Tomoya Fujita

@Voldivh thanks for checking! CI: 🤞 * Linux [](http://ci.ros2.org/job/ci_linux/18944/) * Linux-aarch64 [](http://ci.ros2.org/job/ci_linux-aarch64/13475/) * Windows [](http://ci.ros2.org/job/ci_windows/19694/)

@Voldivh @clalancette Ah, sorry 😓 i will restart the CI.

CI: * Linux [](http://ci.ros2.org/job/ci_linux/18949/) * Linux-aarch64 [](http://ci.ros2.org/job/ci_linux-aarch64/13480/) * Windows [](http://ci.ros2.org/job/ci_windows/19701/)

@Voldivh i think we need to rebase rclcpp, please check https://ci.ros2.org/job/ci_linux/18949/console.

@clalancette can you review and merge these PRs? they need to be go in at the same time, but i do not have merge permission for `system_tests`. - https://github.com/ros2/rcl/pull/1004 -...

Final? CI: * Linux [](http://ci.ros2.org/job/ci_linux/19006/) * Linux-aarch64 [](http://ci.ros2.org/job/ci_linux-aarch64/13531/) * Windows [](http://ci.ros2.org/job/ci_windows/19760/)

@iuhilnehc-ynos > Actual behavior I have the following result, it is slightly different from yours. (with https://github.com/ros2/ros2/commit/ca09b31ce2851773cf722e1fe1e64b70ee651fdd) ``` $ ros2 run examples_rclcpp_minimal_composition composition_composed \ --ros-args \ -r publisher_node:__ns:=/ns1 \ -r...

okay, now i see it. sorry i was too early to ask you.

adding dynamic memory allocation and hashmap operation would be issue for logging as mentioned before. in production, logging is really important to make sure that we can have sufficient information...

@clalancette friendly ping.