/cassandra-cloud-backup

Cassandra backups to Google Cloud Storage

Primary LanguageShellApache License 2.0Apache-2.0

Cassandra Backup and Restore with Google Cloud Storage

Shell script for creating and managing Cassandra Backups using Google Cloud Storage.

Features

  • Take snapshot backups
  • Copy Incremental backup files
  • Compress with gzip or bzip2 to save space
  • Prune old incremental and snapshot files
  • Execute Dry Run mode to identify target files

Requirements

Google Cloud SDK installed with gsutil utility configured for authentication to An existing Google Cloud Storage bucket Linux system with BASH shell. Cassandra 2+

Usage

./cassandra-cloud-backup.sh [ options ] < command>

Examples

  • Take a full snapshot, gzip compress it with nice level 15, use the /var/lib/cassandra/backups directory to stage the backup before uploading it to the GCS Bucket, and clear old incremental and snapshot files

./cassandra-cloud-backup.sh -b gs://cassandra-backups123/ -zCc -N 15 -d /var/lib/cassandra/backups backup

  • Do a dry run of a full snapshot with verbose output and create list of files that would have been copied

./cassandra-cloud-backup.sh -b gs://cassandra-backups123/ -vn backup

  • Backup and bzip2 compress copies of the most recent incremental backup files since the last incremental backup

./cassandra-cloud-backup.sh -b gs://cassandra-backups123/ -ji -d /var/lib/cassandra/backups backup

  • Restore a backup without prompting from given bucket path and keep the old files locally

./cassandra-cloud-backup.sh -b gs://cass-bk123/backups/host01/snpsht/2016-01-20_18-57/ -fk -d /var/lib/cassandra/backups restore

  • List inventory of available backups stored in Google Cloud Store

./cassandra-cloud-backup.sh -b gs://cass-bk123 inventory

  • List inventory of available backups stored in Google Cloud Store for a different server

./cassandra-cloud-backup.sh -b gs://cass-bk123 inventory -a testserver01

Commands:

  • backup --- Backup the Cassandra Node based on passed in options
  • restore --- Restore the Cassandra Node from a specific snapshot backup
  • inventory --- List available backups
  • commands --- List available commands
  • options --- list available options

Options:

Flags:

-a, --alt-hostname Specify an alternate server name to be used in the bucket path construction. Used to create or retrieve backups from different servers

-B, backup Default action is to take a backup

-b, --gcsbucket Google Cloud Storage bucket used in deployment and by the cluster.

-c, --clear-old-ss Clear any old SnapShots taken prior to this backup run to save space additionally will clear any old incremental backup files taken immediately following a successful snapshot. this option does nothing with the -i flag

-C, --clear-old-inc Clear any old incremental backups taken prior to the the current snapshot

-d, --backupdir The directory in which to store the backup files, be sure that this directory has enough space and the appropriate permissions

-D, --download-only During a restore this will only download the target files from GCS

-f, --force Used to force the restore without confirmation prompt

-h, --help Print this help message.

-H, --home-dir This is the $CASSANDRA_HOME directory and is only used if the data_directories, commitlog_directory, or the saved_caches_directory values cannot be parsed out of the yaml file.

-i, --incremental Copy the incremental backup files and do not take a snapshot. Can only be run when compression is enabled with -z or -j

-j, --bzip Compresses the backup files with bzip2 prior to pushing to Google Cloud Storage This option will use additional local disk space set the --target-gz-dir to use an alternate disk location if free space is an issue

-k, --keep-old
Set this flag on restore to keep a local copy of the old data files Set this flag on backup to keep a local copy of the compressed backup and schema dump

-l, --log-dir Activate logging to file 'CassandraBackup${DATE}.log' from stdout Include an optional directory path to write the file Default path is /var/log/cassandra

-n, --noop Will attempt a dry run and verify all the settings are correct

-N, --nice Set the process priority, default 10

-p The Cassandra User Password if required for security

-r, restore Restore a backup, requires a --gcsbucket path and optional --backupdir

-s, --split-size Split the resulting tar archive into the configured size in Megabytes, default 100M

-S, --service-name Specify the service name for cassandra, default is cassandra use to stop and start service

-T, --target-gz-dir Override the directory to save compressed files in case compression is used default is --backupdir/compressed, also used to decompress for restore

-u The Cassandra User account if required for security

-U, --auth-file A file that contains authentication credentials for cqlsh and nodetool consisting of two lines: CASSANDRA_USER=username CASSANDRA_PASS=password

-v --verbose When provided will print additional information to log file

-y, --yaml Path to the Cassandra yaml configuration file default: /etc/cassandra/cassandra.yaml

-z, --zip Compresses the backup files with gzip prior to pushing to Google Cloud Storage This option will use additional local disk space set the --target-gz-dir to use an alternate disk location if free space is an issue

###Cron Examples

  • Full gzip compressed snapshot every day at 1:30 am with nice level 10

30 1 * * * /path_to_scripts/cassandra-cloud-backup.sh -z -N10 -b gs://cass-bk123-vCcj -d /var/lib/cassandra/backups > /var/log/cassandra/$(date +\%Y\%m\%d\%H\%M\%S)-fbackup.log 2>&1

  • Incremental gzip compressed backups copied every hour nice level 10

0 * * * * /path_to_scripts/cassandra-cloud-backup.sh -b -N10 gs://cass-bk123 -vjiz -d /var/lib/cassandra/backups > /var/log/cassandra/$(date +\%Y\%m\%d\%H\%M\%S)-ibackup.log 2>&1

Notes

The script must be run with sufficient privileges to be able to stop/start processes and create/delete directories and files within the working directories.

The restore command is designed to perform a simple restore of a full snapshot. In the event that you want to restore incremental backups you should start by restoring the last full snapshot prior to your target incremental backup file and manually move the files from each incremental backup in chronological order leading up to the target incremental backup file. The schema dump is included in the snapshot backups, but if necessary it must also be restored manually.

Snapshots are taken at the system level, the script currently does not support backup or restore of an individual keyspace or columnfamily.

In order to enable incremental backups, the incremental_backups option has to be set to true in the cassandra.yaml file.

###License Copyright 2016 Google Inc. All Rights Reserved.

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS-IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

This is not an official Google product.