rafaelfelix

#!/bin/bash
cat > registry-values.yaml << EOF
service:
  type: NodePort
  nodePort: 30000
EOF

helm init --wait && \
  helm upgrade --install --wait registry -f registry-values.yaml stable/docker-registry

rafaelfelix

#!/bin/bash
K8S_VERSION=v1.15.4
MINIKUBE_VMDRIVER=virtualbox

minikube start --cpus=4 --memory=4000mb --vm-driver=${MINIKUBE_VMDRIVER} --kubernetes-version=${K8S_VERSION}

rafaelfelix

#!/bin/bash
MIRROR=archive.apache.org
SPARK_VERSION=2.4.4
SPARK_VERSION_SUFFIX=-bin-hadoop2.7

curl -Lo spark.tgz https://${MIRROR}/dist/spark/spark-${SPARK_VERSION}/spark-${SPARK_VERSION}${SPARK_VERSION_SUFFIX}.tgz && \
  tar xvzf spark.tgz && mv spark-${SPARK_VERSION}${SPARK_VERSION_SUFFIX} spark
export SPARK_HOME=$(pwd)/spark

rafaelfelix

rules:
  - pattern: Hadoop<service=HBase, name=RegionServer, sub=Regions><>Namespace_([^\W_]+)_table_([^\W_]+)_region_([^\W_]+)_metric_(\w+)
    name: HBase_RegionServer_Regions_metric_$4
    labels:
      namespace: "$1"
      table: "$2"
      region: "$3"
  - pattern: Hadoop<service=HBase, name=RegionServer, sub=(TableLatencies|Tables)><>Namespace_([^\W_]+)_table_([^\W_]+)_metric_(\w+)
    name: HBase_RegionServer_$1_metric_$4
    labels:

rafaelfelix

#!/bin/bash

# set spark.master to k8s api host:port
sed -i "s/#spark.master.*/spark.master\ k8s\:\/\/https\:\/\/$KUBERNETES_SERVICE_HOST\:$KUBERNETES_PORT_443_TCP_PORT/g" $SPARK_HOME/conf/spark-defaults.conf

# set spark.kubernetes.driver.pod.name to $HOSTNAME (https://spark.apache.org/docs/latest/running-on-kubernetes.html#client-mode-executor-pod-garbage-collection)
sed -i "s/#spark.kubernetes.driver.pod.name.*/spark.kubernetes.driver.pod.name\ $HOSTNAME/g" $SPARK_HOME/conf/spark-defaults.conf

# set spark.driver.host to own pod name
THIS_POD_FQDN=$HOSTNAME.servicename.{{ .Release.Namespace }}.svc.cluster.local

rafaelfelix

{
  "StartAt": "Step1",
  "States": {
    "Step1": {
      "Type": "Task",
      "Resource": "arn:aws:lambda:region:account:function:step1",
      "ResultPath": "$.result",
      "Next": "Step2"
    },
    "Step2": {

rafaelfelix

def lambda_handler(event, context):
   return f'{event["bucket"]}/{event["model"]}/{event["market"]}'

rafaelfelix

{
  "StartAt": "Concatenate Strings",
  "States": {
    "Concatenate Strings": {
      "Type": "Task",
      "Resource": "arn:aws:lambda:region:account:function:concatenate_strings",
      "ResultPath": "$.preparedInput",
      "Parameters": {
        "market.$": "$.market",
        "model.$": "$.model",

rafaelfelix

import boto3
import os

# Read environment variables into a read-only dict-like object. Falls back to AWS SSM 
# in case the variable is not set
# inspiration: https://aws.amazon.com/blogs/compute/sharing-secrets-with-aws-lambda-using-aws-systems-manager-parameter-store/
class AWSLambdaConfig(object):
    def __init__(self, path):
        self._ssm_path = path
        self._ssm_client = boto3.client('ssm')

rafaelfelix

Keybase proof

I hereby claim:

• I am rafaelfelix on github.
• I am rafaelfc (https://keybase.io/rafaelfc) on keybase.
• I have a public key ASCGxsBdB_ZnmKH-lhFUuCs_geoXR236mbQElQs78L7LhQo

To claim this, I am signing this object: