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

6th
Apr 2018
Jon Cluck
@cluckj
Apr 06 2018 00:08
@juanjuanmejias_twitter awesome! congrats on the loop!!!
Martin Haeberli
@mhaeberli
Apr 06 2018 00:14
remind me … Warning: remote port forwarding failed for listen port xxxx when establishing tunnel from openaps to cloud server … what have I forgotten ? passwordless login via ssh already working that direction
alimhassam
@alimhassam
Apr 06 2018 00:20
probably means that port is already in use.
Martin Haeberli
@mhaeberli
Apr 06 2018 00:28
with multiple openaps instances, all going to pushover, does one get flooding of messages? or do they get consolidated?
Eric
@ecc1
Apr 06 2018 01:49
If anyone is looping using Medtronic CGM and has a spare rig to run a "packet capture" on, please let me know. I'm hoping to allow Dexcom CGM data to be graphed on Medtronic pumps by "spoofing" the Enlite transmissions, but I'm missing some low-level details. I'll provide the packet-capture program and whatever help you need.
Dana Lewis
@danamlewis
Apr 06 2018 01:50
@ecc1 😍
@juanjuanmejias_twitter yay!
Tim Gunn
@Tornado-Tim
Apr 06 2018 01:57
@ecc1 that sounds awesome!
LilDucky
@LilDucky
Apr 06 2018 02:04
@ecc1 I am back on deck at last. please send me details and I should be able to do it this weekend. I have a Pi rig with an RFM69 waiting.
I also have a spare 722, enlite transmitters and test plug / sof-sensors (AKA harpoons)
Paul Heltzel
@pheltzel
Apr 06 2018 03:17
@jimrandomh Thanks again for advice on this. I hadn't recorded the last cannula change in NS. More lessons learned. It's a few days old. But it's definitely possible it's a site problem. Going to get everything back to baseline. New site, new insulin, my old basal schedule. Low carb diet. Then start over. And I might start again just overnight first where things were going well and get started more slowly next time. Then add SMB. My IC ratio is higher than most also, 4. Hasn't changed since I was a kid but has worked well so far.
Martin Haeberli
@mhaeberli
Apr 06 2018 03:58
Radio check failed. mmeowlink.exceptions.CommsException: Could not get subg_rfspy state or version. Have you got the right port/device and radio_type? - but I just re-flashed radio w ccprog and rebooted . what am I missing?
Scott Leibrand
@scottleibrand
Apr 06 2018 04:03
Did you fully power cycle? Reseat the Edison?
Martin Haeberli
@mhaeberli
Apr 06 2018 04:04
have NOT fully power cycled - just rebooted - so will try that now.
not yet tried reseating edison - but will ...
Martin Haeberli
@mhaeberli
Apr 06 2018 04:11
still getting error; now suspecting board
i both erased AND reflashed the radio board...
Martin Haeberli
@mhaeberli
Apr 06 2018 04:22
@scottleibrand can you reconfirm the correct file?
ok and a related question - is there an obvious configuration error in oref0 install that can break this?
Scott Leibrand
@scottleibrand
Apr 06 2018 05:05
ERD has some tools in their repo for checking the radio chip
900MHz Explorer tools or similar.
kallnap
@kallnap
Apr 06 2018 05:17
@scottleibrand @danamlewis did you hear any rumours whether the Dexcom G6 will by capable of DIY looping or not? I assume interoperable does not per se mean DIY loopable.
Martin Haeberli
@mhaeberli
Apr 06 2018 05:28
@scottleibrand thx
Ivica Suran
@isuran
Apr 06 2018 05:33
Is it possible to change bluetooth mac, instead of running setup again, if I want to use new phone for bluetooth tethering?
Scott Leibrand
@scottleibrand
Apr 06 2018 06:01
yes, you can change it in crontab -e and pair the new one with bluetoothctl
@kallnap I hear there are some changes on the Bluetooth side of things, but it sounds like online looping via the Share servers should still work in the mean time
Martin Haeberli
@mhaeberli
Apr 06 2018 06:06
Although it -may- be needed to pair with that phone via bluetoothctld or whatever it’s called
Ivica Suran
@isuran
Apr 06 2018 06:14
thanls
thanks
Martin Haeberli
@mhaeberli
Apr 06 2018 06:46
Changes on Bluetooth side? Because new hardware?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 07:04
@cluckj You helped me get there. Thanks!
