error streaming container content for copy up into volume [...] : copier: get: globs [/rabbitmq] #9432
Labels
kind/bug
Categorizes issue or PR as related to a bug.
locked - please file new issue/PR
Assist humans wanting to comment on an old issue or PR with locked comments.
/kind bug
Description
Integration tests from our infra project on GitHub workflows have started to fail due to a pdoman regression. This can be reproduced with:
This only happens on some images (in particular this RabbitMQ one); e.g. docker.io/fedora:latest or quay.io/cockpit/tasks etc. are fine.
This is an Ubuntu 20.04 host (GitHub workflows only offers Ubuntu 18.04 or 20.04) with the semi-official kubic packages.
This uses system podman, as rootless podman does not work very well in that environment -- e.g. the kubic packages don't pull in slirp4network or fuse-overlayfs, so networking is broken by default and performance is very poor.
But interestingly this does work with rootless podman in this environment, the container starts and I see the RabbitMQ messages. This is surprising, the above sounds like a bug in the layer unpacking/handling, which at first sight should not be very dependent on system vs. rootless?
On my current F34 workstation I have podman-3.0.0-0.204.dev.gita086f60.fc34.x86_64 where this works fine. It's a slightly older version. I'll try this again in an F34 VM with latest podman.
Output of
podman version
:Output of
podman info --debug
:Package info (e.g. output of
rpm -q podman
orapt list podman
):Have you tested with the latest version of Podman and have you checked the Podman Troubleshooting Guide?
Yes
Additional environment details (AWS, VirtualBox, physical, etc.):
The text was updated successfully, but these errors were encountered: