podman stats: fix race when ctr process exists

stats read from the cgroup, and in order to know the cgroup we check the
pid for the cgroup. However there is a window where the pid exited and
podman did not yet updated its internal state. In this case the code
returns ErrCtrStopped so we should ignore this error as well.

Fixes #23334

Signed-off-by: Paul Holzinger <pholzing@redhat.com>

pkg/domain/infra/abi/containers.go

@@ -1581,7 +1581,14 @@ func (ic *ContainerEngine) ContainerStats(ctx context.Context, namesOrIds []stri
 			for _, ctr := range containers {
 				stats, err := ctr.GetContainerStats(containerStats[ctr.ID()])
 				if err != nil {
-					if queryAll && (errors.Is(err, define.ErrCtrRemoved) || errors.Is(err, define.ErrNoSuchCtr) || errors.Is(err, define.ErrCtrStateInvalid)) {
+					if queryAll &&
+						// All these errors might happen while we get stats, when we list all
+						// they must be skipped as they cause podman stats to stop and error otherwise.
+						// ErrCtrStopped can happen when the container process exited before we could
+						// update the container state
+						// https://github.com/containers/podman/issues/23334
+						(errors.Is(err, define.ErrCtrRemoved) || errors.Is(err, define.ErrNoSuchCtr) ||
+							errors.Is(err, define.ErrCtrStateInvalid) || errors.Is(err, define.ErrCtrStopped)) {
 						continue
 					}
 					return nil, err