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
When I start a Podman container and attach to it, then exit the container and start it again, its non-loopback network interface is no longer present. When I start and stop the container without attaching to it the network interface is present after the next start, even if it was previously lost.
To Reproduce
Steps to reproduce the behavior:
Run a Podman container and attach to it, e.g.: podman run -it --name test debian.
Check the network interfaces: ip link. lo and tap0 are present.
Run exit.
Start the container again without attaching to it: podman start test.
Check the network interfaces: podman exec test ip link. Only lo is present.
Restart the container: podman stop test; podman start test (note that podman restart may exhibit a similar problem with older versions due to another Podman bug).
Check again: podman exec test ip link. Both lo and tap0 are present again.
Expected behavior
In step 5. above, tap0 interface is present.
Additional context
The problem is present in both 21.05 and unstable.
I'm reporting this issue on NixOS rather than on Podman because I couldn't reproduce it on other distributions.
I think that's an issue with the network handling in the rootless (slirp) case. We probably should move this topic to the Podman repository: https://github.com/containers/podman/issues/new
Describe the bug
When I start a Podman container and attach to it, then exit the container and start it again, its non-loopback network interface is no longer present. When I start and stop the container without attaching to it the network interface is present after the next start, even if it was previously lost.
To Reproduce
Steps to reproduce the behavior:
podman run -it --name test debian
.ip link
.lo
andtap0
are present.exit
.podman start test
.podman exec test ip link
. Onlylo
is present.podman stop test; podman start test
(note thatpodman restart
may exhibit a similar problem with older versions due to another Podman bug).podman exec test ip link
. Bothlo
andtap0
are present again.Expected behavior
In step 5. above,
tap0
interface is present.Additional context
The problem is present in both 21.05 and unstable.
I'm reporting this issue on NixOS rather than on Podman because I couldn't reproduce it on other distributions.
Notify maintainers
@marsam @adisbladis @saschagrunert @vdemeester @zowoq
Metadata
"x86_64-linux"
Linux 5.10.48, NixOS, 21.05.1408.9376bf7b342 (Okapi)
yes
yes
nix-env (Nix) 2.3.12
"nixos-21.05.1408.9376bf7b342, nixos-unstable-21.11pre301414.45fc7d4a35c"
/nix/var/nix/profiles/per-user/root/channels/nixos
Maintainer information:
The text was updated successfully, but these errors were encountered: