EIP failover

eip_failover.sh

#!/bin/sh
# This script will monitor another HA node and take over an Elastic IP (EIP)
# if communication with the other node fails
# Based off the failover scenario and script available here:
# https://aws.amazon.com/articles/2127188135977316

# High Availability IP variables
# Other node's IP to ping and EIP to swap if other node goes down
HA_Node_IP=
EIP=

#Specify the Secondary Private IP for this node
PRIV_IP=

# Specify the EC2 region that this will be running in
REGION=

# Determine the instance, Allocation, and ENI IDs so we can reassign the EIP to the
# correct ENI.  Requires EC2 describe-instances, associate-address, and describe-addresses
# permissions.  The following example EC2 Roles policy will authorize these
# commands:
# {
#  "Statement": [
#    {
#      "Action": [
#        "ec2:DescribeAddresses",
#        "ec2:AssociateAddress",
#        "ec2:DescribeInstances"
#      ],
#      "Effect": "Allow",
#      "Resource": "*"
#    }
#  ]
# }

Instance_ID=`/usr/bin/curl --silent http://169.254.169.254/latest/meta-data/instance-id`
ENI_ID=`/bin/aws ec2 describe-instances --instance-ids $Instance_ID --region eu-central-1|python -c 'import sys, json; print json.load(sys.stdin)["Reservations"][0]["Instances"][0]["NetworkInterfaces"][0]["NetworkInterfaceId"]'`
ALLOC_ID=`/bin/aws ec2 describe-addresses --public-ips $EIP --region eu-central-1| python -c 'import sys, json; print json.load(sys.stdin)["Addresses"][0]["AllocationId"]'`

echo `date` "-- Starting HA monitor"
while [ . ]; do
  pingresult=`ping -c 3 -W 1 $HA_Node_IP | grep time= | wc -l`
  if [ "$pingresult" == "0" ]; then
    echo `date` "-- HA heartbeat failed, taking over EIP"
    /bin/aws ec2 associate-address --network-interface-id $ENI_ID --allocation-id $ALLOC_ID --private-ip-address $PRIV_IP --allow-reassociation --region $REGION
    pingresult=`ping -c 1 -W 1 $EIP | grep time= | wc -l`
    if [ "$pingresult" == "0" ]; then
      echo `date` "-- Restarting network"
      /sbin/service network restart > /dev/null 2>&1
    fi
    sleep 60
    echo `date` "-- Waiting for $HA_Node_IP to come online before resuming heartbeat"
    until ping -c 3 -W 1 $HA_Node_IP > /dev/null 2>&1; do
      sleep 60
    done
    echo `date` "-- $HA_Node_IP now online, resuming heartbeat"
  fi
  sleep 2
done