Switch to Humble/Garden on 22.04 #111
Conversation
Signed-off-by: Louise Poubel <louise@openrobotics.org>
Signed-off-by: Louise Poubel <louise@openrobotics.org>
Signed-off-by: Louise Poubel <louise@openrobotics.org>
Signed-off-by: Louise Poubel <louise@openrobotics.org>
Signed-off-by: Dharini Dutia <dharini@openrobotics.org>
Signed-off-by: Dharini Dutia <dharini@openrobotics.org>
- Update library and namespaces for ign -> gz migration. - Use `python -m em` rather than `empy` for macOS support. - Update dependencies to ros_gz_bridge and ros_gz_sim Signed-off-by: Rhys Mainwaring <rhys.mainwaring@me.com>
Signed-off-by: Dharini Dutia <dharini@openrobotics.org>
Signed-off-by: Dharini Dutia <dharini@openrobotics.org>
Signed-off-by: Michael Carroll <michael@openrobotics.org>
Signed-off-by: Michael Carroll <michael@openrobotics.org>
Signed-off-by: Michael Carroll <michael@openrobotics.org>
Signed-off-by: Michael Carroll <michael@openrobotics.org>
Signed-off-by: Michael Carroll <michael@openrobotics.org>
Signed-off-by: Michael Carroll <michael@openrobotics.org>
Signed-off-by: Michael Carroll <michael@openrobotics.org>
Signed-off-by: Michael Carroll <michael@openrobotics.org>
|
@mabelzhang @mjcarroll I think it works with cyclonedds? |
|
rerunning the jobs |
|
The PR tests succeeded, but the push tests failed. Is it flaky? I see it passing on the PR one. Maybe rerun? |
yeah seems to be flaky. I ran the tests once and everything passed. I re-ran them and this one failed with a timeout. On one of the tests I fixed a timeout by adding a sleep at the end after rclcpp shutdown and then |
|
Sounds promising |
Signed-off-by: Michael Anderson <anderson@mbari.org>
|
seems good? |
|
not sure how to remove the (fortress, galactic) required status |
|
@andermi @mabelzhang it seems like we can either add a branch filter in the workflow or just skip the workflow run temporarily, see here. What would be better? |
|
@quarkytale @mabelzhang well, I'm not sure why we have a check for fortress/galactic in this PR at all. Can we just get rid of that check? I can't find it in the workflow... |
|
Should we remove the extra rosbag things and the extra timeouts originally added for the default RMW, and retest to see if we can do without them? Other than following clean software engineering practice, the practical benefit is that if CI doesn't need to sleep extra time, CI will be quicker. |
Signed-off-by: Michael Anderson <anderson@mbari.org>
|
@mabelzhang I stopped generating rosbags for the two tests I enabled them for. There are only 1 second sleeps in two files I added right before an exe exits. I needed those even with the rmw change to cyclonedds. |
There was a problem hiding this comment.
Okay I've looked at all the files. One minor style comment.
Re timing - okay, sounds good. I see the rosbag being disabled.
I'm approving because I'm done for the week. But there are 2 followup tickets we need to do below.
This test does seem to be flaky. We can merge now because we know it is flaky, but we should open an issue ticket to track it, and so that we know it's flaky in the future. Either we can try in another PR to add more timeout, or disable it so that we're not always merging on red CI.
35 - launch_pc_commands_ros_feedback.launch.py (Failed)
Don't forget to change the github workflow line 31 back to main in a followup PR.
Co-authored-by: Mabel Zhang <mabel.m.zhang@gmail.com>
This reverts commit 93c8384.
|
It seems to be failing more consistently, and the only error is a print of I'll try to debug it a bit more on Monday |
|
Looks like it passed |
Get tests passing
merge along with:
osrf/mbari_wec_utils#38