I can’t get that to work
@scottleibrand The computer no longer recognizes my edsion
@rsilvers129 Gitter troubleshooting 101: paste what you're doing and seeing
@denglfddd "recognizes" how? what exactly are you doing and seeing?
@scottleibrand I think it should be something I deleted，When I plug in EDSION, the device menu will not show edison, so now I can not re-flash
what do you see on console over the UART port when you reboot it?
That console output looks fine. does it eventually boot up, or does it stop there?
It has been stuck here
I turned off my WiFi router, and still got -99 from 69cm away.
bytearray index out of range
Google gives me no info about what this might mean
It usually self resolves
Just noticed I hadn't been looping for an hour. The rig was only a few feet away, and since putting it back in my pocket with the pump, it is fine, but not sure why it didn't connect for so long
seems that sometimes it can be two rooms away no problem, and other times on the other side of the couch and no good
to switch from running my rig online w/ NS to offline with G5 plugged into edison, am I correct that all i have to do is plug it in and add a usb battery? I can't get it to work
Data cables? Plugged in the otg port? Etc
yep otg to g5 receiver and power in uart
What version explorer do you have?
1.2
@rsilvers129 you are absolutely sure the serial number is correct right ? That is really strange
if it matters im running 0.2.0-dev
@kensack It works when it is closer than about a foot.
nevermind- it seems this rig isnt working at all right now. time for some troubleshooting
@PopcornPam_twitter The other side of the couch would be great. If my pump is in my right pocket, and my Explorer is in my left pocket, it probably won't work. Same pocket works.
@danamlewis re-running setup fixed whatever issues I was having. The updates made on setup are nice. I changed the max_iob during setup but it wasn't reflected in my pref file (obviously I was able to update it, just thought you should know)
@rsilvers129 are you making the files to print your case public? looks good.
thx
@kenkotch what version of dev are you on?
openaps 0.2.0-dev
When did you last pull a copy of dev though, complete w npm global install etc?
hmm...not today. this rig has been in a drawer for a week or so, so maybe end of may for complete npm global install
Heh. Ok. It would be great if you can re-do dev on that rig and try it again to confirm if it kept the maxiob you have it in the script, based off the latest code - or not
Made several tweaks even in past week
sure. i'll let you know - if not tonight, tomorrow
:+1:
@danamlewis just re-ran. here's the issue- I use 4.5 which seems to work for my setup:
/root/src/oref0/bin/oref0-setup.sh: line 489: [[: 4.5: syntax error: invalid arithmetic operator (error token is ".5")
/root/src/oref0/bin/oref0-setup.sh: line 489: [[: 4.5: syntax error: invalid arithmetic operator (error token is ".5")
For maxiob or another setting?
maxiob. I also use 1/2's for other settings
K. Hm, was fairly sure we had tested and fixed .5s
Will you log an issue so can check it out further tomorrow?
(And thanks for repulling and rerunning, too, to confirm)
yep- im still new to this, do i log the issue here: https://github.com/openaps/oref0/issues
Yup, exactly. Explain you got a fresh dev and include the error output
k. on it
:+1:
A word of caution related to autosens settings. I tried widening the range for autosens to 0.6 - 1.4 about 2 weeks ago. In my case, this contributed to 3 consecutive low bg's during the night following my change. I dug through the logs and found clearly that autosens was adjusting and causing the over-aggressiveness. I dialed back autosens to 0.95 - 1.05 and haven't had any troubles like that since.
In the offline looping for dexcom CGM users I have reached to the 1- add cron entry to start the microservice at start up "@reboot python /home/root/.xdripAPS/xdripAPS.py" ....... how can I add this ??
One thing that may be nice is to not get a carbsreq pushover if it is less than ISF/10.
@ericfeibelman I will keep that in mind if I have a problem, but I found that I had to widen it to 1.5 or else it was not able to lower me enough at times.
@marafie google is your friend :wink: https://askubuntu.com/questions/609850/what-is-the-correct-way-to-edit-a-crontab-file
@rsilvers129 are you watching the oref0 repo? I just did openaps/oref0#494 instead.
