Skip to content

Instantly share code, notes, and snippets.

@nicosnyman

nicosnyman/change_instance.sh

Created May 19, 2015 08:59
Download ZIP

Revisions

  1. nicosnyman created this gist May 19, 2015.
    207 changes: 207 additions & 0 deletions change_instance.sh
    View file
    Original file line number Diff line number Diff line change
    @@ -0,0 +1,207 @@
    #!/bin/bash
    # Nico Snyman, nico@gammafly.com, 14/05/2015
    # Change an instance type, while keeping snapshots
    # This script will stop an instance identified by instance ID,
    # take snapshots of all atached volumes, and restart it
    # Use:
    # -i instance ID - REQUIRED - stored in instance_id
    # -t instance new instance type - REQUIRED - stored in instance_type
    # -s if set to 0, no snapshot will be taken, dafault is 1
    # (take snapshot of attached volumes) - stored in snapshot
    # -h Print Help
    #
    # Example: change_instance.sh -i i-aeae453e -t t2.small -s 0
    # The script requires AWS CLI tools to be installed and configured
    # If you want to take snapshots BEFORE stopping the instance, edit
    # the "Main" part of the script, moving stop_instance to below create_snapshot
    #
    # NOTE: I do no error checking to see if the instance exists, the script
    # will crash horribly if it doesn't


    #Some variables
    instance_id="0"
    instance_type="0"
    snapshot=1

    PROGNAME=$(basename $0) #Sets PROGNAME to the name of the script
    logfile="~/ec2-instance-change.log" #Set up the log file name
    tempfile="/tmp/volume_info.txt" #Sets up the temp file for volume lists

    clean_up() {

    # Perform program exit housekeeping
    rm $tempfile
    echo "$PROGNAME aborted by user"
    exit
    }

    #usage - print script help
    usage() {

    # Display usage message on standard error
    echo "Usage: $PROGNAME [OPTION] [Args]
    Change an instance type, while keeping snapshots
    This script will stop an instance identified by instance ID,
    take snapshots of all atached volumes, and restart it
    Options:
    -i instance ID - REQUIRED - stored in instance_id
    -t instance new instance type - REQUIRED - stored in instance_type
    -s if set to 0, no snapshot will be taken, dafault is 1
    (take snapshot of attached volumes) - stored in snapshot
    Example: change_instance.sh -i i-aeae453e -t t2.small -s 0
    The script requires AWS CLI tools to be installed and configured
    If you want to take snapshots BEFORE stopping the instance, edit the #Main#
    part of the script, moving stop_instance to below create_snapshot
    Allowed instance types:
    t2.micro | t2.small | t2.medium | m3.medium | m3.large | m3.xlarge | m3.2xlarge
    c4.large | c4.xlarge | c4.2xlarge | c4.4xlarge | c4.8xlarge | c3.large | c3.xlarge | c3.2xlarge | c3.4xlarge | c3.8xlarge
    r3.large | r3.xlarge | r3.2xlarge | r3.4xlarge | r3.8xlarge
    i2.xlarge | i2.2xlarge | i2.4xlarge | i2.8xlarge | d2.xlarge | d2.2xlarge | d2.4xlarge | d2.8xlarge
    g2.2xlarge | g2.8xlarge
    " 1>&2
    }

    #volume_info() - Get a list of all volumes attached to instance-id, write to /tmp/volume_info.txt
    volume_info () {
    aws ec2 describe-volumes --filter Name=attachment.instance-id,Values=$instance_id --query Volumes[*].{ID:VolumeId} --output text | tr '\t' '\n' > $tempfile 2>&1
    }

    #create_snapshots(): Take a snapshot of all volumes with correct tags
    create_snapshots(){
    for volume_id in $(cat $tempfile)
    do
    #Create a decription for the snapshot that describes the volume: servername.device-backup-date
    instance_name=$(aws ec2 describe-instances --instance-ids $instance_id --query Reservations[*].Instances[*].Tags --output text|awk '{if (/[/]/) pat1=$0; if ($1 ~/Name/) pat2=$2;}{print pat2}')
    temp="-backup-$(date +%Y-%m-%d)"
    description=$instance_name"-"$volume_id$temp
    description=${description// /.}
    #echo "Volume ID is $volume_id" >> $logfile
    echo "INFO: Creating snapshot $description"
    #Take a snapshot of the current volume, and capture the resulting snapshot ID
    snapresult=$(aws ec2 create-snapshot --output=text --description $description --volume-id $volume_id --query SnapshotId)

