RE: sd_ble_gap_scan_start() returns NRF_SUCCESS, but no advertising packets get received

Jan Panier — Fri, 30 Nov 2018 08:46:49 GMT

Hi,

There has been some busy time between then and now, but I'm glad for your reaction and wanted to answer your questions:

1. Well, it did. By reorganising my project and learning of the 'scan timeout == connection timeout' anomaly (my scan timeout was 0, my connection timeout != 0) I'm now down to it happening less than once a day. But it might still happen, we currently are in the middle of some longterm testing and one device 'froze' up while scanning. Still trying to reproduce it.

2. No, sometimes it happened 5 minutes into the test, sometimes it was running for hours before showing that behaviour.

3. Yes, I had started a minimal project that was just scanning, connecting, disconnecting in a cycle as fast as possible. That's when I learned about the timeout anomaly mentioned in (1). I took this as a base for reimplementing my main project.

4. I could share my minimal test project. But that didn't freeze up as often as the main project did. Unfortunately, I can't share everything, but as soon as I have isolated the cause, I am willing to share the reproducing code with you.

RE: sd_ble_gap_scan_start() returns NRF_SUCCESS, but no advertising packets get received

MartinBL — Tue, 13 Nov 2018 14:07:23 GMT

Hi,

I have never seen anything like this and I can't find any similar reports in our system.

It could sound like you end up in some sort of dead lock blocking your scanner.

Does this happen every time?
Does it always happen after ~30min?
Have you tried narrowing it down a bit and remove all features not essential to scanning?
It could be useful to reproduce this here. Do you mind sharing your code or describe how I can reproduce it? We can make the case private if you prefer confidentiality.