Now I'm having issues connecting to my rig via SSH at the office wi fi. I'm getting this every time I try to connect:
image.png
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 07:12
This is the IP I'm entering:
image.png
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 07:18
I can't connect via SSH at all. Any input is welcome
th122
@th122
Apr 06 2018 09:24
@juanjuanmejias_twitter well, since you grayed out everything, not much to be said about that. (considered the pc is on a private address range, I’m assuming the rig is as well).
  • broadcast addresses need to match, if they’re on the same subnet
  • if they’re not, they need a defaultroute that knows how to route between the subnets
  • you need to have sshd enabled on the rig - there are enough tutorials on the net how to configure that
  • you could test if you can ping or traceroute between both systems
  • you could run ssh with verbose debugging to see at which stage it fails
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 10:28
Hi @th122. I'm not very knowledgable in security, so I have probably erred on the side of security. If it's safe and useful, I'll be happy to share
kallnap
@kallnap
Apr 06 2018 10:38
@scottleibrand that is sad news regarding the G6. Online looping in Europe is not really a viable option, so it looks like I need to get a G5 from my health insurrance while they are still available.
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 10:56
Bcast for usb0 and for wlan0 don't match. Is this what you mean?
  • broadcast addresses need to match, if they’re on the same subnet
In case it helps, the SSH connection has worked in the past. It's possible that something has changed in the office network
Peter Jeschke
@peterjeschke
Apr 06 2018 11:25
Please post the IP address of your rig and the IP address of the PC you're connecting from. These two need to be on the same subnet. Don't worry, these addresses are internal so we can't abuse them
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 11:25
OK
Peter Jeschke
@peterjeschke
Apr 06 2018 11:26
Also try to ping the rig from your PC. Try entering ping <the address of your rig> in the command prompt of your PC
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 11:26
wlan0 inet addr:192.168.4.178 Bcast:192.168.4.255
Martin Haeberli
@mhaeberli
Apr 06 2018 11:27
@kallnap I missed the sad G6 news on this thread from @scottleibrand . Can someone elaborate?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 11:27
And from my PC: IPv4 Address. . . . . . . . . . . : 192.168.1.76
image.png
Peter Jeschke
@peterjeschke
Apr 06 2018 11:29
Your PC and your rig are on different subnets
You mentioned that you are in your office, it's likely that the connection will not be possible due to the setup of your network. Talk to your office administrator or try it at home.
Mike
@libxmike
Apr 06 2018 11:38
@juanjuanmejias_twitter this may be related to DHCP and dynamic IP you rig is getting, do you know your rig current IP? are you using your phone as hotspot or office WiFi?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:13
@libxmike It's my office Wi Fi. My rig's IP is 192.168.4.178
Mike
@libxmike
Apr 06 2018 12:15
can you do ping 192.168.4.178 , are you sure this the correct and current IP ?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:15
I can try to talk to IT on monday. Maybe they can help
@libxmike Let me do it live
image.png
image.png
Mike
@libxmike
Apr 06 2018 12:18
this is not your rig, this is your PC
wlan0
correct?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:18
Yes, the first screenshot is my rig. I ran ifconfig and got the IP
Then I pinged that IP from my PC
Mike
@libxmike
Apr 06 2018 12:19
what is "rig" for you? PC ?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:20
I mean that's a screenshot of my Putty app. The rig (the edison) is connected via serial to the PC
I want to connect via ssh
Mike
@libxmike
Apr 06 2018 12:21
did you configured SSH on edison ?
is SSH started?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:21
I don't know, I ran the setup script. Is that it?
Mike
@libxmike
Apr 06 2018 12:22
openAPS setup?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:22
Yes
Mike
@libxmike
Apr 06 2018 12:22
this is not enough
you just installed oref0 but you have to configure edison SSH
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:23
I think I did, let me check
Mike
@libxmike
Apr 06 2018 12:25
do you have Firewall in the office? it may block port 22 also SSH connection
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:25
@peterjeschke mentioned that my rig and my PC are on different subnets
Maybe that's it
Setting up SSH on the rig: is that in the openaps docs?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:28
My Edison came with Jubilinux. Is that enough?
Mike
@libxmike
Apr 06 2018 12:28
no
you have to set it up and configure
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:30
But I mean for the SSH thing. Are you suggesting I reflash the edison?
Mike
@libxmike
Apr 06 2018 12:30
I would suggest to go through the docs to get familiar with the setup if you not done it already
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:30
I have, I thought :worried:
Mike
@libxmike
Apr 06 2018 12:30
I'm not sure at what stage of the configuration you are
is this first/initial setup?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:31
My loop is working, and I've connected via SSH at the office in the past
Mike
@libxmike
Apr 06 2018 12:31
did it work before ?
how do you know edisons IP? where did you check?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:32
Yes, but I had to run the setup script twice, and I the network at the office may have changed
I connected via serial and ran ifconfig
Mike
@libxmike
Apr 06 2018 12:33
thats the problem in office environment, you have DHCP and your rig IP may and will change
and what is your PC IP?
192.168.4.xx ?
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:35
192.168.1.76
Mike
@libxmike
Apr 06 2018 12:35
and thats the problem
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:35
Different subnets?
Mike
@libxmike
Apr 06 2018 12:35
you on a different network
yes diff subnet
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:35
Yes, I've told you :smile:
Mike
@libxmike
Apr 06 2018 12:36
you wont be able to connect unless you network admin configures IP forwarding between nets
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:36
OK, I'll talk to him on Monday
Mike
@libxmike
Apr 06 2018 12:36
but this may be security issue
Juan Mejías
@juanjuanmejias_twitter
Apr 06 2018 12:36
Thanks a lot
Mike
@libxmike
Apr 06 2018 12:37
no problem
James Babcock
@jimrandomh
Apr 06 2018 14:19
Something I noticed, wondering if other people also see this pattern. At 11pm I have a high-carb high-fat slow-digesting meal and matching meal bolus. BG goes up a bit, crests, BG starts falling and OpenAPS starts zero-temping, and I get a carbs-required alert because it isn't digesting fast enough to keep up with the insulin in the meal bolus. I have some sugar to avert the low, then it starts rising again, goes high, and doesn't get back in range until 8am.
NightscoutScreenshot.jpg
I believe traditionally this would be handled with a square or dual-wave bolus, but as I understand it OpenAPS doesn't mix with those very well
Luis Betancourt
@betluis
Apr 06 2018 14:22
hello all... quick question. Do we have any experience using the "zero figersticks" G6 CGM system from Dexcom with Openaps?
James Babcock
@jimrandomh
Apr 06 2018 14:23
Other odd thing: the temp basals OpenAPS sets for me seem to max out at 3.50U/hr, but the limits in the config are higher than that (base daily is 33.95, so 3x daily would be 4.24; overnight basal is 1.2, so 4x current would be 4.8. I think a slightly higher max might be a good idea.
As far as I know G6 hasn't shipped yet and no one has one
Luis Betancourt
@betluis
Apr 06 2018 14:26
could this be used with Openaps?
th122
@th122
Apr 06 2018 14:29
@jimrandomh I’m having the same problem - Both OpenAPS and AndroidAPS seem to limit the IOB to the amount that would be a safe to bring the BG back into range. I’ve been running AndroidAPS the last two weeks, and there I’m setting a TBR for the duration I would have set for the square part of the bolus.
James Babcock
@jimrandomh
Apr 06 2018 14:31
@betluis Expect a few weeks to a few months between when an OpenAPS developer gets one, and when non-developers are able to use them with OpenAPS. And there's some possibility of unforeseen problems that prevent them from ever being compatible.
Luis Betancourt
@betluis
Apr 06 2018 14:33
I see
thanks. If precision is good, this sounds like a very good option
Tore Bjørndalen
@BjoTore_twitter
Apr 06 2018 15:19
@jimrandomh I didn't see you registering any correcting carbs? That would give leverage for more SMBs on the way up. i.e. higher IOB.
I see this problem of getting the BG down when my ISF is too high. Autosens enabled?
Fat is something you need to figure out how affects you.
Add 20-40% carbs to your calculation for a later bolus maybe.
James Babcock
@jimrandomh
Apr 06 2018 15:27
Right, I didn't enter the carbs I ate in response to the carbs-required alert (it was 10g), but I see this happen without that, too. Autosens is indeed enabled.
