I'm setting up a self-managed Kubernetes cluster where I'm limited by the number of choices for the backend storage for Pods.

I should use a remote storage, so that multiple Pods can read/write at the same time, and Pods can get migrated easily to other nodes. Of course the reliability of storage is important.

Due to hardware limitations, Ceph is not an option. GlusterFS is also depreciated.

NFS seems to be my only option. So my question is, how reliable is NFS for using in Kubernetes? Would applications (mostly Java-based) and databases (mysql, postgres) behave abnormally?

from googleapiclient import discovery
from oauth2client.client import OAuth2Credentials as creds
crm = discovery.build(
 'cloudresourcemanager', 'v3', 
  http=creds.authorize(httplib2.Http()))

project = crm.projects().get(name="projects/"+projectId).execute()
project_name = project['displayName']

These are the default https://github.com/terraform-linters/tflint-ruleset-terraform/blob/main/docs/rules/README.md for tflint. Most of these rules will have an example of how to trigger it and remediate it. A simple one you could use is the terraform_comment_syntax https://github.com/terraform-linters/tflint-ruleset-terraform/blob/main/docs/rules/terraform_comment_syntax.md .

You can trigger it by using a // comment inside a *.tf file.

On inspection

$ kubectl describe pod hello-minikube
$ kubectl get nodes

I found that I was running x86 container on ARM Macbook M1. Installed on x86 Linux and everything works as expected

https://github.com/kubernetes/minikube/issues/13129

Helm upgrade is making revisions to the deployment which means, you can roll back to the last deployment. regarding the restart of current pods, it will happen based on your definition of Deployment.spec.strategy.rollingUpdate will recreate the pods!

I'm having doubts when trying to run a query with JQ. I have an SQS queue with Tags applied and when I run

aws sqs list-queue-tags --region sa-east-1 --queue-url --output json | jq

returns the output:

{
  "Tags": {
    "owner": "foo",
    "Name": "bar-queue"
  }
}

So...I just filter using like this:

aws sqs list-queue-tags --region sa-east-1 --queue-url --output json | jq '.Tags[] | [.Name]'

or

aws sqs list-queue-tags --region sa-east-1 --queue-url --output json | jq '.Tags[].Name'

and other methods without success, always output: jq: error (at :6): Cannot index string with string "Name"

Maybe I'm having a noobie hahaha, but someone can help me?

I am having issues with my Kubernetes cluster daemonSet. I have deployed this yaml file on my cluster. It looks like the daemonSet is failing to pull the docker image from the registry.

apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: oci-la-fluentd-daemonset
  namespace: kube-system
  labels:
    app: oci-la-fluentd-logs
    version: v1
spec:
  selector:
    matchLabels:
      app: oci-la-fluentd-logs
      version: v1
  template:
    metadata:
      labels:
        app: oci-la-fluentd-logs
        version: v1
    spec:
      serviceAccountName: oci-la-fluentd-serviceaccount
      tolerations:
        - key: node-role.kubernetes.io/master
          effect: NoSchedule
      # Uncomment the following section if a secret is associated to pull the image
      imagePullSecrets:
      - name: regcred
      containers:
        - name: oci-la-fluentd-logs
          # Replace this value with actual docker image url
          image: iad.ocir.io/kobie/oke_logging/fluentd_oci_la
          # Replace this value with desired value for image pull policy
          imagePullPolicy: Always
          env:
            - name: awx_lower
              valueFrom:
                fieldRef:
                  fieldPath: spec.nodeName

As you could see I have attached the secrets on my yaml files to pull the image.

This is the error message. This is what I get on the cluster when I look at the pods.

Failed to pull image iad.ocir.io/kobie/oke_logging/fluentd_oci_la:latest rpc error: code = Unknown desc denied: Anonymous users are only allowed read access on public repos

I am in some desperate help. I am atempting to translate this docker run command:

docker run -d -p 5434:5432 --name postgres-2 livingdocs/postgres:14.4 standby -d "host=host.docker.internal port=5433 user=postgres target_session_attrs=read-write"

which works flawlessly, but when I attempt to convert it to docker_compose.yml:

  postgres-2:
    image: livingdocs/postgres:14.4
    container_name: postgres-SLAVE
    ports:
      - "5434:5432"
    volumes:
      - postgres2:/tmp/data
    entrypoint: "/scripts/entrypoint standby"
    command:
      - host=host.docker.internal
      - port=5433
      - user=postgres
      - target_session_attrs=read-write"

It throws:

pg_basebackup: error: too many command-line arguments (first is "host=host.docker.internal")
Try "pg_basebackup --help" for more information.

After I save the yml file and execute: docker-compose up -d --remove-orphans

I've tried the online conversion tools and about 300 variations, but I cannot get to function the same as docker run.

I was able to figure out how to update those records. This how the update_one command should be written:

collection_one.update_one(
                           {"group_wallet_id": ObjectId(data["group_wallet_id"])},
                                {
                                    "$set": {"update_at": now},
                                    "$inc": {"total_donations": data["amount"]},
                                    "$push": {"categories": {
                                        "category_id": category_id,
                                        "categories.$.donors_info": {
                                            "user_id": ObjectId(user_id),
                                            "total_balance": convert_to_decimal128(data["amount"]),
                                            "d_created_at": now,
                                            "d_updated_at": now
                                        },
                                        "c_created_at": now, "c_updated_at": now,
                                        "amount": convert_to_decimal128(data["amount"]),
                                    }}}, session=session)

All the operations should be wrapped in a {} as seen above.

How can I retrieve the root token from a KubeApps deployment?