Strange Logging Behaviour when using bt_enable(NULL)

Question

I am trying to run the peripheral_uart example on a custom nrf52833 board and the RTT logs get interrupted with corrupted output unless a long sleep is placed after bt_enable(NULL). This is the corrupted output: SEGGER J-Link V7.94a - Real time terminal output SEGGER J-Link (unknown) V1.0, SN=801036780 Process: JLink.exe [00:00:00.013,977] fs_nvs: 6 Sectors of 4096 bytes [00:00:00.014,038] fs_nvs: alloc wra: 0, fd0 [00:00:00.014,068] fs_nvs: data wra: 0, 1c [00:00:00.014,312] bt_sdc_hci_driver: SoftDevice Controller build revision: c5 93 ba a9 14 4d 8d 05 30 4e 9b 92 d7 71 1e e8 |.....M.. 0N...q.. aa 02 50 3c |..P< [00:00:00.019,592] bt_hci_core: HW Platform: Nordic Semiconductor (0x0002) [00:00:00.019,683] bt_hci_core: HW Variant: nRF52x (0x0002) [00:00:00.019,714] bt_hci_core: Firmware: Standard Bluetooth controller (0x00) Version 197.47763 Build 2370639017 [00:00:00.020,629] bt_hci_core: No ID address. App must call settings_load() [00:00:00.020,660] peripheral_uart: Bluetooth initialized [00:00:00.021,667] bt_hci_core: Identi0m [0m [00:00:38.922,607] peripheral_uart: Connected 6A:68:37:DD:F5:1A (random) The only place I could find a similar error was in this post where the identity line was also interrupted with a "0m[0m" however the post didn't focus on this error. After doing some searching I found this post where they found adding a 1ms sleep changed the logging behaviour however I required a much longer sleep than 1ms to get the intended behaviour in my case. When I add a 10000msec sleep after calling bt_enable(NULL) I get the intended logging behaviour: err = bt_enable(NULL); k_sleep(K_MSEC(10000)); if (err) { error(); } SEGGER J-Link V7.94a - Real time terminal output SEGGER J-Link (unknown) V1.0, SN=801036780 Process: JLink.exe [00:00:00.013,702] fs_nvs: 6 Sectors of 4096 bytes [00:00:00.013,732] fs_nvs: alloc wra: 0, fd0 [00:00:00.013,732] fs_nvs: data wra: 0, 1c [00:00:00.014,038] bt_sdc_hci_driver: SoftDevice Controller build revision: c5 93 ba a9 14 4d 8d 05 30 4e 9b 92 d7 71 1e e8 |.....M.. 0N...q.. aa 02 50 3c |..P< [00:00:00.019,195] bt_hci_core: HW Platform: Nordic Semiconductor (0x0002) [00:00:00.019,226] bt_hci_core: HW Variant: nRF52x (0x0002) [00:00:00.019,317] bt_hci_core: Firmware: Standard Bluetooth controller (0x00) Version 197.47763 Build 2370639017 [00:00:00.020,172] bt_hci_core: No ID address. App must call settings_load() [00:00:10.020,477] peripheral_uart: Bluetooth initialized [00:00:10.021,575] bt_hci_core: Identity: F8:53:36:C3:DB:07 (random) [00:00:10.021,636] bt_hci_core: HCI: version 5.4 (0x0d) revision 0x1102, manufacturer 0x0059 [00:00:10.021,697] bt_hci_core: LMP: version 5.4 (0x0d) subver 0x1102 [00:00:26.143,707] peripheral_uart: Connected 6A:68:37:DD:F5:1A (random) Any idea what is causing this and any idea to fix it without using such a long sleep.

Vidar Berg · Accepted Answer

I can only reproduce this issue with the 'peripheral_uart' sample if I connect to the RTTViewer after the application has booted up. This is because the RTT client does not connect until the buffer has been filled up. 
 Vidar Berg said: CONFIG_LOG_BUFFER_SIZE - increase default buffer size when using deffered mode. 
 I should have suggested this symbol instead: CONFIG_SEGGER_RTT_BUFFER_SIZE_UP. This is the symbol that controls the size of the RTT buffer in RAM.

Strange Logging Behaviour when using bt_enable(NULL)

Top Replies