Splitting boluses would be a lot easier if I could schedule it in advance, which square-bolus sort of does, rather than having to deal with setting a timer and responding to it
Tore Bjørndalen
@BjoTore_twitter
Apr 06 2018 15:31
Even with autosens enabled I see wrong ISF values mostly because I have too much carb absorption over a long time period and autosens is then not able to correct ISF.
Ok, test with a higher carb count initially and see how it performs using the same bolus up front.
James Babcock
@jimrandomh
Apr 06 2018 15:36
What do you mean by with a higher carb count initially?
Use the bolus wizard to enter a higher carb quantity and override the insulin amount to simulate a different carb ratio?
Tore Bjørndalen
@BjoTore_twitter
Apr 06 2018 15:38
Ok , you use the bolus wizard. Then reduce the bolus size it suggests.
James Babcock
@jimrandomh
Apr 06 2018 15:38
Will SMB just give the missing bolus component immediately?
That seems like something that could happen
Tore Bjørndalen
@BjoTore_twitter
Apr 06 2018 15:39
It can be allowed to be more aggressive
If no other constraints hit you. Check the OpenAPS pill.
Go RETRO mode and check the pill during this event
James Babcock
@jimrandomh
Apr 06 2018 15:41
The OpenAPS pill only shows me time since last loop; should it have more information than that?
Tore Bjørndalen
@BjoTore_twitter
Apr 06 2018 15:42
If ISF is a problem you could also set a low temp target to speed things up.
You can check the pill if you go back in time. You know retro mode in NS?
Slide back to different points in time and check the pill.
James Babcock
@jimrandomh
Apr 06 2018 15:44
When I go back the OpenAPS pill shows time of last loop prior to simulated-time, but nothing else
Tore Bjørndalen
@BjoTore_twitter
Apr 06 2018 15:49
Ok I made an example. Does it look like this to you?
Capture.JPG
James Babcock
@jimrandomh
Apr 06 2018 15:50
Ohh, it's in the hover popup
Yeah, I have the same info as that
Tore Bjørndalen
@BjoTore_twitter
Apr 06 2018 15:52
Step back in time and check what is going on then. Look at the predictions also
hover popup == pill :)
James Babcock
@jimrandomh
Apr 06 2018 15:58
At 4:15am (5 hours post-meal), BG is 215, COB is 0, IOB is 1.7U, all the predictions are that BG is going to come down--and then it doesn't. I can't see any reason to fault it for that prediction, though.
Does digestion slow down during highs? If so that would explain a few things
Scott Leibrand
@scottleibrand
Apr 06 2018 16:14

Sounds like your CR might be a bit too high/weak. What does autotune say about that?

You probably should reduce the upfront bolus to prevent the postprandial low and the need for correction carbs, and let SMB fill in as needed. If you still go high and autotune doesn’t recommend lowering CR, entering carb equivalents without an up front bolus might help.

You might also consider enableSMB_after_carbs to allow it to keep SMBing if COB goes to zero and you’re still absorbing carbs.
Also check what autotune says about ISF and basals.
Dana Lewis
@danamlewis
Apr 06 2018 16:18
Also, check what max basal rate is set to on your pump.
James Babcock
@jimrandomh
Apr 06 2018 16:22
Autotune agrees the CR should be stronger. I'll try having the problematic food/quantity again with the stronger CR, a reduced up-front bolus and enableSMB_after_carbs, earlier in the day so I'm not asleep, and see how it goes.
James Babcock
@jimrandomh
Apr 06 2018 16:31
Actually, rather than enableSMB_after_carbs I think I'll just set enableSMB_always. I see from a source-code comment that enableSMB_after_carbs extends 6h after any carb entry, and for me that's almost all of the time
Scott Leibrand
@scottleibrand
Apr 06 2018 16:50
:+1:
Dana Lewis
@danamlewis
Apr 06 2018 16:50
Check your max basal on your pump, too.
James Babcock
@jimrandomh
Apr 06 2018 16:59
Ah, yes, the max basal on the pump was 3.5, that's what was capping it
Martin Haeberli
@mhaeberli
Apr 06 2018 17:40
dumb question - would it be possible to ‘preflight’ - have orefX warn you in its logs , for example, that pump max basal and max_basal are not consistent with each other?
