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

22nd
Nov 2018
mrboletmrpot
@mrboletmrpot
Nov 22 2018 00:15
Hello suddenly my daughter's openaps says cc111x error, uneable to preflight. " cannot connect to CC111x radio on /dev/spidev5.1
2018/11/22 00:56:51 cc111x: no response" I tried to switch explorer board, done but no success, I tried to reinstall oref0, done but no improvement. And I tried to flash it to wipe for a clean start, but was unable to do it, flashall.bat does not seems to work. Any Idea, what should I try? (pump battery changed, pump in good range, she is not between pump and edison circuit, circuit not really hot, putted appart and back together to check connection)
Ebgineer
@Ebgineer
Nov 22 2018 02:13
@mrboletmrpot I haven't seen that error before, but it sounds like you're on the right track looking at pump to rig communication via radio. Can you give a bit more info on the type of pump, and list a full loop cycle from start up through the error message.
jgslade
@jgslade
Nov 22 2018 02:40
@mrboletmrpot I got that error earlier today, I corrected it by shutting down the system and disconnecting all power sources for a few seconds then booted it back up.
Eric
@ecc1
Nov 22 2018 03:53
@mrboletmrpot that error means the Edison isn't getting the response it expects from the firmware on the cc111x radio chip, so you might try reflashing that (not the Edison), with a cc-programmer if you have one or with ccprog. But no idea why it would suddenly stop if nothing else changed
Ebgineer
@Ebgineer
Nov 22 2018 04:14
@audiefile I'm able to call up the offline web page by entering the rig IP address in chrome on windows 10, no port number or anything else specified.
live4sw
@live4sw
Nov 22 2018 11:53
Has anyone ever had an A52 error while manually priming? I know it is common with Bolus Wizard (which I don't use), but I was taken aback to get one while priming a new reservoir.
Just re-inputted settings and am back on track (I think), but not sure what happened.
I had no COB, so it wasn't trying to SMB or anything
live4sw
@live4sw
Nov 22 2018 11:58
Oddly, though I did need to re-set the time, I didn't lose my basals or other history.
live4sw
@live4sw
Nov 22 2018 12:35
From looking at the logs, my rig attempted to re-set the time at 6:49:04am, and the A52 error occurred at 6:49am, while I was doing the manual prime. Is it a plausible/reasonable/likely theory that the A52 error occurred because a time change was attempted at the same time the prime was being done? Kind of like the main cause of A52 errors (bolus wizard while SMB being attempted)?

Here's the exact language from the logs from the rig that was about 2 feet away from me when the A52 error happened (sometime between 6:49 and 6:50am):

Nov 22 06:49:04 [] pump-loop.log: Checking pump clock: "2018-11-22T06:43:18-05:00" is within 90s of current time: Thu Nov 22 06:49:04 EST 2018 Nov 22 06:49:04 [] pump-loop.log: Pump clock is more than 55s off: attempting to reset it Nov 22 06:49:04 [] pump-loop.log: Waiting for ntpd to synchronize... OK! Nov 22 06:49:04 [] pump-loop.log: Setting pump time to Thu Nov 22 06:49:04 EST 2018 Nov 22 06:49:18 [] pump-loop.log: IndexError: bytearray index out of range Nov 22 06:49:18 [] pump-loop.log: Setting CGM time to Thu Nov 22 06:49:18 EST 2018 Nov 22 06:49:20 [] pump-loop.log: AttributeError: 'NoneType' object has no attribute 'ReadSystemTime' Nov 22 06:49:20 [] pump-loop.log: Error: pump clock refresh error / mismatch Nov 22 06:50:40 [] pump-loop.log: oref0-pump-loop failed. Listening for 40s silence before mmtuning: .............................No interfering pump comms detected from other rigs (this is a good thing!)

Scott Leibrand
@scottleibrand
Nov 22 2018 14:16
That sounds very similar to what openaps/oref0#1140 fixes. If you're running dev, are you up to date with that commit?
live4sw
@live4sw
Nov 22 2018 14:17
I'm on master, though I'd be happy to update docs to talk about this. But does that sound like a reasonable cause for an A52?
mrboletmrpot
@mrboletmrpot
Nov 22 2018 16:25
@Ebgineer @ecc1 I look for it, and in between, I tried the ancient openaps I made, it says the same error, but reboot on and on to fix it. And I'm note able to flash them with jubilinus. flashall.bat does not seems to work. The pump is a 754 canadian, it works from 2 month without problem, and the ancient openaps was working on a european 754, that died on the floor with several falls. And if needed I also can test the pump ability to be radio connected with loop and riley link, but I bet my hat it is still working. I post a full error text asap.
mrboletmrpot
@mrboletmrpot
Nov 22 2018 17:00
image.png
image.png
the ancient one wit radio cc111x located, but unknown error, on oref0 06...
How far have you gotten with those steps?
mrboletmrpot
@mrboletmrpot
Nov 22 2018 17:08
and the new one with its error cc111x
image.png
I discover it as you show me. But I've never had a successful command with cron, I know it's used to schedule things in edison, but I still discover plenty of things, I'm not used to it at all. It still is chinese to me, I have to learn.
mrboletmrpot
@mrboletmrpot
Nov 22 2018 17:13
-by the way, could a battery linked to the edison explorer board empeach it to be flashed with windows? Don't throw me rocks for my perhaps stupid questions, I'm all new to those electronic things-
Ebgineer
@Ebgineer
Nov 22 2018 17:14
don't worry, I won't throw rocks. I'm in the glass house with you. As I mentioned, I haven't had this error myself so I'm working from theory only.
I believe you said you're running an edison rig.
One of the first steps would be to check for a good electrical connection between the edison and the explorer board, by removing it and attaching it again.
mrboletmrpot
@mrboletmrpot
Nov 22 2018 17:19
yes, it's an edison. I did, it, I switched the edison chipset and motherboard for the two I have, same errors.
(they are edison in fact)
Ebgineer
@Ebgineer
Nov 22 2018 17:21
if the battery in the pump is low, the pump will keep running but shut off radio communication. so installing a new pump battery is another early troubleshooting step.
if you are getting radio errors with two different rigs that both used to work, it would tend to point more toward a pump radio problem rather than a problem with the radios on both rigs.
mrboletmrpot
@mrboletmrpot
Nov 22 2018 17:25
(I spoke about battery while trying to flash it, to wipe everithing and do it back to scratch no data, no programming problem I guess/hope. More work perhaps, but I've done it already)
I also changed the pump in between, it went broken.
Ebgineer
@Ebgineer
Nov 22 2018 17:27
the battery temperature errors in your first screenshot occurred during initial boot of the edison, so I believe they are related to the rig battery and not the pump battery
mrboletmrpot
@mrboletmrpot
Nov 22 2018 17:28
well, I tried to flash it again with flashall.bat without battery on the rig : no better result. it just reboots as usual.
Ebgineer
@Ebgineer
Nov 22 2018 17:29
is it stuck rebooting over and over again?
mrboletmrpot
@mrboletmrpot
Nov 22 2018 17:31
I missed that battery temperature error. special, I just ran it to test it.

it is not stuck, I reboots as usual. my dos session tells me this : C:\jubilinux>FLASHALL.BAT
Using U-boot target: edison-blankrndis
Now waiting for dfu device 8087:0a99
Please plug and reboot the board
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

