Scott Baker | e702d12 | 2019-10-22 11:54:12 -0700 | [diff] [blame] | 1 | /* |
| 2 | * Copyright 2018-present Open Networking Foundation |
| 3 | |
| 4 | * Licensed under the Apache License, Version 2.0 (the "License"); |
| 5 | * you may not use this file except in compliance with the License. |
| 6 | * You may obtain a copy of the License at |
| 7 | |
| 8 | * http://www.apache.org/licenses/LICENSE-2.0 |
| 9 | |
| 10 | * Unless required by applicable law or agreed to in writing, software |
| 11 | * distributed under the License is distributed on an "AS IS" BASIS, |
| 12 | * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 13 | * See the License for the specific language governing permissions and |
| 14 | * limitations under the License. |
| 15 | */ |
| 16 | |
| 17 | package afrouterd |
| 18 | |
| 19 | import ( |
| 20 | "fmt" |
| 21 | "github.com/golang/protobuf/ptypes/empty" |
Scott Baker | f579f13 | 2019-10-24 14:31:41 -0700 | [diff] [blame] | 22 | "github.com/opencord/voltha-lib-go/v2/pkg/log" |
Scott Baker | e702d12 | 2019-10-22 11:54:12 -0700 | [diff] [blame] | 23 | pb "github.com/opencord/voltha-protos/go/afrouter" |
| 24 | cmn "github.com/opencord/voltha-protos/go/common" |
| 25 | vpb "github.com/opencord/voltha-protos/go/voltha" |
| 26 | "golang.org/x/net/context" |
| 27 | "k8s.io/api/core/v1" |
| 28 | metav1 "k8s.io/apimachinery/pkg/apis/meta/v1" |
| 29 | "k8s.io/client-go/kubernetes" |
| 30 | "time" |
| 31 | ) |
| 32 | |
| 33 | func getVolthaPods(cs *kubernetes.Clientset) ([]*volthaPod, error) { |
| 34 | pods, err := cs.CoreV1().Pods(podNamespace).List(metav1.ListOptions{LabelSelector: podLabelSelector}) |
| 35 | if err != nil { |
| 36 | return nil, err |
| 37 | } |
| 38 | |
| 39 | var rwPods []*volthaPod |
| 40 | items: |
| 41 | for _, v := range pods.Items { |
| 42 | // only pods that are actually running should be considered |
| 43 | if v.Status.Phase == v1.PodRunning { |
| 44 | for _, condition := range v.Status.Conditions { |
| 45 | if condition.Status != v1.ConditionTrue { |
| 46 | continue items |
| 47 | } |
| 48 | } |
| 49 | |
| 50 | if group, have := v.Labels[podAffinityGroupLabel]; have { |
| 51 | log.Debugf("Namespace: %s, PodName: %s, PodIP: %s, Host: %s\n", v.Namespace, v.Name, v.Status.PodIP, v.Spec.NodeName) |
| 52 | rwPods = append(rwPods, &volthaPod{ |
| 53 | name: v.Name, |
| 54 | ipAddr: v.Status.PodIP, |
| 55 | node: v.Spec.NodeName, |
| 56 | devIds: make(map[string]struct{}), |
| 57 | backend: afrouterRWClusterName + group, |
| 58 | }) |
| 59 | } else { |
| 60 | log.Warnf("Pod %s found matching % without label %", v.Name, podLabelSelector, podAffinityGroupLabel) |
| 61 | } |
| 62 | } |
| 63 | } |
| 64 | return rwPods, nil |
| 65 | } |
| 66 | |
| 67 | func reconcilePodDeviceIds(ctx context.Context, pod *volthaPod, ids map[string]struct{}) { |
| 68 | ctxTimeout, _ := context.WithTimeout(ctx, time.Second*5) |
| 69 | conn, err := Connect(ctxTimeout, fmt.Sprintf("%s:%d", pod.ipAddr, podGrpcPort)) |
| 70 | if err != nil { |
| 71 | log.Debugf("Could not reconcile devices from %s, could not connect: %s", pod.name, err) |
| 72 | return |
| 73 | } |
| 74 | defer conn.Close() |
| 75 | |
| 76 | var idList cmn.IDs |
| 77 | for k := range ids { |
| 78 | idList.Items = append(idList.Items, &cmn.ID{Id: k}) |
