@fjungman do a cd /var/log/openaps and then do a ls and you will see all logs available
fjungman
@fjungman
Mar 02 2017 22:30
Thanks, John!
tazitoo
@tazitoo
Mar 02 2017 22:46
@Bender1061 - yup I've printed it twice. Once in hot pink, and once in a lighter blue. It's fine while operating - no apparent problems. I pop the lid off to recharge and stand it upright with the usb ports at the top. Temp obviously goes up when recharging - but no noticeable issues. One bit of customization you might want to edit - I've got the hostname embossed on the lid (my rigs are 'Ernie' and 'Bert'). YMMV.
Dana Lewis
@danamlewis
Mar 02 2017 22:46
:)
John Benjamin
@Bender1061
Mar 02 2017 22:47
Nice!
tazitoo
@tazitoo
Mar 02 2017 22:49
@danamlewis - I put some scripts up on github I use to reduce the amount of data I pull over shareble - https://github.com/tazitoo/openaps_reduce_comms . For me, it makes bluetooth/offline mode stable and reliable - needing a reboot about once a week. Thought you and @scottleibrand might be interested.
<of course your definition of 'stable' and 'reliable' may differ>
Dana Lewis
@danamlewis
Mar 02 2017 22:50
hehe.
is this for use with your self-built loop, or with use with a loop built from oref0-setup.sh?
I’m also surprised you’re finding openxshareble stable - wasn’t able to ever achieve that, but now that we have bluetooth tethering fairly solid it might be time to take another stab over there...
tazitoo
@tazitoo
Mar 02 2017 22:52
my own loop - I started using openaps/oref0 before the slick setup loop and just continued using my own shell script
but it's still pretty normal stuff - it just uses the python to save history to CSV, figure out how much (new) data to pull, run the openaps report, and merge to get enough lenght for autosens, etc.
Dana Lewis
@danamlewis
Mar 02 2017 22:55
neat, will take a look. thanks!
tazitoo
@tazitoo
Mar 02 2017 22:56
The real thanks goes to you guys! A1C would not be what it is without looping....
mm22dl
@mm22dl
Mar 02 2017 22:57
Hey loopers! I've been trying lately to understand a little more of how decocare works in order to reproduce its results independently and hopefully help doing some more decoding. I just ran pyreverse on the whole thing and here is what I got:
classes_decocare.png
I just thought I'd share that for the ones of you that didn't know how big the project is! :D
packages_decocare.png
pretty neat I find :)