The BT teething works with the phone
I followed the instruction mentioned on the Bluetooth tethering on Edison (optional) but when I restart the Edison it doesn't automatically connect
I followed the instruction mentioned on the Bluetooth tethering on Edison (optional) but when I restart the Edison it doesn't automatically connect
And the Bluetooth is not On
@ericfeibelman 5% is probably too narrow for autosens to do any good. Agreed that allowing it to detect 40% resistance might be a bit much, but why didn't you just revert it back to the defaults from before you widened it? Also, you can safely allow it to detect extra sensitivity without giving it the same leeway on resistance. The defaults already allow that somewhat: 0.7 to allow detecting 42% sensitivity (1/0.7) vs. only 1.2 to allow only 20% resistance.
@marafie What phone are you using and do you have wifi setup locally?
@tim2000s I am using Samsung S8 and yes I have a local wifi setup... I installed on my andriod auto BT
Seems to me the BT on the Edison does not work when reboot
I keep getting this error:"BG data is too old (it's probably this), or clock set incorrectly. The last BG data was read at Wed Jun 07 2017 06:50:23 GMT-0400 (EDT) but your system time currently is Fri Jun 09 2017 14:04:45 GMT-0400 (EDT)"}
Waiting for ntpd to synchronize... No!
ntpd did not synchronize.
Restarting ntp (via systemctl): ntp.service.
Waiting for ntpd to synchronize... No!
ntpd did not synchronize.
Setting system time to later of Fri Jun 9 14:05:00 EDT 2017 or "2017-06-09T14:03:20-04:00":
(date +%s; date -d 2017-06-09T14:03:20-04:00 +%s) | sort -g | tail -1 | while read line; do sudo date -s @$line; done;
Fri Jun 9 14:05:00 EDT 2017
Waiting for ntpd to synchronize... No!
ntpd did not synchronize.
Restarting ntp (via systemctl): ntp.service.
Waiting for ntpd to synchronize... No!
ntpd did not synchronize.
Setting system time to later of Fri Jun 9 14:05:00 EDT 2017 or "2017-06-09T14:03:20-04:00":
(date +%s; date -d 2017-06-09T14:03:20-04:00 +%s) | sort -g | tail -1 | while read line; do sudo date -s @$line; done;
Fri Jun 9 14:05:00 EDT 2017
Where and what should I be looking at to fix?
@jtsackdex it looks like your rig may not be able to connect to the internet to pull data (glucose) or sync it's date/time. Can you ping google.com from a terminal session on your rig?
It's like every other run in the loop?
@ericmtx 144 packets transmitted, 144 received, 0% packet loss, time 143214ms
rtt min/avg/max/mdev = 9.492/29.590/136.213/18.317 ms
rtt min/avg/max/mdev = 9.492/29.590/136.213/18.317 ms
huh - it gets glucose data OK every other loop also?
and you've checked the time across your various devices to make sure they're all the same?
Yes like every other. Sometimes every 3rd
your nightscout is showing recent blood glucose readings?
(assuming you're pulling BG from NS in your rig)
Yes gets each change no problem within seconds of xDrip+
OK within 1st minute
Now it spit this out in the log
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 52, in call
self.epilog( )
File "/usr/local/bin/openaps-report", line 69, in epilog
super(ReportToolApp, self).epilog( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 75, in epilog
self.create_git_commit( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 98, in create_git_commit
self.repo.index.commit(msg)
File "/usr/local/lib/python2.7/dist-packages/git/index/base.py", line 954, in commit
author_date=author_date, commit_date=commit_date)
File "/usr/local/lib/python2.7/dist-packages/git/objects/commit.py", line 393, in create_from_tree
repo.head.set_commit(new_commit, logmsg=message)
File "/usr/local/lib/python2.7/dist-packages/git/refs/symbolic.py", line 211, in set_commit
self.set_object(commit, logmsg)
File "/usr/local/lib/python2.7/dist-packages/git/refs/symbolic.py", line 240, in set_object
return self._get_reference().set_object(object, logmsg)
File "/usr/local/lib/python2.7/dist-packages/git/refs/reference.py", line 64, in set_object
super(Reference, self).set_object(object, logmsg)
File "/usr/local/lib/python2.7/dist-packages/git/refs/symbolic.py", line 237, in set_object
return self.set_reference(object, logmsg)
File "/usr/local/lib/python2.7/dist-packages/git/refs/symbolic.py", line 308, in set_reference
fd = lfd.open(write=True, stream=True)
File "/usr/local/lib/python2.7/dist-packages/gitdb/util.py", line 333, in open
raise IOError("Lock at %r could not be obtained" % self._lockfilepath())
IOError: Lock at '/root/myopenaps/.git/refs/heads/master.lock' could not be obtained
{"bg":105,"temp":"absolute","snoozeBG":100,"recieved":true,"predBGs":{"IOB":[105,104,103,103,102,101,101,100,100,100,100,100,100]},"rate":0.55,"reason":"COB: 0, Dev: -5,
File "/usr/local/bin/openaps-report", line 82, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 52, in call
self.epilog( )
File "/usr/local/bin/openaps-report", line 69, in epilog
super(ReportToolApp, self).epilog( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 75, in epilog
self.create_git_commit( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 98, in create_git_commit
self.repo.index.commit(msg)
File "/usr/local/lib/python2.7/dist-packages/git/index/base.py", line 954, in commit
author_date=author_date, commit_date=commit_date)
File "/usr/local/lib/python2.7/dist-packages/git/objects/commit.py", line 393, in create_from_tree
repo.head.set_commit(new_commit, logmsg=message)
File "/usr/local/lib/python2.7/dist-packages/git/refs/symbolic.py", line 211, in set_commit
self.set_object(commit, logmsg)
File "/usr/local/lib/python2.7/dist-packages/git/refs/symbolic.py", line 240, in set_object
return self._get_reference().set_object(object, logmsg)
File "/usr/local/lib/python2.7/dist-packages/git/refs/reference.py", line 64, in set_object
super(Reference, self).set_object(object, logmsg)
File "/usr/local/lib/python2.7/dist-packages/git/refs/symbolic.py", line 237, in set_object
return self.set_reference(object, logmsg)
File "/usr/local/lib/python2.7/dist-packages/git/refs/symbolic.py", line 308, in set_reference
fd = lfd.open(write=True, stream=True)
File "/usr/local/lib/python2.7/dist-packages/gitdb/util.py", line 333, in open
raise IOError("Lock at %r could not be obtained" % self._lockfilepath())
IOError: Lock at '/root/myopenaps/.git/refs/heads/master.lock' could not be obtained
{"bg":105,"temp":"absolute","snoozeBG":100,"recieved":true,"predBGs":{"IOB":[105,104,103,103,102,101,101,100,100,100,100,100,100]},"rate":0.55,"reason":"COB: 0, Dev: -5,
@jtsackdex I assume you used xDrip as your CGM source in oref0-setup?
I get the "BG Data too old" error every so often too. I believe it's from when you switch back to wifi from BT- The rig looks for data directly from xDrip+, but doesn't see it. Since everything is still hitting nightscout, it pulls that history and gets it, and keeps looping.
I get the "BG Data too old" error every so often too. I believe it's from when you switch back to wifi from BT- The rig looks for data directly from xDrip+, but doesn't see it. Since everything is still hitting nightscout, it pulls that history and gets it, and keeps looping.
@dramageek That is how I set it up. I was actually trying to do the xDrip Offline and haven't been sucessful yet. I think I may see where I went wrong. i.e. enter bluetooth address instead of WiFi address Would I then remove wifi access to stop confusion?
@jtsackdex once you found out how to work the x drip offline via blutooth please do tell me how .... I got it working on local wifi no idea who it would use the blutooth
@marafie There is a lot of self teaching. I fully back and understand the reasoning. But it can be painful for someone just learning...
I have always learned better from mistakes
@marafie @jtsackdex I'm trying to find it. Is the readthedocs site acting up for anyone else?
Today I faced several times up and down
Were it mentions page does. It exist
You'll have to set xDrip to upload to nightscout using the old MongoDB method, then use REST for loading directly to the rig.
@dramageek my BT is not working
Ahaa
So if the rig is on BT it gets updates directly from xDrip+, if it's online it pulls from Nightscout but complains sometimes. Still works, though.
It works fine with local wifi
How far did you make it with the BT instructions? Whenever I end up doing them (after a reflash), it takes a bunch of times. Just keep following them, and if it tells you to go back and do something again, do it.
