publish-resource-regions.sh

#!/bin/bash
#
# Publish a resource to all regions
#
# Run this from the resource folder

regions_to_publish=(us-east-1 af-south-1 ap-east-1 ap-northeast-1 ap-northeast-2 ap-northeast-3 ap-south-1 ap-southeast-1 ap-southeast-2 ap-southeast-3 ca-central-1 eu-central-1 eu-north-1 eu-south-1 eu-west-1 eu-west-2 eu-west-3 me-central-1 me-south-1 sa-east-1 us-east-2 us-west-1 us-west-2 eu-central-2 eu-south-2 ap-south-2)

# Use this to test succeed-fail locally
#regions_to_publish=(us-east-1 us-seattle)

successes=()
failures=()

cfn validate
cfn generate

# Create the package
echo "About to run cfn submit --dry-run to create the package"
echo ""
cfn submit --dry-run
echo ""

TYPE_NAME=$(cat .rpdk-config | jq -r .typeName)
TYPE_NAME_LOWER="$(echo $TYPE_NAME | sed s/::/-/g | tr '[:upper:]' '[:lower:]')"
# For example, awscommunity-s3-deletebucketcontents
echo "TYPE_NAME_LOWER is $TYPE_NAME_LOWER"

ZIPFILE="${TYPE_NAME_LOWER}.zip"
echo "ZIPFILE is $ZIPFILE"

ACCOUNT_ID=$(aws sts get-caller-identity|jq -r .Account)
echo "ACCOUNT_ID is $ACCOUNT_ID"

HANDLER_BUCKET="cep-handler-${ACCOUNT_ID}"

# We only need to copy the handler zip once, since it's not regional
# We use this bucket for logs also. The only drawback is if there are 
# failures, it can be hard to sort out which log file belongs to which region.

echo "Copying schema package handler to $HANDLER_BUCKET"
aws s3 cp $ZIPFILE s3://$HANDLER_BUCKET/$ZIPFILE

for region in ${regions_to_publish[@]}
do
    echo "About to start publishing to $region"

    ../deregister-all.sh $region
    ../publish-resource.sh $region

    if [ "$?" -eq 0 ]
    then
        echo "Publishing to $region succeeded"
        successes+=($region)
    else
        echo "Publishing to $region failed"
        failures+=($region)
    fi

done

echo "The following regions succeeded:"
for s in "${successes[@]}"
do
    echo $s
done

echo "The following regions failed:"
for f in "${failures[@]}"
do
    echo $f
done