Delete CNI configuration file when SDN pod is stopped

The openshift sdn CNI configuration file /etc/cni/net.d/80-openshift-network.conf
is written on the node upon first time of sdn pod deployment and remains to be there
on the node during its lifetime. This makes multus to assume sdn is in ready state though
sdn pod is coming up during reboot scenarios. Because of this, for a pod delete request,
CNI DELETE request fails with connection refused error when sdn cni server socket is being
created, this may end up in entries for the pod may not get cleaned up (example: IP address
allocated for the pod by host-local IPAM plugin). Hence this commit cleans up the cni config
file as soon as sigterm signal is received, it makes multus to wait until sdn plugin is
actually ready and then invoke CNI requests.

Signed-off-by: Periyasamy Palanisamy <[email protected]>

Author: pperiyasamy

pkg/cmd/openshift-sdn-node/cmd.go

@@ -2,12 +2,13 @@ package openshift_sdn_node
 
 import (
 	"fmt"
-	"github.com/spf13/pflag"
 	"io"
 	"os"
 	"path/filepath"
 	"time"
 
+	"github.com/spf13/pflag"
+
 	"github.com/fsnotify/fsnotify"
 	"github.com/spf13/cobra"
 	"k8s.io/klog/v2"
@@ -120,6 +121,10 @@ func (sdn *openShiftSDN) run(c *cobra.Command, errout io.Writer, stopCh chan str
 	}
 
 	<-stopCh
+	err = sdn.deleteConfigFile()
+	if err != nil {
+		klog.Errorf("unable to delete sdn cni configuration file: %v", err)
+	}
 	time.Sleep(500 * time.Millisecond) // gracefully shut down
 }
 

pkg/cmd/openshift-sdn-node/sdn.go

@@ -2,6 +2,7 @@ package openshift_sdn_node
 
 import (
 	"io/ioutil"
+	"os"
 
 	corev1 "k8s.io/api/core/v1"
 	"k8s.io/client-go/kubernetes/scheme"
@@ -56,3 +57,8 @@ func (sdn *openShiftSDN) writeConfigFile() error {
 }
 `), 0600)
 }
+
+func (sdn *openShiftSDN) deleteConfigFile() error {
+	// Delete CNI config file from disk if it's no longer able to serve CNI requests.
+	return os.Remove(openshiftCNIFile)
+}