Make c.networks() list include the default network

This makes things a lot more clear - if we are actually joining a
CNI network, we are guaranteed to get a non-zero length list of
networks.

We do, however, need to know if the network we are joining is the
default network for inspecting containers as it determines how we
populate the response struct. To handle this, add a bool to
indicate that the network listed was the default network, and
only the default network.

Signed-off-by: Matthew Heon <mheon@redhat.com>

4 files changed, 31 insertions, 23 deletions

diff --git a/libpod/container.go b/libpod/container.go
index 31c958959..e954d84eb 100644
--- a/libpod/container.go
+++ b/libpod/container.go
@@ -13,6 +13,7 @@ import (
 	"github.com/containers/image/v5/manifest"
 	"github.com/containers/podman/v2/libpod/define"
 	"github.com/containers/podman/v2/libpod/lock"
+	"github.com/containers/podman/v2/pkg/rootless"
 	"github.com/containers/storage"
 	"github.com/cri-o/ocicni/pkg/ocicni"
 	spec "github.com/opencontainers/runtime-spec/specs-go"
@@ -1095,13 +1096,17 @@ func (c *Container) Umask() string {
 // values at runtime via network connect and disconnect.
 // If the container is configured to use CNI and this function returns an empty
 // array, the container will still be connected to the default network.
-func (c *Container) Networks() ([]string, error) {
+// The second return parameter, a bool, indicates that the container container
+// is joining the default CNI network - the network name will be included in the
+// returned array of network names, but the container did not explicitly join
+// this network.
+func (c *Container) Networks() ([]string, bool, error) {
 	if !c.batched {
 		c.lock.Lock()
 		defer c.lock.Unlock()
 
 		if err := c.syncContainer(); err != nil {
-			return nil, err
+			return nil, false, err
 		}
 	}
 
@@ -1109,19 +1114,22 @@ func (c *Container) Networks() ([]string, error) {
 }
 
 // Unlocked accessor for networks
-func (c *Container) networks() ([]string, error) {
+func (c *Container) networks() ([]string, bool, error) {
 	networks, err := c.runtime.state.GetNetworks(c)
 	if err != nil && errors.Cause(err) == define.ErrNoSuchNetwork {
-		return c.config.Networks, nil
+		if len(c.config.Networks) == 0 && !rootless.IsRootless() {
+			return []string{c.runtime.netPlugin.GetDefaultNetworkName()}, true, nil
+		}
+		return c.config.Networks, false, nil
 	}
 
-	return networks, err
+	return networks, false, err
 }
 
 // networksByNameIndex provides us with a map of container networks where key
 // is network name and value is the index position
 func (c *Container) networksByNameIndex() (map[string]int, error) {
-	networks, err := c.networks()
+	networks, _, err := c.networks()
 	if err != nil {
 		return nil, err
 	}
diff --git a/libpod/container_internal.go b/libpod/container_internal.go
index 108954bad..823e5fb3a 100644
--- a/libpod/container_internal.go
+++ b/libpod/container_internal.go
@@ -641,18 +641,13 @@ func (c *Container) removeIPv4Allocations() error {
 		cniDefaultNetwork = c.runtime.netPlugin.GetDefaultNetworkName()
 	}
 
-	networks, err := c.networks()
+	networks, _, err := c.networks()
 	if err != nil {
 		return err
 	}
 
-	switch {
-	case len(networks) > 0 && len(networks) != len(c.state.NetworkStatus):
+	if len(networks) != len(c.state.NetworkStatus) {
 		return errors.Wrapf(define.ErrInternal, "network mismatch: asked to join %d CNI networks but got %d CNI results", len(networks), len(c.state.NetworkStatus))
-	case len(networks) == 0 && len(c.state.NetworkStatus) != 1:
-		return errors.Wrapf(define.ErrInternal, "network mismatch: did not specify CNI networks but joined more than one (%d)", len(c.state.NetworkStatus))
-	case len(networks) == 0 && cniDefaultNetwork == "":
-		return errors.Wrapf(define.ErrInternal, "could not retrieve name of CNI default network")
 	}
 
 	for index, result := range c.state.NetworkStatus {
diff --git a/libpod/networking_linux.go b/libpod/networking_linux.go
index 8dce7c9fe..7a0bebd95 100644
--- a/libpod/networking_linux.go
+++ b/libpod/networking_linux.go
@@ -110,10 +110,15 @@ func (r *Runtime) configureNetNS(ctr *Container, ctrNS ns.NetNS) ([]*cnitypes.Re
 
 	podName := getCNIPodName(ctr)
 
-	networks, err := ctr.networks()
+	networks, _, err := ctr.networks()
 	if err != nil {
 		return nil, err
 	}
+	// All networks have been removed from the container.
+	// This is effectively forcing net=none.
+	if len(networks) == 0 {
+		return nil, nil
+	}
 
 	// Update container map of interface descriptions
 	if err := ctr.setupNetworkDescriptions(networks); err != nil {
@@ -224,7 +229,7 @@ func (r *Runtime) setupRootlessNetNS(ctr *Container) error {
 	if ctr.config.NetMode.IsSlirp4netns() {
 		return r.setupSlirp4netns(ctr)
 	}
-	networks, err := ctr.networks()
+	networks, _, err := ctr.networks()
 	if err != nil {
 		return err
 	}
@@ -744,13 +749,13 @@ func (r *Runtime) teardownNetNS(ctr *Container) error {
 
 	logrus.Debugf("Tearing down network namespace at %s for container %s", ctr.state.NetNS.Path(), ctr.ID())
 
-	networks, err := ctr.networks()
+	networks, _, err := ctr.networks()
 	if err != nil {
 		return err
 	}
 
 	// rootless containers do not use the CNI plugin directly
-	if !rootless.IsRootless() && !ctr.config.NetMode.IsSlirp4netns() {
+	if !rootless.IsRootless() && !ctr.config.NetMode.IsSlirp4netns() && len(networks) > 0 {
 		var requestedIP net.IP
 		if ctr.requestedIP != nil {
 			requestedIP = ctr.requestedIP
@@ -863,7 +868,7 @@ func (c *Container) getContainerNetworkInfo() (*define.InspectNetworkSettings, e
 	settings := new(define.InspectNetworkSettings)
 	settings.Ports = makeInspectPortBindings(c.config.PortMappings)
 
-	networks, err := c.networks()
+	networks, isDefault, err := c.networks()
 	if err != nil {
 		return nil, err
 	}
@@ -872,7 +877,7 @@ func (c *Container) getContainerNetworkInfo() (*define.InspectNetworkSettings, e
 	if c.state.NetNS == nil {
 		// We still want to make dummy configurations for each CNI net
 		// the container joined.
-		if len(networks) > 0 {
+		if len(networks) > 0 && !isDefault {
 			settings.Networks = make(map[string]*define.InspectAdditionalNetwork, len(networks))
 			for _, net := range networks {
 				cniNet := new(define.InspectAdditionalNetwork)
@@ -893,7 +898,7 @@ func (c *Container) getContainerNetworkInfo() (*define.InspectNetworkSettings, e
 	}
 
 	// If we have CNI networks - handle that here
-	if len(networks) > 0 {
+	if len(networks) > 0 && !isDefault {
 		if len(networks) != len(c.state.NetworkStatus) {
 			return nil, errors.Wrapf(define.ErrInternal, "network inspection mismatch: asked to join %d CNI networks but have information on %d networks", len(networks), len(c.state.NetworkStatus))
 		}
@@ -1101,7 +1106,7 @@ func (c *Container) NetworkConnect(nameOrID, netName string, aliases []string) e
 		return err
 	}
 
-	ctrNetworks, err := c.networks()
+	ctrNetworks, _, err := c.networks()
 	if err != nil {
 		return err
 	}
diff --git a/libpod/rootless_cni_linux.go b/libpod/rootless_cni_linux.go
index 1d6158cc2..2c2977f9f 100644
--- a/libpod/rootless_cni_linux.go
+++ b/libpod/rootless_cni_linux.go
@@ -40,7 +40,7 @@ const (
 //
 // AllocRootlessCNI does not lock c. c should be already locked.
 func AllocRootlessCNI(ctx context.Context, c *Container) (ns.NetNS, []*cnitypes.Result, error) {
-	networks, err := c.networks()
+	networks, _, err := c.networks()
 	if err != nil {
 		return nil, nil, err
 	}
@@ -81,7 +81,7 @@ func AllocRootlessCNI(ctx context.Context, c *Container) (ns.NetNS, []*cnitypes.
 //
 // DeallocRootlessCNI does not lock c. c should be already locked.
 func DeallocRootlessCNI(ctx context.Context, c *Container) error {
-	networks, err := c.networks()
+	networks, _, err := c.networks()
 	if err != nil {
 		return err
 	}