Cfrgtkky

I was looking to setup automatic EBS snapshot at a particular time interval (let say once every week), for this I did Google and found that this task can be done using shell script and I found the same at this link.

    #!/bin/bash



# Volume list file will have volume-id:Volume-name format



VOLUMES_LIST = /var/log/volumes-list

SNAPSHOT_INFO = /var/log/snapshot_info

DATE = `date +%Y-%m-%d`

REGION = "ap-south-1a"



# Snapshots Retention Period for each volume snapshot

RETENTION=6



SNAP_CREATION = /var/log/snap_creation

SNAP_DELETION = /var/log/snap_deletion



EMAIL_LIST = shishupal.shakya@itsmysun.com



echo "List of Snapshots Creation Status" > $SNAP_CREATION

echo "List of Snapshots Deletion Status" > $SNAP_DELETION



# Check whether the volumes list file is available or not?



if [ -f $VOLUMES_LIST ]; then



# Creating Snapshot for each volume using for loop



for VOL_INFO in `cat $VOLUMES_LIST`

do

# Getting the Volume ID and Volume Name into the Separate Variables.



VOL_ID = `echo $VOL_INFO | awk -F":" '{print $1}'`

VOL_NAME = `echo $VOL_INFO | awk -F":" '{print $2}'`



# Creating the Snapshot of the Volumes with Proper Description.



DESCRIPTION = "${VOL_NAME}_${DATE}"



/usr/local/bin/aws ec2 create-snapshot --volume-id $VOL_ID --description "$DESCRIPTION" --region $REGION &>> $SNAP_CREATION

done

else

echo "Volumes list file is not available : $VOLUMES_LIST Exiting." | mail -s "Snapshots Creation Status" $EMAIL_LIST

exit 1

fi



echo >> $SNAP_CREATION

echo >> $SNAP_CREATION



# Deleting the Snapshots which are 10 days old.



for VOL_INFO in `cat $VOLUMES_LIST`

do



# Getting the Volume ID and Volume Name into the Separate Variables.



VOL_ID = `echo $VOL_INFO | awk -F":" '{print $1}'`

VOL_NAME = `echo $VOL_INFO | awk -F":" '{print $2}'`



# Getting the Snapshot details of each volume.



/usr/local/bin/aws ec2 describe-snapshots --query Snapshots[*].[SnapshotId,VolumeId,Description,StartTime] --output text --filters "Name=status,Values=completed" "Name=volume-id,Values=$VOL_ID" | grep -v "CreateImage" > $SNAPSHOT_INFO



# Snapshots Retention Period Checking and if it crosses delete them.



while read SNAP_INFO

do

SNAP_ID=`echo $SNAP_INFO | awk '{print $1}'`

echo $SNAP_ID

SNAP_DATE=`echo $SNAP_INFO | awk '{print $4}' | awk -F"T" '{print $1}'`

echo $SNAP_DATE



# Getting the no.of days difference between a snapshot and present day.



RETENTION_DIFF = `echo $(($(($(date -d "$DATE" "+%s") - $(date -d "$SNAP_DATE" "+%s"))) / 86400))`

echo $RETENTION_DIFF



# Deleting the Snapshots which are older than the Retention Period



if [ $RETENTION -lt $RETENTION_DIFF ];

then

/usr/local/bin/aws ec2 delete-snapshot --snapshot-id $SNAP_ID --region $REGION --output text> /tmp/snap_del

echo DELETING $SNAP_INFO >> $SNAP_DELETION

fi

done < $SNAPSHOT_INFO

done



echo >> $SNAP_DELETION



# Merging the Snap Creation and Deletion Data



cat $SNAP_CREATION $SNAP_DELETION > /var/log/mail_report



# Sending the mail Update



cat /var/log/mail_report | mail -s "Volume Snapshots Status" $EMAIL_LIST

But when I ran it over terminal, it is showing me following errors.

Since I am new in this type of work so I am little uncomfortable in resolving this. Please suggest the fix, I am on this since last few days.

I followed below steps to resolve it.

1. Created /var/log/volumes-list and wrote volume-id:Volume-name inside it.


2. Installed AWS CLI using this link.


3. Changed /usr/local/bin/aws with /home/centos/.local/bin/aws (in my case).


4. Ran it using following command.

bash myscript.sh

It worked like a charm.