C:\jubilinux>

-yesm the one with battery heat error reboots again and again
Ebgineer
@Ebgineer
Nov 22 2018 17:37
ideally you could interrupt the loop before it reboots and enter sudo service cron stop && killall -g oref0-pump-loop && shutdown -c
I'm not going to be much help for you getting your other rig flashed again. My edison broke about a year ago.
mrboletmrpot
@mrboletmrpot
Nov 22 2018 17:55
You reminded me something, speaking about battery heat. It's the rig with whom my wife reversed batterie's polarity when changing it. It heat up all of a sudden, and anly worked then on 5v supply on micro usb. It's why it does not like when battery is on it, even on a short period. (and if I remember well, it worked without battery and it could be a plan b for the night until the new explorer board)
Ebgineer
@Ebgineer
Nov 22 2018 17:56
That sounds like a good starting point.
Ebgineer
@Ebgineer
Nov 22 2018 18:07
On a different topic, has anyone had trouble installing dev branch on pi0 with an error about node 8.x being incompatible with armv6 hardware? You appear to be running on ARMv6 hardware. Unfortunately this is not currently supported by the NodeSource Linux distributions. Please use the 'linux-armv6l' binary tarballs available directly from nodejs.org for Node.js v4 and later.
mrboletmrpot
@mrboletmrpot
Nov 22 2018 18:07
well : I cannot access to it in serial, but I see it on wifi. I hope it is still good, just have to update it for nights
I never tried, cannot help.
Scott Leibrand
@scottleibrand
Nov 22 2018 18:28
Andrew: Yes, I suspect your A52 was caused by the time change during the prime.
live4sw
@live4sw
Nov 22 2018 18:45
thanks and happy thanksgiving!
i am thankful for openaps and you guys
mrboletmrpot
@mrboletmrpot
Nov 22 2018 18:52
the burned one keeps repeting this and rebooting : any idea of it's origin?
retry 0 (Do you need to run 'export RFSPY_RTSCTS=0' first?)
retry 0 (Do you need to run 'export RFSPY_RTSCTS=0' first?)
mrboletmrpot
@mrboletmrpot
Nov 22 2018 19:03
Ok, I'm so stupid : I think it was programmed to be linked to the ancient pump. I'll update it.
Cas Eliëns
@cascer1
Nov 22 2018 19:06
Hey, Happy Thanksgiving to all those celebrating :)
I've got an issue.. my cob seems to go down way too slow (about 20g per hour) and it's causing some scary SMB behavior
I was wondering where I might change the carb absorption rate
Running the most recent version of Dev btw
mrboletmrpot
@mrboletmrpot
Nov 22 2018 19:14
When I had that, I think it was with the max smb basal minutes=xx
And in worst case, I cut smb for cob, time to make trials safely
Cas Eliëns
@cascer1
Nov 22 2018 19:18
I feel like the root cause is the weird cob decay
Ebgineer
@Ebgineer
Nov 22 2018 19:20
There's a qualitative description of the decay algorithm in the glossary https://openaps.readthedocs.io/en/latest/docs/Resources/glossary.html?highlight=cob
mrboletmrpot
@mrboletmrpot
Nov 22 2018 19:25
at the beginning, I had also a weird smb, and it was because of a bad calibration process between different freestyle.
Dana Lewis
@danamlewis
Nov 22 2018 19:34
@cascer1 your carb ratio may be too low. Make sure also you're on later version where min5carbimpact is 8 and not 3

@live4sw 💜

happy thanksgiving everyone!

Cas Eliëns
@cascer1
Nov 22 2018 20:08
@danamlewis My carb ratio works fine for meal boluses, and autotune seems to agree with it. I'm not sure what min5carbimpact is
Martin Haeberli
@mhaeberli
Nov 22 2018 20:54
i will cross post to cgm-remote-monitor - but I’m trying to capture my mongo data before clearing collections and continuing (mongo db near or at full). But it’s failing (actually, I cleared the db s yesterday, so I was just trying to validate the process going forward. Advice sought. The data capture tool is failing for me at the moment… :-(
Cas Eliëns
@cascer1
Nov 22 2018 21:06
I just ran oref0-runagain to make sure it wasn't some weird setting, and it's still doing the odd carb absorption thing
entered 65 carbs over 4 hours ago and there are still 23 cob according to openaps
can't find min5carbimpact at all in the repo
Just see if it's at the default 8, the very earliest versions were 3
Cas Eliëns
@cascer1
Nov 22 2018 21:32
The setting isn't in my preferences file,I guess that means it's using the default?
Dana Lewis
@danamlewis
Nov 22 2018 22:38
Yep. Do you know what version you're using?
Martin Haeberli
@mhaeberli
Nov 22 2018 23:11
@cluckj - thanks for the debugging advice re cd ~/myopenaps; killall -g oref0-pump-loop; OREF0_DEBUG=1 oref0-pump-loop (I also tried with 2, vs 1). but it doesn’t tell me anything about cgm-loop.log
Martin Haeberli
@mhaeberli
Nov 22 2018 23:24

I know I’m not understanding the full context, but the only relevant matches to cgm-loop.log I seem to be able to find in oref0

./bin/oref0-setup.sh:            (crontab -l; crontab -l | grep -q "cd $directory-cgm-loop && ps aux | grep -v grep | grep -q 'openaps monitor-cgm'" || echo "* * * * * cd $directory-cgm-loop && ps aux | grep -v grep | grep -q 'openaps monitor-cgm' || ( date; openaps monitor-cgm) | tee -a /var/log/openaps/cgm-loop.log; cp -up monitor/glucose-raw-merge.json $directory/cgm/glucose.json ; cp -up $directory/cgm/glucose.json $directory/monitor/glucose.json") | crontab -

and

./bin/oref0-setup.sh:            (crontab -l; crontab -l | grep -q "cd $directory && ps aux | grep -v grep | grep -q 'openaps get-bg'" || echo "* * * * * cd $directory && ps aux | grep -v grep | grep -q 'openaps get-bg' || ( date; openaps get-bg ; cat cgm/glucose.json | jq -r  '.[] | \"\\(.sgv) \\(.dateString)\"' | head -1 ) | tee -a /var/log/openaps/cgm-loop.log") | crontab -
looking at master; I’ll try looking at dev
it’s almost as if the substitution for $directory or $directory-cgm-loop is wrong
Jon Cluck
@cluckj
Nov 22 2018 23:25
I'm not sure the debugging hooks are in cgm-loop
Martin Haeberli
@mhaeberli
Nov 22 2018 23:25
where would I even find the code for cgm-loop
locate oref0-cron-every-minute
/home/.rootfs/usr/bin/oref0-cron-every-minute
/home/.rootfs/usr/local/bin/oref0-cron-every-minute
/root/src/oref0/bin/oref0-cron-every-minute.sh
/root/src/oref0/node_modules/.bin/oref0-cron-every-minute
(which of these is operative?)
Jon Cluck
@cluckj
Nov 22 2018 23:27
Third one down?
Martin Haeberli
@mhaeberli
Nov 22 2018 23:28
empirically, that’s right
/usr/local/bin/oref0-cron-every-minute
(based on cd /root/myopenaps ; which oref0-cron-every-minute
trying by hand:
~/myopenaps# oref0-monitor-cgm
Failed to get string preference .cgm_loop_path
Not an openaps environment, run: openaps init
so it looks like I’m warm
Jon Cluck
@cluckj
Nov 22 2018 23:32
Yes
Is that checking preferences.json for cgm_loop_path and not finding it?
Martin Haeberli
@mhaeberli
Nov 22 2018 23:34
maybe; don’t know yet.
tried .cgm_loop_path=/root/myopenaps oref0-monitor-cgm but it says no such path or directory
so I’m ‘holding it wrong!'
Martin Haeberli
@mhaeberli
Nov 22 2018 23:42
looks like it: CGM_LOOPDIR="$(get_pref_string .cgm_loop_path)”
Jon Cluck
@cluckj
Nov 22 2018 23:45
Yes
Martin Haeberli
@mhaeberli
Nov 22 2018 23:45
but i’m still puzzling ...
Jon Cluck
@cluckj
Nov 22 2018 23:46
Put "cgm_loop_path": "/path/to/appropriate/dir/" in ~/myopenaps/preferences.json
Martin Haeberli
@mhaeberli
Nov 22 2018 23:47
ok thx; but I assume “path/to/appropriate/dir” would be like “/root/myopenaps/"
Jon Cluck
@cluckj
Nov 22 2018 23:47
I have no clue :laughing:
Martin Haeberli
@mhaeberli
Nov 22 2018 23:52
what is the purpose of cgm-loop, and why is this behavior different in dev vs master (I THINK it’s different in master)
so - that ‘solved’ the problem