Last active
August 29, 2024 20:27
-
-
Save lukeplausin/b64c10f8b524bb310e0083756c42caf6 to your computer and use it in GitHub Desktop.
AWS, JQ and bash command cheat sheet. How to query, cut and munge things in JSON generally.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# Count total EBS based storage in AWS | |
aws ec2 describe-volumes | jq "[.Volumes[].Size] | add" | |
# Count total EBS storage with a tag filter | |
aws ec2 describe-volumes --filters "Name=tag:Name,Values=CloudEndure Volume qjenc" | jq "[.Volumes[].Size] | add" | |
# Describe instances concisely | |
aws ec2 describe-instances | jq '[.Reservations | .[] | .Instances | .[] | {InstanceId: .InstanceId, State: .State, SubnetId: .SubnetId, VpcId: .VpcId, Name: (.Tags[]|select(.Key=="Name")|.Value)}]' | |
# Wait until $instance_id is running and then immediately stop it again | |
aws ec2 wait instance-running --instance-id $instance_id && aws ec2 stop-instances --instance-id $instance_id | |
# Get 10th instance in the account | |
aws ec2 describe-instances | jq '[.Reservations | .[] | .Instances | .[]] | .[10]' | |
# List the private IP addresses of all instances | |
aws ec2 describe-instances | jq '[.Reservations | .[] | .Instances | .[] | .PrivateIpAddress] | sort' | |
# Do that, but only on non-terminated instances | |
aws ec2 describe-instances | jq '[.Reservations | .[] | .Instances | .[] | select(.State.Name!="terminated") | {Name: (.Tags[]|select(.Key=="Name")|.Value), PrivateIp: .PrivateIpAddress}]' | |
# JQ export to csv command / suffix | |
export_csv_suffix='| map([.Name, .PrivateIp] | join(",")) | join("\n")' | |
# Get all production instances | |
instance_ids=$(aws ec2 describe-instances | jq '[.Reservations | .[] | .Instances | .[] | select(.State.Name!="terminated") | select((.Tags[]|select(.Key=="Environment")|.Value) =="prod") | {Name: (.Tags[]|select(.Key=="Name")|.Value), InstanceId: .InstanceId}]' | jq ".[] | .InstanceId") | |
# Add a backup tag to those instances | |
echo $instance_ids | sed "s/\"//g" | grep i- | parallel --delay 3 aws ec2 create-tags --resources {} --tags Key=Backup,Value=PolicyA | |
# Attach multiple new ebs volumes to an instance | |
instance_id="i-0d42888191f597bb8" | |
volume_size="8" | |
for x in {a..h} | |
do | |
volume_id=$(aws ec2 create-volume --size $volume_size --volume-type gp2 --availability-zone eu-west-1a | jq -r ".VolumeId") | |
aws ec2 wait volume-available --volume-ids $volume_id | |
aws ec2 attach-volume --volume-id $volume_id --instance-id $instance_id --device /dev/xvd$x | |
done | |
# Produce a summary of instances | |
jq '[.Reservations | .[] | .Instances | .[] | select(.State.Name!="terminated") | {Name: (.Tags[]|select(.Key=="Name")|.Value), InstanceId: .InstanceId}]' | |
# Check instances for ones which are missing required tags | |
instances=$(cat "./scripts/prod-instances.json") | |
required_tags='["Environment","Backup","Owner","AppName","Name"]' | |
echo $instances | jq "[.Reservations | .[] | .Instances | .[] | select(.Tags | [.[] | .Key] | contains($required_tags) | not)]" | jq ' | |
[.[] | select(.State.Name!="terminated") | select(([.Tags | .[] | .Key]) | contains(["CloudEndure creation time"]) | not) | { | |
InstanceId: .InstanceId, | |
InstanceName: (.Tags | from_entries | .Name), | |
MissingTags: (('$required_tags') - ([.Tags | .[] | .Key])) | |
}]' | |
# Get the 'Live & Tagged' instances | |
instances=$(aws ec2 describe-instances) | |
live=$(echo $instances | jq "[.Reservations | .[] | .Instances | .[] | select(.Tags | [.[] | .Key] | contains($required_tags))]") | |
# Enable termination protection from a list of instances stored in $list | |
echo $live | jq -r ".[] | .InstanceId" | while read id | |
do | |
echo "Enabling termination proection on machine: $id" | |
aws ec2 modify-instance-attribute --disable-api-termination --instance-id $id | |
done | |
# Attach unused EBS Volumes to an instance | |
instance_id="i-abcd1234" | |
letters=({a..j}) | |
volumes=$(aws ec2 describe-volumes | jq -r ".Volumes[] | select(.State==\"available\") | .VolumeId") | |
lc=1 | |
echo $volumes | while read id | |
do | |
echo "Attaching volume on: $id" | |
aws ec2 attach-volume --instance-id $instance_id --volume-id $id --device /dev/sd${letters[++lc]} | |
done | |
# Detach and delete secondary volumes on a machine | |
aws ec2 describe-instances --instance-ids $instance_id | jq -r ".Reservations[0].Instances[0].BlockDeviceMappings | .[] | select(.DeviceName != \"/dev/sda1\") | .Ebs.VolumeId" | while read volume_id | |
do | |
aws ec2 detach-volume --volume-id $volume_id && \ | |
aws ec2 wait volume-available --volume-ids $volume_id && \ | |
aws ec2 delete-volume --volume-id $volume_id | |
done | |
# Copy everything from an account into an OSX clipboard | |
aws ec2 describe-volumes | jq "[.Volumes[] | select(.State==\"available\") | .VolumeId]" | pbcopy | |
# Tell me ALL my instances in ALL regions across ALL accounts (from CLI file) | |
echo -e 'Profile \t Region \t InstanceId \t Name Tag' | |
for profile in $(grep "^\[.*\]" ~/.aws/config | sed 's/\[//g' | sed 's/\]//g' | cut -d ' ' -f 2) ; do | |
for region in `aws --profile $profile --region us-east-1 ec2 describe-regions | jq -r '.Regions | .[] | .RegionName'`; do | |
instances=$(aws --profile $profile --region $region ec2 describe-instances) | |
filtered=$(echo $instances | jq "[.Reservations | .[] | .Instances | .[] | select(.State.Name!=\"terminated\")]") | |
summary=$(echo $filtered | jq "[ .[] | {Name: (.Tags // {} | from_entries | .Name ), InstanceId: .InstanceId, Profile: \"$profile\", Region: \"$region\"} ]") | |
# JSON format: echo $summary | |
# Tabular format: | |
echo "$summary" | jq -r '.[] | [.Profile, .Region, .InstanceId, .Name] | @tsv' | |
done | |
done | |
# Iterating all profiles / regions is useful as a tool for account scanning. Let's define a useful alias for doing this: | |
function awsloop() { | |
for profile in $(grep "^\[.*\]" ~/.aws/config | sed 's/\[//g' | sed 's/\]//g' | cut -d ' ' -f 2) ; do | |
for region in `aws --profile $profile --region us-east-1 ec2 describe-regions | jq -r '.Regions | .[] | .RegionName'`; do | |
echo "--------------------------------------------" | |
echo "| profile: $profile, region: $region" | |
echo "--------------------------------------------" | |
AWS_PROFILE=$profile AWS_REGION=$region $SHELL -c "$@" | |
done | |
done | |
} | |
function awsgloop() { | |
for profile in $(grep "^\[.*\]" ~/.aws/config | sed 's/\[//g' | sed 's/\]//g' | cut -d ' ' -f 2) ; do | |
echo "--------------------------------------------" | |
echo "| profile: $profile, region: us-east-1" | |
echo "--------------------------------------------" | |
AWS_PROFILE=$profile AWS_REGION=us-east-1 $SHELL -c "$@" | |
done | |
} | |
# Use our all accounts / all regions shorthand to list out all VPC CIDR ranges in use in all regions in all accounts | |
awsloop 'aws ec2 describe-vpcs | jq -r ".Vpcs | .[] | {\"ID\": .VpcId, \"CIDR\": .CidrBlock}"' | |
# Let's use the awsgloop function to locate which account a particular S3 bucket lives in | |
awsgloop 'aws s3 list-buckets | grep my-s3-bucket-name' | |
# Iterate all AWS profiles and regions, reporting on EKS clusters running there | |
# (Using ~/.aws/config instead of ~/.aws/credentials file) | |
for profile in $(grep "^\[.*\]" ~/.aws/config | sed 's/\[//g' | sed 's/\]//g' | cut -d ' ' -f 2) ; do | |
for region in `aws --profile $profile --region us-east-1 ec2 describe-regions | jq -r '.Regions | .[] | .RegionName'`; do | |
clusters=$(aws --profile $profile --region $region eks list-clusters) | |
clusters=$(echo $clusters | jq -r '.clusters | .[]') | |
if [ ! -z "$clusters" ]; then | |
echo ">> profile: $profile | region: $region" | |
echo $clusters | |
fi | |
done | |
done | |
Thanks for this awesome work!
I've found a case where "Tags" array is completely missing, so your suggestion to afsheenfatima is still not working.
I've cloned your gist and updated that command to set InstanceName as null if the Tags array is missing. You can find my forked gist here
Maybe you can merge this command with the given suggestion so it'll cover all the cases.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Brilliant thanks for sharing! <3