You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is this a BUG REPORT or FEATURE REQUEST? (leave only one on its own line)
/kind bug
Description
I would like to create a volume for a container that is owned by a specific user before it is mounted into a container.
When using the podman volume create --opt "o=uid=XXX,gid=YYY" test_volume2 command to pre-create the volume the UID/GID of the volume gets set correctly. The issue is that this is also seems to be forwarded as a mount option to the underlying mount with the local driver and it doesn't seem to support this.
The documentation seems to imply that when you use the local driver the uid/gid options will not be passed through to the underlying mount command because they are not supported. source
When not using the local driver, the given options will be passed directly to the volume plugin. In this case, supported options will be dictated by the plugin in question, not Podman.
# podman run --rm --subuidname=test_user --subgidname=test_user -v test_volume2:/test_volume2 alpine
Describe the results you received:
Podman run output:
Error: error mounting volume test_volume2 for container 4354295c9d8e075eb7295d1a9583da4e03ccb2971b49b90ac57034186f98f915: error mounting volume test_volume2: mount: /var/lib/containers/storage/volumes/test_volume2/_data: wrong fs type, bad option, bad superblock on , missing codepage or helper program, or other error.
Describe the results you expected:
I expected no output.
I also expected the lack of the line "o": "uid=231172,gid=231172" in the volume inspect output.
Example: When creating the volume inline with podman run --rm --subuidname=test_user --subgidname=test_user -v test_volume:/test_volume alpine
The volume inspect looks like:
Additional information you deem important (e.g. issue happens only occasionally):
All of this is run from the root account. The goal is to run user namespaced containers from the root account.
I included a 100 UID offset in the example. But this happens without it too. (I would like to ultimately keep the offset for actual use)
dmesg contains messages like:
[ 326.196858] fuseblk: Bad value for 'source'
Output of podman version:
Version: 3.1.2
API Version: 3.1.2
Go Version: go1.16.3
Built: Wed May 12 19:27:59 2021
OS/Arch: linux/amd64
Is this a BUG REPORT or FEATURE REQUEST? (leave only one on its own line)
/kind bug
Description
I would like to create a volume for a container that is owned by a specific user before it is mounted into a container.
When using the
podman volume create --opt "o=uid=XXX,gid=YYY" test_volume2
command to pre-create the volume the UID/GID of the volume gets set correctly. The issue is that this is also seems to be forwarded as a mount option to the underlying mount with the local driver and it doesn't seem to support this.The documentation seems to imply that when you use the local driver the uid/gid options will not be passed through to the underlying mount command because they are not supported. source
Steps to reproduce the issue:
Describe the results you received:
Podman run output:
Podman volume inspect:
Describe the results you expected:
I expected no output.
I also expected the lack of the line
"o": "uid=231172,gid=231172"
in the volume inspect output.Example: When creating the volume inline with
podman run --rm --subuidname=test_user --subgidname=test_user -v test_volume:/test_volume alpine
The volume inspect looks like:
Notably the Options are set to
{}
Additional information you deem important (e.g. issue happens only occasionally):
All of this is run from the root account. The goal is to run user namespaced containers from the root account.
I included a 100 UID offset in the example. But this happens without it too. (I would like to ultimately keep the offset for actual use)
dmesg contains messages like:
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? (https://github.com/containers/podman/blob/master/troubleshooting.md)
master
branch.Additional environment details (AWS, VirtualBox, physical, etc.):
Using libvirt based vagrant box:
fedora/34-cloud-base (libvirt, 34.20210423.0)
The text was updated successfully, but these errors were encountered: