Apparent multiple callbacks after repetitive stop/start cycles about android-ibeacon-service

[jamesfalkner]

Also, this is with version 0.7.6 of the open source library on a Nexus 7.

[jamesfalkner]

Yeah, I pasted a large chunk of adb output (with timestamps) here:

https://gist.github.com/jamesfalkner/b777e3145f4e91b1020c

You can see in there where I am cycling monitoring/not monitoring and several big chunks where what appear to be multiple threads trying to start/stop scanning and reporting results.

[davidgyoung]

I believe I have a fix as of 1a2aa13

James, I have rolled a new build at the links below. Please try it when you have a chance and let me know if it resolves your issue.

Thanks,
David

http://s3.amazonaws.com/android-ibeacon-library.radiusnetworks.com/AndroidIBeaconLibrary-0.7.6.1.aar

http://s3.amazonaws.com/android-ibeacon-library.radiusnetworks.com/AndroidIBeaconLibrary-0.7.6.1.tar.gz

[jamesfalkner]

Hey David, just wanted to thank you for taking the time to check into and fix this.. Using 0.7.6.1 I no longer see the tons of scanLE attempts, however after a few cycles of starting/stopping monitoring and ranging, there still seems to be something lingering. Just before I call iBeaconManager.stopMonitoringBeaconsInRegion(r) in a loop (to stop monitoring all regions that the iBeaconManager thinks it is managing), I printed out the number of regions returned from iBeaconManager.getMonitoredRegions().size() - this size seems to be constantly going up (after several cycles of turning on/off monitoring, the size of iBeaconManager.getMonitoredRegions() is currently up to 66 and increasing).

What's odd is that even if I back out of and fully close my app, when I re-launch it, the number of regions reported by iBeaconManager.getMonitoredRegions() is huge (now up to 97), even before I turn on monitoring for any regions. Even if I exit app, toggle bluetooth off and back on, and restart app, it still thinks it is monitoring 100+ regions!

[davidgyoung]

OK, glad I fixed ONE problem :) Do you mind opening a new issue for this since it is a little different? I'll investigate that separately and close this one.

[jamesfalkner]

Haha, sure, no problem. I'll include some more debugging info in the next one :) Thanks again for your responsiveness!