| 79 | } |
| 80 | |
| 81 | client := vpb.NewVolthaServiceClient(conn) |
| 82 | _, err = client.ReconcileDevices(ctx, &idList) |
| 83 | if err != nil { |
| 84 | log.Errorf("Attempt to reconcile ids on pod %s failed: %s", pod.name, err) |
| 85 | return |
| 86 | } |
| 87 | } |
| 88 | |
| 89 | func queryPodDeviceIds(ctx context.Context, pod *volthaPod) map[string]struct{} { |
| 90 | ctxTimeout, _ := context.WithTimeout(ctx, time.Second*5) |
| 91 | conn, err := Connect(ctxTimeout, fmt.Sprintf("%s:%d", pod.ipAddr, podGrpcPort)) |
| 92 | if err != nil { |
| 93 | log.Debugf("Could not query devices from %s, could not connect: %s", pod.name, err) |
| 94 | return nil |
| 95 | } |
| 96 | defer conn.Close() |
| 97 | |
| 98 | client := vpb.NewVolthaServiceClient(conn) |
| 99 | devs, err := client.ListDeviceIds(ctx, &empty.Empty{}) |
| 100 | if err != nil { |
| 101 | log.Error(err) |
| 102 | return nil |
| 103 | } |
| 104 | |
| 105 | var ret = make(map[string]struct{}) |
| 106 | for _, dv := range devs.Items { |
| 107 | ret[dv.Id] = struct{}{} |
| 108 | } |
| 109 | return ret |
| 110 | } |
| 111 | |
| 112 | // coreMonitor polls the list of devices from all RW cores, pushes these devices |
| 113 | // into the affinity router, and ensures that all cores in a backend have their devices synced |
| 114 | func CoreMonitor(ctx context.Context, client pb.ConfigurationClient, clientset *kubernetes.Clientset) { |
| 115 | // map[backend]map[deviceId]struct{} |
| 116 | deviceOwnership := make(map[string]map[string]struct{}) |
| 117 | loop: |
| 118 | for { |
| 119 | // get the rw core list from k8s |
| 120 | rwPods, err := getVolthaPods(clientset) |
| 121 | if err != nil { |
| 122 | log.Error(err) |
| 123 | continue |
| 124 | } |
| 125 | |
| 126 | // for every pod |
| 127 | for _, pod := range rwPods { |
| 128 | // get the devices for this pod's backend |
| 129 | devices, have := deviceOwnership[pod.backend] |
| 130 | if !have { |
| 131 | devices = make(map[string]struct{}) |
| 132 | deviceOwnership[pod.backend] = devices |
| 133 | } |
| 134 | |
| 135 | coreDevices := queryPodDeviceIds(ctx, pod) |
| 136 | |
| 137 | // handle devices that exist in the core, but we have just learned about |
| 138 | for deviceId := range coreDevices { |
| 139 | // if there's a new device |
| 140 | if _, have := devices[deviceId]; !have { |
| 141 | // add the device to our local list |
| 142 | devices[deviceId] = struct{}{} |
| 143 | // push the device into the affinity router |
| 144 | setAffinity(ctx, client, deviceId, pod.backend) |
| 145 | } |
| 146 | } |
| 147 | |
| 148 | // ensure that the core knows about all devices in its backend |
| 149 | toSync := make(map[string]struct{}) |
| 150 | for deviceId := range devices { |
| 151 | // if the pod is missing any devices |
| 152 | if _, have := coreDevices[deviceId]; !have { |
| 153 | // we will reconcile them |
| 154 | toSync[deviceId] = struct{}{} |
| 155 | } |
| 156 | } |
| 157 | |
| 158 | if len(toSync) != 0 { |
| 159 | reconcilePodDeviceIds(ctx, pod, toSync) |
| 160 | } |
| 161 | } |
| 162 | |
| 163 | select { |
| 164 | case <-ctx.Done(): |
| 165 | // if we're done, exit |
| 166 | break loop |
| 167 | case <-time.After(10 * time.Second): // wait a while |
| 168 | } |
| 169 | } |
| 170 | } |