Dana Lewis
@danamlewis
Apr 06 2018 17:50
Max basal on the pump is read in automatically by the rig as part of settings read - it's not adjusted in the rig. What situation are you trying to solve for?
Tore Bjørndalen
@ToreBj
Apr 06 2018 18:02
@jimrandomh I agree with Scott that you should reduce the upfront bolus since you are probably not absorbing more than 20-30 carbs/h and your insulin bolus peaks after 1 hour which covers 72 g of carbs. Your CR is 10 and seems high.
statuur
@statuur
Apr 06 2018 18:30

@scottleibrand adjusted pump settings. Stangely enough it took some time for the previous errors to dissappear. But now i have new error in return which don't seem to add up.

root@edisonhost:~/myopenaps# oref0-pump-loop

Starting oref0-pump-loop at Fri Apr 6 20:24:03 CEST 2018 with 27 second wait_for_silence:
MDT CGM configured; not waiting
Radio ok. Listening: .No interfering pump comms detected from other rigs (this is a good thing!)
Preflight OK.
Attempting to retrieve MDT CGM data from pump
MDT CGM data retrieved
MDT New cgm data reformatted
Profile less than 60m old; Profile valid. Refreshed pumphistoryskipping bolus wizard entry with 30 g carbs and no insulin
skipping bolus wizard entry with 55 g carbs and no insulin
skipping bolus wizard entry with 20 g carbs and no insulin
skipping bolus wizard entry with 40 g carbs and no insulin
skipping bolus wizard entry with 0 g carbs and no insulin
skipping bolus wizard entry with 45 g carbs and no insulin
skipping bolus wizard entry with 25 g carbs and no insulin

Warning: setting mealCOB to 0 because currentDeviation is null/undefined
and meal.json
Checking pump clock: "2018-04-06T20:25:06+02:00" is within 1m of current time: Fri Apr 6 20:25:38 CEST 2018
Temp refreshed
Warning: Autotune has not been run. All microboluses will be disabled until you manually run autotune or add it to run nightly in your loop.
{"carbs":0,"nsCarbs":0,"bwCarbs":0,"mealCOB":0,"currentDeviation":null,"maxDeviation":0,"minDeviation":999,"slopeFromMaxDeviation":0,"slopeFromMinDeviation":999,"allDeviations":[],"lastCarbTime":0,"bwFound":false}
{"iob":1.18,"activity":0.0134,"basaliob":-0.005,"bolusiob":1.186,"netbasalinsulin":-0.1,"bolusinsulin":2.35,"time":"2018-04-06T18:25:09.000Z","iobWithZeroTemp":{"iob":1.18,"activity":0.0134,"basaliob":-0.005,"bolusiob":1.186,"netbasalinsulin":-0.1,"bolusinsulin":2.35,"time":"2018-04-06T18:25:09.000Z"},"lastBolusTime":1523032915000,"lastTemp":{"rate":0,"timestamp":"2018-04-06T16:30:12+02:00","started_at":"2018-04-06T14:30:12.000Z","date":1523025012000,"duration":30}}
{"delta":10,"glucose":164,"short_avgdelta":10.22,"long_avgdelta":9.58,"date":1522949520000}
Checking deliverAt: null is within 1m of current time: Fri Apr 6 20:25:45 CEST 2018
date: invalid date ‘null’
Couldn't smb_verify_suggested
oref0-pump-loop failed. {"reason":"If current system time Fri Apr 06 2018 20:25:45 GMT+0200 (CEST) is correct, then BG data is too old. The last BG data was read 1493.8m ago at Thu Apr 05 2018 19:32:00 GMT+0200 (CEST). Temp 0 <= current basal 0.275U/hr; doing nothing. "}
MDT CGM configured; not waiting
Unsuccessful oref0-pump-loop (BG too old) at Fri Apr 6 20:25:45 CEST 2018

root@edisonhost:~/myopenaps# openaps use pump iter_glucose_hours 1
[
{
"date_type": "none",
"packet_size": 0,
"_tell": 1,
"name": "DataEnd",
"op": "0x01"
},
{
"name": "GlucoseSensorData",
"date_type": "relative",
"_tell": 2,
"sgv": 242,
"date": "2018-04-06T20:22:00",
"packet_size": 0,
"op": 121
},
{
"name": "GlucoseSensorData",
"date_type": "relative",
"_tell": 3,
"sgv": 242,
"date": "2018-04-06T20:17:00",
"packet_size": 0,
"op": 121
},
{
"name": "GlucoseSensorData",
"date_type": "relative",
"_tell": 4,
"sgv": 240,
"date": "2018-04-06T20:12:00",
"packet_size": 0,
