:farmer: wait_for_topic_launch_test failing in Rolling Zenoh for extra items in topic receive

[miguelgonrod]

Generated by Generative AI

No

Operating System:

Ubuntu Noble

ROS version or commit hash:

Jazzy Jalisco and Kilted Kaiju

RMW implementation (if applicable):

Zenoh

RMW Configuration (if applicable):

No response

Client library (if applicable):

No response

'ros2 doctor --report' output

ros2 doctor --report

<COPY OUTPUT HERE>

Steps to reproduce issue

1. Go to https://build.ros2.org/view/Kci/job/Kci__nightly-zenoh_ubuntu_noble_amd64 or https://build.ros2.org/job/Rci__nightly-zenoh_ubuntu_noble_amd64
2. Build one of them
3. See test failing

Expected behavior

Test not failing

Actual behavior

Test reporting a failure

Additional information

Reference builds:

• Rci__nightly-zenoh_ubuntu_noble_amd64#247 Failing since #236
• Kci__nightly-zenoh_ubuntu_noble_amd64#59 First fail in last 20 days

Test Failing:

• launch_testing_ros.test.examples.wait_for_topic_launch_test.wait_for_topic_launch_test

Rolling Log Output:

==============================
FAIL: wait_for_topic_launch_test.TestFixture.test_trigger_function
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/ws/src/ros2/launch_ros/launch_testing_ros/test/examples/wait_for_topic_launch_test.py", line 122, in test_trigger_function
    assert wait_for_node_object.wait(is_trigger_called)
AssertionError: assert False
 +  where False = <bound method WaitForTopics.wait of <launch_testing_ros.wait_for_topics.WaitForTopics object at 0x7fead80e7e90>>([True])
 +    where <bound method WaitForTopics.wait of <launch_testing_ros.wait_for_topics.WaitForTopics object at 0x7fead80e7e90>> = <launch_testing_ros.wait_for_topics.WaitForTopics object at 0x7fead80e7e90>.wait

Rolling Test Log Output:

----------------------------- Captured stderr call -----------------------------
test_topics_successful (wait_for_topic_launch_test.TestFixture.test_topics_successful)
All the supplied topics should be read successfully. ... ok
test_topics_unsuccessful (wait_for_topic_launch_test.TestFixture.test_topics_unsuccessful)
All topics should be read except for the 'invalid_topic'. ... FAIL
test_trigger_function (wait_for_topic_launch_test.TestFixture.test_trigger_function) ... [INFO] [1762284216.263001042] [_test_node_YN64F4N54J]: Trigger function called with argument: False
FAIL

======================================================================
FAIL: test_topics_unsuccessful (wait_for_topic_launch_test.TestFixture.test_topics_unsuccessful)
All topics should be read except for the 'invalid_topic'.
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/ws/src/ros2/launch_ros/launch_testing_ros/test/examples/wait_for_topic_launch_test.py", line 104, in test_topics_unsuccessful
    assert wait_for_node_object.topics_received() == expected_topics
AssertionError: assert {'chatter_0',..., 'chatter_4'} == {'chatter_0',..., 'chatter_4'}
  Extra items in the right set:
  'chatter_3'
  'chatter_1'

Kilted Log Output:

==========================================================================================================================================
FAIL: wait_for_topic_launch_test.TestFixture.test_topics_unsuccessful
------------------------------------------------------------------------------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/ws/src/ros2/launch_ros/launch_testing_ros/test/examples/wait_for_topic_launch_test.py", line 104, in test_topics_unsuccessful
    assert wait_for_node_object.topics_received() == expected_topics
AssertionError: assert {'chatter_0', 'chatter_3'} == {'chatter_0',..., 'chatter_4'}
  Extra items in the right set:
  'chatter_1'
  'chatter_4'
  'chatter_2'
  Use -v to get more diff

Flakyness last 15 days:

job_name	last_fail	first_fail	build_count	failure_count	failure_percentage
Rci__nightly-zenoh_ubuntu_noble_amd64	2025-11-12	2025-10-30	14	12	85.71
Kci__nightly-zenoh_ubuntu_noble_amd64	2025-11-01	2025-11-01	5	1	20.0

It is failing more consistent in Rolling as it have been happening for some builds, in Kilted it is inconsistent as I found only one time this fail in last 20 days
Diffs doesn't show any major changes, as it is eprosima changes, and some ros2 messages, vendors and cli changes, most of them for documentation.

Maybe an issolation issue? @cottsay

CC: @Crola1702

[kscottz]

@JEnoch we discussed this issue in the waffle meeting. We think it is RMW Zenoh related. Can you take a look please?

[JEnoch]

I tried to reproduce the test failure in a local Docker container - the test never failed, even after 1 hour loop on wait_for_topic_launch_test.py.

Looking at the logs I find weird that the test_topics_successful test before is successful, meaning that the 5 topics are well received, and just after the test_topics_unsuccessful no longer receives those same 5 topics...

Is it possible on the build farm to run only this wait_for_topic_launch_test.py test, activating RCL debug logs and Zenoh debug logs via RUST_LOG=debug ?

[kscottz]

@miguelgonrod can you get one of the build farmers to help @JEnoch with his request?

[miguelgonrod]

Sure, let me write to them