Use EnableRmwIsolation in launch tests#724
Conversation
…covery" This reverts commit 88828e9. Signed-off-by: Scott K Logan <logans@cottsay.net>
Signed-off-by: Scott K Logan <logans@cottsay.net>
Yadunund
left a comment
There was a problem hiding this comment.
This is a great reference for updating the other repos! Thanks!
|
@Mergifyio backport kilted |
✅ Backports have been createdDetails
|
* Revert "Set envars to run tests with rmw_zenoh_cpp with multicast discovery" This reverts commit 88828e9. Signed-off-by: Scott K Logan <logans@cottsay.net> * Use EnableRmwIsolation in launch tests Signed-off-by: Scott K Logan <logans@cottsay.net> --------- Signed-off-by: Scott K Logan <logans@cottsay.net> (cherry picked from commit 2d283be)
|
@Yadunund @cottsay so we should apply the same changes else where like this? if that is the plan, i can help that. just a question, i see rmw_zenoh_cpp is unstable sometimes especially tests for |
|
@fujitatomoya that is indeed the plan! We would want to make similar changes for repos listed here ros2/ros2#1664
We did see some issues with ros2cli tests on multicast that go away when switching to isolated testing. But I'm unclear if the error you linked is a new regression or something we've seen before. Let's apply similar changes and see how it goes! |
|
Another good reference PR is ros2/system_tests#571. I've been plugging away at the ros2cli conversion. It's going to look a little different because the test themselves perform some communication - not just the launch side. So the isolation has to happen "sooner" in the process. |
This reverts (most of) #711 in favor of using the recently introduced
rmw_test_fixture.This should be backported to
kilted, but no further.