"op": 120
},
{
"name": "GlucoseSensorData",
"date_type": "relative",
"_tell": 5,
"sgv": 238,
"date": "2018-04-06T20:07:00",
"packet_size": 0,
"op": 119
},
{
"name": "GlucoseSensorData",
"date_type": "relative",
"_tell": 6,
"sgv": 236,
"date": "2018-04-06T20:02:00",
"packet_size": 0,
"op": 118
},
{
"name": "GlucoseSensorData",
"date_type": "relative",
"_tell": 7,
"sgv": 234,
"date": "2018-04-06T19:57:00",
"packet_size": 0,
"op": 117
},
{
"name": "GlucoseSensorData",
"date_type": "relative",
"_tell": 8,
"sgv": 232,
"date": "2018-04-06T19:52:00",

humalogwolfgang
@humalogwolfgang
Apr 06 2018 18:42
I'm trying to change my nightscout from Dexcom G4 to G5. that works well with the G5 upload. But with the bridge function does not work. The share connection created with the G5 app does not need a password at all. And which Bridge_server should I enter in Europe? Who can help?
David
@davidkeddydb
Apr 06 2018 19:01
Good afternoon, one of my rigs that was working fine for weeks, now isn't. For some reason the date/time on the rig is wacky. It is set now to Current default time zone: 'America/New_York'
Local time is now: Fri Dec 31 20:00:19 EST 1999.
Universal Time is now: Sat Jan 1 01:00:19 UTC 2000.
This is after I have run dpkg-reconfigure tzdata
Where should I start trying to resolve this date/time issue with that already tried?
philipgo
@philipgo
Apr 06 2018 19:04
@humalogwolfgang BRIDGE_SERVERis EU. I do not understand your password question. If you did not change your password, it should be the same as before.
James Babcock
@jimrandomh
Apr 06 2018 19:04
Assuming this is an Edison, if it loses all power it'll forget what time it is. There are mechanisms in OpenAPS that try to set the time from the network, or from a pump, so the first thing is to either give those a chance to work or to try to diagnose why those aren't working.
David
@davidkeddydb
Apr 06 2018 19:05
Sorry @jimrandomh is is an Edison with a Ti stick. It has been trying for the last 30 minutes but no luck
James Babcock
@jimrandomh
Apr 06 2018 19:05
@statuur Wherever your rig was getting BG data from, it isn't currently getting it.
David
@davidkeddydb
Apr 06 2018 19:06
It basically keeps cycling Pump clock is more than 55s off: attempting to reset it
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.
Error: pump clock refresh error / mismatch
James Babcock
@jimrandomh
Apr 06 2018 19:08
Ok, quick-fix to get back online is to set the system clock to the right time manually. Top answer on https://superuser.com/questions/302396/how-to-set-current-time-on-linux shows the right syntax.
Scott Leibrand
@scottleibrand
Apr 06 2018 19:09
What kind of network are you on? Many of them block NTP.
I think there might also be a bug preventing oref0 from getting time from the pump, if you want to try troubleshooting that.
David
@davidkeddydb
Apr 06 2018 19:11
I am running off a netgear box at work. I don't believe I have ntp blocked. I will look and check getting time from pump
th122
@th122
Apr 06 2018 20:33
@davidkeddydb if ntp failed to set the date/time at startup (ntp may have been up before networking had a usable connection), it probably is stymied by the time difference being way too large to be corrected by a change in slew. It tries to prevent “time jumps” on a running system unless explicitely told so.
see the manual page for ntpdate on how to set the date in a larger jump (use -b flag)
Scott Leibrand
@scottleibrand
Apr 06 2018 21:21
PRs welcome for making oref0 automatically use -b when appropriate
th122
@th122
Apr 06 2018 21:34
I’ll write an issue and tackle that when I’ve popped through my stack of open tasks (too many in parallel blocking each other, both *APS and work related)
Scott Leibrand
@scottleibrand
Apr 06 2018 21:35
:+1: