iOS 11 causing issues #601
Comments
|
This was the second meal of the day. Again, BGs were flowing just fine...no error message was available in the Loop pill. Unfortunately, Anna messed with it before getting an issue report or any other info to help troubleshoot.
|
|
[see update] Not during a bolus...but here's the next failure. Captured an issue report.
My bad...it was after a bolus. Loop failed again after a bolus. what in the heck is going on?
|
|
Looks like your launchDate was right before generating the issue report, so Loop had stopped running. Do you have any crash reports? (You can find these in xcode under Windows -> Devices & Simulators -> View Device Logs |
|
Loop-2017-10-08-103823.ips.pdf this is from the crash report in my iphone, couldn't get it out of Xcode (or at least not as easily...it's been scanning for awhile now, apparently lots of logs) |
|
Google talks about this potentially being a CloudKit communications issue? The symptoms are that the local reading of BGs stops in Loop app, but Dexcom has BGs in the cloud and on the apps. Loop gets stuck...doesn't read locally, and doesn't go to the cloud. If i click on the BG value in Loop app, takes me to Share2app where everything looks good. As soon as I go back to Loop app, it briefly fetches from the cloud successfully and then goes back to normal local readings. I think my initial problems with this being coincidental with bolusing is just that...coincidental. It's now been reproduced without bolusing incidents |
|
I noticed this one at 12:56pm EDT. My Dexcom app was fine (no warning screens, and no missing data) but my loop was red for at least 20 mins, and when I tapped it, it said it was missing CGM data. I'll continue to upload them as I notice them, along with the state of all relevant apps. I'd be happy to collect more data if anyone feels something else would be beneficial - just let me know what to grab. |
|
I guess I spoke too soon. After about a day and a half on the new iPhone with 10.3.3 and Loop v1.5, just got the missing CGM data source/exclamation red loop again. BGs were flowing fine... Loop didn't have any crash logs in iphone or xcode. BUT, this time there is a Dexcom Share2 crash report in Xcode (not on iphone). The Loop lost cgm data at 8:12pm based on the HUD timestamp. The dexcom crash report says the launch time was 8:18pm. Of course, I know the app was launch by us a long time before that since it's been running all day. But the date/time of crash report is for 9:23pm...not sure how to interpret that as I don't remember anything happening at that particular time. We had Loop app back and running around 8:52pm based on the time stamp in the issue report.
|
|
BTW...just wanted to say I think this is very low priority. The red loop isn't overwhelming and I know how to easily get it back to green now super fast. Not happening enough to yield poor results or angry teen...so we are good here. I'm sure whatever it is can be worked out later (or will get fixed in iOS or such). Almost tempted to just close the issue altogether. I'll keep playing with things and see what happens. |
|
I don't know what this means, but I just watched Loop die right in front of my eyes at 10:05pm. It always seems to happen on the 5's - all of my crash reports end in either 0 or 5. I had it open on my lap because I was transcribing settings and it just went away. Re-opened it, and it had the exclamation mark next to the BG readings. Dexcom app was and is fine. I've attached the crash report. |
|
Over the past few days, I've been watching Analytics on my IPhone 6s running Loop 1.5 on iOS 10.3.2 (I've just realized I'd not updated even to 10.3.3), and I've found that Loop does crash occasionally, perhaps not as often as on iOS 11.x. Over the past 5 days my Loop produced 4 crash reports. Three of those look like this: Exception Type: EXC_BREAKPOINT (SIGTRAP) and the last one looks like this: Exception Type: EXC_BAD_ACCESS (SIGSEGV) It appears that the intermittent red Loop problems may not be related to Share2 iOS11 compatibility, or any Bluetooth issues, at least not directly. Looks like these problems might be caused by Loop crashing, possibly due to different reasons. The frequency of crashes on iOS prior to iOS11 might be low enough so that a user would not normally notice. |
|
I would agree that moving to iOS11 caused issues- very similar to what's described above. I updated to the latest version hoping some bugs might get fixed with the OS...bad idea. I've seen the Gray Loop now for 18 hours. Tried to push the loop again: Could not locate device support files. Looks like iOS 11.1 is not quite there yet. |
|
@schulzedj That isn't an issue with loop. You need to update your version of Xcode and build with the latest version of Loop following the instructions that Katie posted on facebook if you are using master branch. |
|
I’m going to close this issue because the updates to iOS appear to have been making this issue resolve for the most part. Still have occasional issues, but not nearly as many as before. |
We were using v1.5 with iOS 10.3.3 with no problems. Now, updated to iOS 11.0.1 and then 11.0.2 several days ago and having multiple daily issues.
Twice today we've had Loop go stale immediately after Anna has bolused. This is an example of this morning's situation, she bolused at 7:28am (that's the log time in NS). I captured my NS site, asked her to screenshot her Loop main screen, then she went to dexcom screen to screenshot that. By the time she got back to Loop app to get an issue report, the Loop had turned green again. The screenshots seem to indicate Loop lost CGM data, but NS was tracking data fine via bridge the whole time. We do have share account info entered into her app, but it did not apparently go to cloud for BG data.
I'll work on trying to get an issue report while this is happening. I do have mLab logging errors, but there are no time/date stamps on the error logs so I'm not entirely sure which would be matched with the issue.
Edited to add: Interestingly noticed no pump event for the bolus in the Issue report that was pulled later in the day. It skipped right over the 7:28am bolus
createdAt = "2017-10-05 07:43:29 +0000";
date = "2017-10-05 07:38:27 +0000";
doseType = tempBasal;
duration = 1800;
raw = <16019ba6 004511>;
title = "TempBasalDurationPumpEvent(length: 7, rawData: 7 bytes, duration: 30, timestamp: calendar: gregorian (fixed) year: 2017 month: 10 day: 5 hour: 0 minute: 38 second: 27 isLeapMonth: false )";
type = TempBasal;
unit = "U/hour";
uploaded = 1;
value = 0;
})
createdAt = "2017-10-05 07:23:43 +0000";
date = "2017-10-05 07:18:40 +0000";
doseType = nil;
duration = 0;
raw = <3300a892 00451100>;
title = "TempBasalPumpEvent(length: 8, rawData: 8 bytes, rateType: MinimedKit.TempBasalPumpEvent.RateType.Absolute, rate: 0.0, timestamp: calendar: gregorian (fixed) year: 2017 month: 10 day: 5 hour: 0 minute:";
type = nil;
unit = nil;
uploaded = 1;
value = nil;
The text was updated successfully, but these errors were encountered: