AWS instance state collection & merge script. requires python3 & awscli

collect.sh

#!/bin/bash
DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
REG=$(aws configure get region)
OPDIR=$DIR/$REG/

function retrieve() {
    mkdir -p $OPDIR && \
	cd $OPDIR && \
	(aws ec2 describe-volumes | tee $OPDIR/ec2-describe-volumes.json | wc) && \
	(aws ec2 describe-instances | tee $OPDIR/ec2-describe-instances.json | wc) && \
	(aws ec2 describe-addresses | tee $OPDIR/ec2-describe-addresses.json | wc) && \
	(aws elb describe-load-balancers | tee $OPDIR/elb-describe-load-balancers.json | wc) && \
    echo 'all done retrieving!'     
}
function cat_instances() {
    cat $OPDIR/ec2-describe-instances.json  | jq -c '.Reservations[].Instances|.[]|.'
}
function instance_names() {
    jq -c '.Tags|map(select(.Key | contains ("Name")))[].Value'
}

function instance_table_header() {
    echo '["InstanceId","InstanceName","InstanceType","SecurityGroup","NetGroup"]'
    }
function instance_table_body() {
    jq -r '[.InstanceId,
(.Tags|map(select(.Key | contains ("Name")))[].Value)
,.InstanceType
,([.SecurityGroups[].GroupName]|join(";"))
,([.NetworkInterfaces[].Groups[].GroupName]|join(";"))
]'
}


function load_balancers() {
    cat $OPDIR/elb-describe-load-balancers.json \
	| jq '.LoadBalancerDescriptions[]' \
	| jq -c '.Instances[].InstanceId as $i | .LoadBalancerName as $lb | [$i,$lb]'
    }
function instance_table() {
    (instance_table_header && \
	 cat_instances \
	     | instance_table_body) \
	| jq -r '@tsv'
    }

function balancers_instances_table_header() {
    echo '["InstanceId","InstanceName","InstanceType","SecurityGroup","NetGroup","Balancers"]'
}

function balancers_instances_table() {
    (
	balancers_instances_table_header && \
	    balancers_instances \
		| python3 $DIR/join.py) \
	    | jq -r '@tsv'
    }
function balancers_instances() {
    echo "{"'"'"balancers"'"'":$(load_balancers | jq -r -s '@json'),"'"'"instances"'"'":$(cat_instances | instance_table_body | jq -s -r '@json')}"    
    }
# function instances() {
#     jq -c '.Tags' | jq -c '.Groups' #,map(select(.Key | contains ("Name")))[].Value'
# }

function analyze() {
    cd $OPDIR && \
	echo '*** BALANCERS' && \
	(cat elb-describe-load-balancers.json \
	     | jq '.LoadBalancerDescriptions[]|[.LoadBalancerName,.Instances[].InstanceId]' \
	     | tee balancers-digest.txt) && \
	echo '*** IMAGE TYPES' && \
	( cat ec2-describe-instances.json \
	      | jq '.Reservations[].Instances|.[]|.' \
	      | grep -i imageid \
	      | sort \
	      | uniq -c \
	      | sort -n \
	      | tee imagetypes.txt) && \
	echo '*** NETWORK GROUPS' && \
	( cat ec2-describe-instances.json \
	      | jq -c '.Reservations[].Instances|.[]|.NetworkInterfaces[].Groups' \
	      | sort \
	      | uniq -c \
	      | sort -n \
	      | tee networkgroups.txt) && \
	echo '*** SECURITY GROUPS' && \
	( cat ec2-describe-instances.json \
	      | jq -c '.Reservations[].Instances|.[]|.SecurityGroups' \
	      | sort \
	      | uniq -c \
	      | sort -n \
	      | tee securitygroups.txt) && \
	echo '*** overview' && \
	( balancers_instances_table > $OPDIR/$REG.tsv ) && \
	echo 'all done analyzing.' && \
	cd -
    }
function run() {
    retrieve && analyze
}

if [[ "$0" != "-bash" ]] && [[ "$0" != "bash" ]] ; then
    run
else
    echo "not running - script was sourced."
fi

join.py

#!/usr/bin/env python
import json,sys,collections
i = json.loads(sys.stdin.read())
balancers = collections.defaultdict(list)
for iid,bname in i['balancers']:
    balancers[iid].append(bname)
for row in i['instances']:
    row.append(','.join(balancers[row[0]]))
    print(json.dumps(row))
#print(json.dumps(i['instances']))

requirements.txt

awscli