Fix CNI server state behavior on reload.

This patch fixes the behavior of CNI server state
reload on error. It addresses the two following issues.
- If the state file is corrupted and parsing errors,
we should return, not proceed as the parsed cniServerState
might contain partial data.
- When VPP re-programmation completes, we overwrite the
state file so that we remove pods that yielded errors or
for which the linux ns was removed.

Signed-off-by: Nathan Skrzypczak <[email protected]>

Author: sknat

calico-vpp-agent/cni/cni_server.go

@@ -191,6 +191,8 @@ func (s *Server) rescanState() {
 		if err != nil {
 			s.log.Errorf("Could not remove %s, %s", config.CniServerStateFilename, err)
 		}
+		// if the cniServerState file is corrupted, we remove it and give up.
+		return
 	}
 
 	s.log.Infof("RescanState: re-creating all interfaces")
@@ -216,6 +218,13 @@ func (s *Server) rescanState() {
 			}
 		}
 	}
+	err = model.PersistCniServerState(
+		model.NewCniServerState(s.podInterfaceMap),
+		config.CniServerStateFilename,
+	)
+	if err != nil {
+		s.log.Errorf("CNI state persist errored %v", err)
+	}
 }
 
 func (s *Server) DelRedirectToHostOnInterface(swIfIndex uint32) error {