Sidecar consul envoy not balancing traffic

Consul
1

Hi All,

I has been setup consul version 0.30.0 and envoy 1.14.6 from helm on kubernet
this is my value.yaml

global:
  name: consul
  imageEnvoy: "envoyproxy/envoy-alpine:v1.14.6"
  imagePullSecrets:
    - name: regcred
client:
  grpc: true
connectInject:
  enabled: true
  centralConfig:
    enabled: "true"
  sidecarProxy:
    resources:
      requests:
        memory: "100Mi"
        cpu: "100m"
      limits:
        memory: "100Mi"
        cpu: "100m"
server:
  resources:
    requests:
      memory: "100Mi"
      cpu: "100m"
    limits:
      memory: "100Mi"
      cpu: "300m"

syncCatalog:
  enabled: true

on my service i just add 'consul.hashicorp.com/connect-inject': 'true'
i have some probelm, traffic from my apigateway not balancing to the my service pods.

on my apigateway manifest config for consul i just add

'consul.hashicorp.com/connect-inject': 'true'
'consul.hashicorp.com/connect-service-upstreams': myservice:1001,myservice:1002

Screenshot from 2021-07-14 09-29-15

How to trace root cause this issue, and solve the probelm

Thanks
andi

2

Hi, so if you exec into your API gateway and curl localhost:1001 what’s the response?

3

Hi,

Thanks for your respone,
sorry i’m foirget to give information from apigateway to backend service use grpc protocol

/tmp $ ./grpc_health_probe --addr=127.0.0.1:10016
status: SERVING

Thanks
Andi

4

Hi Ikysow,
I changed my values.yaml configuration to something like this,

global:
  name: consul
  imageEnvoy: "envoyproxy/envoy-alpine:v1.16.4"
  imagePullSecrets:
    - name: regcred
client:
  grpc: true
  extraConfig: |
    {"enable_central_service_config": false}

connectInject:
  enabled: true
  sidecarProxy:
    resources:
      requests:
        memory: "100Mi"
        cpu: "100m"
      limits:
        memory: "100Mi"
        cpu: "100m"

controller:
  enabled: true

server:
  resources:
    requests:
      memory: "100Mi"
      cpu: "100m"
    limits:
      memory: "100Mi"
      cpu: "300m"
  nodeSelector: |
  extraConfig: |
    {"enable_central_service_config": false}

syncCatalog:
  enabled: true

I followed the instructions from the tutorial https://www.consul.io/docs/k8s/crds/upgrade-to-crds#connect-service-protocol-annotation
and I created a servicedefaults.yaml

apiVersion: consul.hashicorp.com/v1alpha1
kind: ServiceDefaults
metadata:
  name: microapi-stream
spec:
  protocol: 'grpc'

and servicedefaults created

NAME              SYNCED   AGE
microapi-stream  True     12h

But after describe my service pods, service-defaults with grpc protocol doesn’t showing,

    EOF
      
      /bin/consul services register \
        /consul/connect-inject/service.hcl
      
      # Generate the envoy bootstrap code
      /bin/consul connect envoy \
        -proxy-id="${PROXY_SERVICE_ID}" \
        -bootstrap > /consul/connect-inject/envoy-bootstrap.yaml
      
      # Copy the Consul binary
      cp /bin/consul /consul/connect-inject/consul

How to configure and enable these servicedefaults, and ensure that the protocol used is grpc, not tcp

Thanks
Andi

5

Hi, you won’t see anything about grpc when you describe the service pod but under the hood Consul knows that your service is using gRPC.