@dramageek Dram I followed the instructions many time... one of them was ok... but not sure why it didn't connect to the mobile ... I am not sure if blutooth is ok after reboot
Did you type y or yes @marafie
sorry late to the convo here but some phones won't allow bluetooth tethering. I'm in that category and struggled for days - weeks - before figuring it out... just fyi-ing y'all
I had a chance to watch my Edison boot while connected to a high resolution Tektronix current probe and a scope, so we could see the spikes in current usage. Although the rig idles at around 60 mA, and goes up to around 180 mA during my loop, I was quite surprised to see occasional spikes up to 750 mA (not correlated with looping). My hypothesis is that these are when the kernel is writing back data to the filesystems on eMMC, but I can't prove that yet.
The other thing that was easy to reproduce (by limiting the current on a bench power supply) was the Edison "boot loop", where it doesn't boot the kernel but keeps saying "waiting for application": that's due to an initial high current demand. If the USB port or cable can't supply that, the Edison exhibits that behavior which many of us have experienced.
@rsilvers129 Tried to do the openaps mmtune command as well. Explorer Board 1.2. 15cm distance "868.486", 5, -78. 30cm "868.390", 5, -82, 1m distance "868.438", 5, -82. Could be better but could also be worse.
@jtsackdex I types yes not y
Typed
@dramageek I did not realize this at first as the offline docs are not clear. You do not have to use the old MongoDB upload at all. The REST API can upload to both the rig and to NS. Enter your rig's local address AND your NS upload address in the REST API Base URL. The two address must be space delimited (type one, space, type the other). Works like a charm for me.
@dramageek I did this too two address one for local wifi and one for blutooth
@applehat showed me how to do that. http://secret@myaddress/api/v1/ http://secret@192.168.44.107:5000/api/v1
Should both of them have 5000/
???
???
During paring process I just got serviceresolved: yes paired: yes ..... then both of them became no
@marafie no. "http://secret@myaddress/api/v1/" points to your NS site and "http://secret@192.168.44.107:5000/api/v1" points to your rig's local address.
@danimaniac please PR in any improvements you can to the offline docs
@marafie Sweet! I was editing mine depending on if I was on my primary or backup rig. Great to know!
Once I get the BT solved I will move to the next step .. use it on my main pump
But be careful when setup URLs, for NS https, for xdrip BG uploading w/o "s" only http :exclamation:
Otherwise xdrip BG upload won't work
Not planning to use NS
Only FYI :wink:
Thanks will keep it in mind once I set it up for my kid after he kd out of the honey moon period
But recommended for reporting stuff :+1:
True... I am only lazy to go through it
I just set up NS + HTTPS + username/password auth on a virtual linux server - 3h total work (starting from a plain ubuntu vserver) -- now it could probably be done in way less than that since I managed to stumble into every pitfall possible ;)
(No azure/mongolab/whatever)
Will try to set it up once I have all other things ready
@danimaniac If you enter both addresses for n the REST api setting, when you're offline, and can't reach Nightscout, it stops the upload getting to your rig as well.
Hey Tim :wink: ...evening
👋
@tim2000s When my phone is offline? Or when my rig is offline?
When you are operating with no internet access.
Really the only reason to use xdripaps is so that you have glucose data when there isn't internet access.
I have the newest version of the explorer board on back-order. My current is from mid/late 2016 and it's having issues with the two red leds staying on and locking during mmtune from time to time. Anyone find that the newest board design is more stable/reliable with that? I've reflashed the radio a couple of times but still see the issue
hello all, I am starting to explore the functions of Oref1 and have a practical question. If a regular meal is announced by entering a bolo, what would be the ideal timing and amount of this bolo. Would this be some sort of pre-bolo?
also if the regular meal is announced instead by entering a rough amount of carbs, the question is how long in advance it is ideal to do this and then, right before the meal, the actual bolo should be reduced by certain amount?
@ericmtx I have four explorer boards also from mid/late 2016, no issues at all so far
@betluis that is very much a YDMV thing that is highly dependent on how you normally bolus for whatever you normally eat
but as a general rule, getting some insulin activity going before the meal helps limit the postprandial BG spike. that can mean an "eating soon" bolus or temp target, a regular pre-bolus, or pre-entry of carbs
if you do pre-enter the carbs, it's good to look at your resulting IOB and reduce your meal bolus accordingly
all of this advice is the same advice I'd give for using oref0 with AMA. oref1 isn't really any different in terms of what you need to do. the only real difference is it can get more insulin started faster if you pre-enter your carbs before you bolus