Editing configmap.yml and elasticsearch.yml in order to change password

[bgarcial]

Update elasticsearch default password.

Problem

I want to implement a jaeger installation with persistent storage using elasticsearch like backend on my Kubernetes cluster on Google cloud platform.

I am using the jaeger kubernetes templates and I start with elasticsearch production setup.

I've downloaded and modified the configmap.yml file in order to change the password field value and the elasticsearch.yml file in order to fix the password value which I've changed.

My customized .yml files has stayed of this way:

• configmap.yml

apiVersion: v1
kind: ConfigMap
metadata:
  name: jaeger-configuration
  labels:
    app: jaeger
    jaeger-infra: configuration
data:
  span-storage-type: elasticsearch
  collector: |
    es:
      server-urls: http://elasticsearch:9200
      username: elastic
      password: **my-password-value**
    collector:
      zipkin:
        http-port: 9411
  query: |
    es:
      server-urls: http://elasticsearch:9200
      username: elastic
      password:  **my-password-value**
  agent: |
    collector:
      host-port: "jaeger-collector:14267"

• elasticsearch.yml

apiVersion: v1
kind: List
items:
- apiVersion: apps/v1beta1
  kind: StatefulSet
  metadata:
    name: elasticsearch
    labels:
      app: jaeger
      jaeger-infra: elasticsearch-statefulset
  spec:
    serviceName: elasticsearch
    replicas: 1
    template:
      metadata:
        labels:
          app: jaeger-elasticsearch
          jaeger-infra: elasticsearch-replica
      spec:
        containers:
          - name: elasticsearch
            image: docker.elastic.co/elasticsearch/elasticsearch:5.6.0
            imagePullPolicy: Always
            command:
              - bin/elasticsearch
            args:
              - "-Ehttp.host=0.0.0.0"
              - "-Etransport.host=127.0.0.1"
            volumeMounts:
              - name: data
                mountPath: /data
            readinessProbe:
              exec:
                command:
                - curl
                - --fail
                - --silent
                - --output
                - /dev/null
                - --user
                - elastic:**my-password-value**
                - localhost:9200
              initialDelaySeconds: 5
              periodSeconds: 5
              timeoutSeconds: 4
        volumes:
          - name: data
            emptyDir: {}
- apiVersion: v1
  kind: Service
  metadata:
    name: elasticsearch
    labels:
      app: jaeger
      jaeger-infra: elasticsearch-service
  spec:
    clusterIP: None
    selector:
      app: jaeger-elasticsearch
    ports:
    - port: 9200
      name: elasticsearch
    - port: 9300
      name: transport

• And then, I've created the kubernetes cluster configuration with the new password value from my machine to my KGE via kubectl command

~/w/j/AddPersistVolumToPods ❯❯❯ kubectl create -f configmap.yml
configmap/jaeger-configuration created
~/w/j/AddPersistVolumToPods ❯❯❯

• And I've created the elasticsearch service via StatefulSet specialized pod (also with the new password value) from my machine to my KGE via kubectl command

~/w/j/AddPersistVolumToPods ❯❯❯ kubectl create -f elasticsearch.yml statefulset.apps/elasticsearch created service/elasticsearch created ~/w/j/AddPersistVolumToPods ❯❯❯

I can see that I have the elasticsearch service created on my GKE cluster

~/w/j/A/production-elasticsearch ❯❯❯ kubectl get services

NAME            TYPE        CLUSTER-IP    EXTERNAL-IP   PORT(S)             AGE
elasticsearch   ClusterIP   None          <none>        9200/TCP,9300/TCP   41m
kubernetes      ClusterIP   10.39.240.1   <none>        443/TCP             1h
~/w/j/A/production-elasticsearch ❯❯❯

And I have the elasticsearch-0 pod which have the docker container of elasticsearch service

~/w/j/A/production-elasticsearch ❯❯❯ kubectl  get pod elasticsearch-0
NAME              READY   STATUS    RESTARTS   AGE
elasticsearch-0   0/1     Running   0          25m
~/w/j/A/production-elasticsearch ❯❯❯

But when I can detail my pod on KGE, I see that my pod have some warnings and is not healthy ...

I get the pod description detail and I get this warning

Warning Unhealthy 2m6s (x296 over 26m) kubelet, gke-jaeger-persistent-st-default-pool-d72f7fde-ggrk Readiness probe failed:

