PRETTY_NAME="Ubuntu 24.04.5 LTS"
NAME="Ubuntu"
VERSION_ID="24.04"
VERSION="24.04.5 LTS (Noble Numbat)"
VERSION_CODENAME=noble
ID=ubuntu
ID_LIKE=debian
HOME_URL="https://www.ubuntu.com/"
SUPPORT_URL="https://help.ubuntu.com/"
BUG_REPORT_URL="https://bugs.launchpad.net/ubuntu/"
PRIVACY_POLICY_URL="https://www.ubuntu.com/legal/terms-and-policies/privacy-policy"
UBUNTU_CODENAME=noble
LOGO=ubuntu-logo
aarch64
Python 3.12.3
ros-jazzy-geometry-msgs	5.3.8-1noble.20260902.122255
ros-jazzy-rclpy	7.1.12-1noble.20260902.143245
ros-jazzy-rmw-fastrtps-cpp	8.4.4-1noble.20260902.140744
ros-jazzy-ros2bag	0.26.11-1noble.20260903.023509
ros-jazzy-rosbag2	0.26.11-1noble.20260903.023557
ros-jazzy-rosbag2-py	0.26.11-1noble.20260902.232852
ros-jazzy-rosbag2-storage-mcap	0.26.11-1noble.20260903.022942
ros-jazzy-rosbag2-transport	0.26.11-1noble.20260902.231544
ros-jazzy-rosgraph-msgs	2.0.4-1noble.20260902.131519

Validation date: 2026-10-12
Image: docker.io/library/ros:jazzy-ros-base
Pinned image: ros:jazzy-ros-base@sha256:066420e07f60aa18262f2479981def87ebcfcec42eefb0c0c57c4a46098348ca
Image OS/architecture: linux/arm64
Host: macOS ARM64, Docker Desktop
Network: none (source, recorder, player, and consumer inside one disposable container)
Read-only mount: exercise directory -> /lesson
Optional writable mount: separate artifact directory -> /output
ROS_DOMAIN_ID=73
RMW_IMPLEMENTATION=rmw_fastrtps_cpp

Two final-script end-to-end runs passed: one with retained artifacts and one
with default temporary cleanup. sample-output.txt is the captured stdout of
the default run. Both recordings contained exactly six PointStamped messages
with header and recording timestamps 100 through 105 seconds. In each run,
both replays delivered all six original payloads, frames, and headers. The
consumer clock was zero without /clock and initialized to 100 seconds before
advancing with the player's --clock 20 option.

An independent host-side check read the retained report.json and compared all
18 stored/replayed payloads with separately calculated fixture values. It also
checked recorded timestamps and the contrasting consumer clock behavior.
A second --output invocation on the same path exited with FileExistsError;
all existing artifact hashes were unchanged. The README's separate ros2 bag
info command also succeeded on the retained recording.

Periodic replay clock callbacks are not synchronized to point callbacks. An
earlier draft's strict upper clock bound of 105 seconds failed during a run
because consumer time can advance slightly past the final message stamp. The
final checks require initialization at 100 seconds, nondecreasing callback
time, and positive progress; they do not impose that incorrect upper bound
or require callback time to equal header time. Raw values are retained in
report.json and replay-with-clock.json when --output is used.

All subprocesses use finite wall-clock waits. Those waits are stall guards,
not latency measurements. The fixture does not estimate motion, run SLAM,
measure hardware behavior, or claim a general recording completeness result.
No dependencies were installed; all packages were present in the pinned image.
Docker --rm removed each completed verification container.