As it looks like when we kill/stop geckodriver we do not wait for the process to be gone. As such the next re-use check for geckodriver fails in `poll()` still returning `None`, causing us to use the same instance again. But because the port was set to `None` already we fail the connection. I pushed a try build with an additional `proc.wait()` after joining the stdout/stderr threads: https://treeherder.mozilla.org/jobs?repo=try&landoInstance=lando-prod-2025&landoCommitID=65303
Bug 2052383 Comment 6 Edit History
Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.
As it looks like when we kill/stop geckodriver we do not wait for the process to be gone. As such the next re-use check for geckodriver fails in `poll()` still returning `None`, causing us to use the same instance again. But because the port was set to `None` already we fail the connection. I pushed a try build with an additional `proc.wait()` after joining the stdout/stderr threads: https://treeherder.mozilla.org/jobs?repo=try&landoInstance=lando-prod-2025&landoCommitID=65331