Here, my entire output to describe command

~/w/j/A/production-elasticsearch ❯❯❯ kubectl describe pod elasticsearch-0
Name:           elasticsearch-0
Namespace:      default
Node:           gke-jaeger-persistent-st-default-pool-d72f7fde-ggrk/10.164.0.2
Start Time:     Tue, 08 Jan 2019 13:57:52 +0100
Labels:         app=jaeger-elasticsearch
                controller-revision-hash=elasticsearch-c684bb745
                jaeger-infra=elasticsearch-replica
                statefulset.kubernetes.io/pod-name=elasticsearch-0
Annotations:    kubernetes.io/limit-ranger: LimitRanger plugin set: cpu request for container elasticsearch
Status:         Running
IP:             10.36.2.7
Controlled By:  StatefulSet/elasticsearch
Containers:
  elasticsearch:
    Container ID:  docker://54d935f3e07ead105464a003745b80446865eb2417da593857d21c56610f704b
    Image:         docker.elastic.co/elasticsearch/elasticsearch:5.6.0
    Image ID:      docker-pullable://docker.elastic.co/elasticsearch/elasticsearch@sha256:f95e7d4256197a9bb866b166d9ad37963dc7c5764d6ae6400e551f4987a659d7
    Port:          <none>
    Host Port:     <none>
    Command:
      bin/elasticsearch
    Args:
      -Ehttp.host=0.0.0.0
      -Etransport.host=127.0.0.1
    State:          Running
      Started:      Tue, 08 Jan 2019 13:58:14 +0100
    Ready:          False
    Restart Count:  0
    Requests:
      cpu:        100m
    Readiness:    exec [curl --fail --silent --output /dev/null --user elastic:L4I^1xxhcu#S localhost:9200] delay=5s timeout=4s period=5s#success=1 #failure=3
    Environment:  <none>
    Mounts:
      /data from data (rw)
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-vkxnj (ro)
Conditions:
  Type           Status
  Initialized    True
  Ready          False
  PodScheduled   True
Volumes:
  data:
    Type:    EmptyDir (a temporary directory that shares a pod's lifetime)
    Medium:
  default-token-vkxnj:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-vkxnj
    Optional:    false
QoS Class:       Burstable
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute for 300s
                 node.kubernetes.io/unreachable:NoExecute for 300s
Events:
  Type     Reason                 Age                   From                                                          Message
  ----     ------                 ----                  ----                                                          -------
  Normal   Scheduled              27m                   default-scheduler                                             Successfully assigned elasticsearch-0 to gke-jaeger-persistent-st-default-pool-d72f7fde-ggrk
  Normal   SuccessfulMountVolume  27m                   kubelet, gke-jaeger-persistent-st-default-pool-d72f7fde-ggrk  MountVolume.SetUp succeeded for volume "data"
  Normal   SuccessfulMountVolume  27m                   kubelet, gke-jaeger-persistent-st-default-pool-d72f7fde-ggrk  MountVolume.SetUp succeeded for volume "default-token-vkxnj"
  Normal   Pulling                27m                   kubelet, gke-jaeger-persistent-st-default-pool-d72f7fde-ggrk  pulling image "docker.elastic.co/elasticsearch/elasticsearch:5.6.0"
  Normal   Pulled                 26m                   kubelet, gke-jaeger-persistent-st-default-pool-d72f7fde-ggrk  Successfully pulledimage "docker.elastic.co/elasticsearch/elasticsearch:5.6.0"
  Normal   Created                26m                   kubelet, gke-jaeger-persistent-st-default-pool-d72f7fde-ggrk  Created container
  Normal   Started                26m                   kubelet, gke-jaeger-persistent-st-default-pool-d72f7fde-ggrk  Started container
  Warning  Unhealthy              2m6s (x296 over 26m)  kubelet, gke-jaeger-persistent-st-default-pool-d72f7fde-ggrk  Readiness probe failed:
~/w/j/A/production-elasticsearch ❯❯❯

I go to the container log section on GCP and I get the following:

And in the audit log section I can see something like this:

resourceName: "core/v1/namespaces/default/pods/elasticsearch-0"     
response: {
   @type: "core.k8s.io/v1.Status"    
   apiVersion: "v1"    
   code: 500    
   
details: {…}    
   kind: "Status"    
   message: "The POST operation against Pod could not be completed at this time, please try again."    
   
metadata: {…}    
   reason: "ServerTimeout"    
   status: "Failure"    
  }
  serviceName: "k8s.io"   
  
status: {
   code: 13    
   message: "The POST operation against Pod could not be completed at this time, please try again."    
  }
 }

If I try with the original files and I change the password via KGE on GCP I get this error:

Pod "elasticsearch-0" is invalid: spec: Forbidden: pod updates may not change fields other than `spec.containers[*].image`, `spec.initContainers[*].image`, `spec.activeDeadlineSeconds` or `spec.tolerations` (only additions to existing tolerations)

After that I've create a pod, is not possible update or perform some changes? kubectl apply -f ..... ? I suposse

How to can I change the elasticsearch password?

If I want configure a persistent volume claim on this pod, can I perform this before the kubectl create -f command and my volume and mountPath will be created on container and KGE?

[pavolloffay]

After that I've create a pod, is not possible update or perform some changes? kubectl apply -f ..... ? I suposse

It depeds on the changes, you will have to restart pods in some situations.

How to can I change the elasticsearch password?

The authentication is managed by xpack. This document https://www.elastic.co/guide/en/elasticsearch/reference/current/security-api-change-password.html describes how to change the password.

If I want configure a persistent volume claim on this pod, can I perform this before the kubectl create -f command and my volume and mountPath will be created on container and KGE?

Just follow the standard k8s docs and it should work on GCE.

[bgarcial]

It depends on the changes, you will have to restart pods in some situations.

@pavolloffay What is the best strategic to restart the pod?

have been reading some posts blogs and speak about that, but I haven't clear how to restart the pods, because many sites talk about it depending of the way in that they has been created, and when I delete some pods via kubectl GKE inmediately create other pod with similar characteristics.

Even recommend go to the specific container (s) inside the pod and restart or kill the process container inside the pod, like strategic to restart ..https://stackoverflow.com/a/47799862/2773461 (edited)

[jpkrohling]

What is the best strategic to restart the pod?

Just kill the pod and let the Kubernetes controller start a new one. You might lose any inflight data, like spans that were received but not persisted.

Ideally, we would implement a file watcher in Jaeger itself, to restart the internal process without the need to kill the pod, but I don't think we have it in place already.

[bgarcial]

@jpkrohling I will take it in my mind when I need it.