Describe the bug
Background
Offboard mode test flight. Drone was briefly spinning (between ~T+1028 and T+1032after a offboard yaw setpoint was sent, but the vehicle recovered.
At T+1037.54, remote channel 11 (Kill switch) and channel 7 (mode switch) were actuated. On the remote, the kill switch was engaged and mode switch was shifted to Position. This is the first log timestamp to acknowledge the control inputs.
In addition, in the log file at 1037 (0:17:17), The following entry appears:
[commander] action_request lost, generation 17 -> 19
The flight controller continues running and logging until it was manually unplugged after reaching the ground.
Bug
The flight controller received two switch inputs but seemingly did not act on the mode switch.
We expected the flight controller to process both inputs. While I realize that the kill switch should take priority, I am creating this issue on the off chance the inverse result is also possible where the mode switch change is processed but not the kill switch one.
Flight Log / Additional Information
Flight log link: https://review.px4.io/plot_app?log=9fe4b6f4-b8da-4652-826b-86a5b4f80b6a
PX4 Version: v1.17.0
Hardware: CubePilot Cube Orange+
Vehicle Type: Octorotor Coaxial
Remote: CRSF (with MAVLink) on TBS Crossfire
MAVLink Shell ver all:
HW arch: CUBEPILOT_CUBEORANGEPLUS
PX4 git-hash: d6f12ad1c4f70ad3230afd7d86e971421e02fef4
PX4 version: Release 1.17.0 (17891583)
OS: NuttX
OS version: Release 11.0.0 (184549631)
OS git-hash: fb2fadf6f599c1406f052db013efd00a2518e72c
Build datetime: May 19 2026 17:03:57
Build uri: localhost
Build variant: default
Toolchain: GNU GCC, 13.2.1 20231009
PX4GUID: 000600000000323336303233510800200021
MCU: STM32H7[4|5]xxx, rev. V
Describe the bug
Background
Offboard mode test flight. Drone was briefly spinning (between ~
T+1028andT+1032after a offboard yaw setpoint was sent, but the vehicle recovered.At
T+1037.54, remote channel 11 (Kill switch) and channel 7 (mode switch) were actuated. On the remote, the kill switch was engaged and mode switch was shifted toPosition. This is the first log timestamp to acknowledge the control inputs.In addition, in the log file at
1037(0:17:17), The following entry appears:The flight controller continues running and logging until it was manually unplugged after reaching the ground.
Bug
The flight controller received two switch inputs but seemingly did not act on the mode switch.
We expected the flight controller to process both inputs. While I realize that the kill switch should take priority, I am creating this issue on the off chance the inverse result is also possible where the mode switch change is processed but not the kill switch one.
Flight Log / Additional Information
Flight log link: https://review.px4.io/plot_app?log=9fe4b6f4-b8da-4652-826b-86a5b4f80b6a
PX4 Version: v1.17.0
Hardware: CubePilot Cube Orange+
Vehicle Type: Octorotor Coaxial
Remote: CRSF (with MAVLink) on TBS Crossfire
MAVLink Shell
ver all: