This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

CC2652R: Scanning becomes unresponsive when connection is terminated

Part Number: CC2652R

If I keep scanning enabled when there is a connection, there will be no more GAP_EVT_ADV_REPORT when the connection is terminated.

To reproduce the issue, please follow the procedure below:

1. Enable scanning with duration 0(indefinite) on device A.
2. Make a connection from device A to device B. Don't disable scanning.
3. Check if device A application is still getting GAP_EVT_ADV_REPORT.
4. Turn off device B to terminate the connection by supervision timeout.
5. Check if device A application is still getting GAP_EVT_ADV_REPORT.

A simple workaround is to disable scanning and re-enable on device A. However, I'm afraid that there might be any side effect or more fundamental problem.

- Cetri

  • Hi Cetri,
    Thanks for sharing the details, we will reproduce and investigate internally.
  • Hello Cetri,
    I could reproduce this on simepl_central in simplelink_cc13x2_26x2_sdk_3_10_00_53. Interestingly after about 4 minutes or so after supervision timeout was reported (step 5.) the GAP_EVT_ADV_REPORT events was steadily reported again. Do you also observe this behavior?
  • Hi Eirik,

    No I didn't wait that long. Can you please check the workaround of disabling/re-enabling right away is safe?

    - Cetri

  • Hello Cetri,
    I have filed a bug ticket (ref. BLE_AGAMA-740) and will get back with a status update. To restart scan seems like a good solution.
  • Eirik,

    Unfortunately the workaround doesn't always work. Did you get any feedback from the development team? Can you please let me know ASAP if there is any update? This is an urgent issue to us. Thanks.

    - Cetri

  • Hello Cetri,
    How to you restart scan?
    Do you check the return code of GapScan_enable?
    Do you disable and then enable?

    I will discuss it this week with R&D.

    The issue is fixed for the next release targeted Q4 2019.

    The