    # Add some tags to the snapshot
    tagresult=$(aws ec2 create-tags --resource $snapresult --tags Key=CreatedBy,Value=ChangeInstanceType)
    echo "INFO: Adding tags CreatedBy:$PROGNAME = $tagresult"
    done
    }

    read_parameters() {
    while [ "$1" != "" ]; do
    case $1 in
    -i | --instance ) shift
    instance_id=$1
    ;;
    -s | --snapshot ) shift
    snapshot=$1
    retention_date_in_seconds=`date +%s --date "$days days ago"` #Set oldest date snapshots should be kept
    ;;
    -t | --type ) shift
    instance_type=$1
    ;;
    -h | --help ) shift
    usage
    exit 0
    ;;
    * ) usage
    exit 1
    esac
    shift
    done

    if [ $instance_id == "0" ]; then
    echo "Instance_id is required, use -h for Help." 1>&2
    exit 1
    fi

    # TCheck that instance type is specified
    if [ $instance_type == "0" ]; then
    echo "Instance_type is required, use -h for Help." 1>&2
    exit 1
    fi
    }

    stop_instance () {
    ##Code to stop instance, notify user, wait till instance has stopped before returning
    ##First check if the instance is actually running
    instance_state=$(aws ec2 describe-instances --instance-ids $instance_id --query Reservations[*].Instances[*].State.Name --output text)
    echo "INFO: Instance state is: $instance_state"
    if [ $instance_state = "running" ]; then
    echo "INFO: Stopping instance $instance_id"
    aws ec2 stop-instances --instance-ids $instance_id
    fi
    i=0
    state=$(aws ec2 describe-instances --instance-ids $instance_id --query Reservations[*].Instances[*].State.Name --output text)
    until [ $state = "stopped" ]; do
    sleep 1s
    i=$((i+1))
    state=$(aws ec2 describe-instances --instance-ids $instance_id --query Reservations[*].Instances[*].State.Name --output text)
    printf "."
    if [ "$i" -ge "180" ]; then
    echo "ERROR: Instance took too long to stop, exiting"
    exit 1
    fi
    done
    printf "\n"
    echo "INFO: Instance stopped"
    }

    start_instance () {
    ##Code to start instance, notify user, wait till instance is running before returning
    echo "INFO: Starting instance $instance_id"
    aws ec2 start-instances --instance-ids $instance_id
    i=0
    state=$(aws ec2 describe-instances --instance-ids $instance_id --query Reservations[*].Instances[*].State.Name --output text)
    until [ $state = "running" ]; do
    sleep 1s
    i=$((i+1))
    printf "."
    if [ "$i" -ge "180" ]; then
    echo "ERROR: Instance took too long to start, exiting"
    exit 1
    fi
    state=$(aws ec2 describe-instances --instance-ids $instance_id --query Reservations[*].Instances[*].State.Name --output text)
    done
    new_instance_type=$(aws ec2 describe-instance-attribute --instance-id $instance_id --attribute instanceType)
    printf "\n"
    echo "INSTANCEID $new_instance_type"
    echo
    echo "INFO: Completed."
    }

    change_instance_type () {
    ##Code to change the instance type, notify user, return
    instance_state=$(aws ec2 describe-instances --instance-ids $instance_id --query Reservations[*].Instances[*].State.Name --output text)
    if [ $instance_state == "stopped" ]; then
    echo "INFO: Changing $instance_id to $instance_type"
    aws ec2 modify-instance-attribute --instance-id $instance_id --instance-type "{\"Value\": \"$instance_type\"}"
    else
    echo "ERROR: Instance state is: $instance_state"
    echo "ERROR: Cannot change instance type in this state, exiting" 1>&2
    exit 1
    fi
    }

    ### Main ###
    ############

    read_parameters $@

    if [ "$snapshot" -eq "1" ]; then
    volume_info
    fi

    stop_instance

    if [ "$snapshot" -eq "1" ]; then
    create_snapshots
    fi

    change_instance_type

    start_instance
    #clean_up