Improve container runtime detection #32309
Conversation
|
I will whip up the test env (Windows Podman, Windows Docker, Windows both, Windows none, Linux none, Linux Docker, Linux Docker rootless, Linux Podman, Linux Podman rootless) and report back next week... |
I'll record the calls again. One is fine. One each class init or something like in the past...not so much. e.g. I usually run test suites en bloc, so things tend to add up... |
This comment has been minimized.
This comment has been minimized.
|
If it is decided we want this, it will need a rebase |
geoand
added
the
triage/needs-rebase
Fully resolve the container runtime once and for all. It costs us one more call when the rootless info is not needed but I prefer things to be resolved entirely once, rather than using patterns that are a bit brittle. Fixes quarkusio#32246
Karm
force-pushed
the
rootless-detection
branch
from
49d764b to
f2f24be
Compare
|
My Quarkus time was eaten by the AWT monster. I'm switching context to this now... I rebased the branch. |
|
✔️ The latest workflow run for the pull request has completed successfully. It should be safe to merge provided you have a look at the other checks in the summary. |
|
👍 https://karms.biz/pastebin/podman-linux-lkjfdlkfsh.txt
|
Fully resolve the container runtime once and for all. It costs us one more call when the rootless info is not needed but I prefer things to be resolved entirely once, rather than using patterns that are a bit brittle.
Fixes #32246
@Karm any chance you could have a look at this? I /think/ it's the last rabbit hole. It's probably worth having a look at the result, rather than the diff as the diff is not very readable. Things should be much simpler to follow.