layout-3 layout-6 layout-1 layout-top
NVIDIA Docs Hubchevron_right NVIDIA Networkingchevron_right Networking Softwarechevron_right Management Softwarechevron_right

NVIDIA NetQ

layout-top layout-6 layout-1

If you are using the current version of Cumulus NetQ, the content on this page may not be up to date. The current version of the documentation is available here. If you are redirected to the main page of the user guide, then this page may have been renamed; please search for it there.

Back Up Your NetQ Data

NetQ data is stored in a Cassandra database. A backup is performed by running scripts provided with the software and located in the /usr/sbin directory. When a backup is performed, a single tar file is created. The file is stored on a local drive that you specify and is named netq_master_snapshot_<timestamp>.tar.gz. Currently, only one backup file is supported, and includes the entire set of data tables. It is replaced each time a new backup is created.

If the rollback option is selected during the lifecycle management upgrade process (the default behavior), a backup is created automatically.

To manually create a backup:

  1. If you are backing up data from NetQ 2.4.0 or earlier, or you upgraded from NetQ 2.4.0 to 2.4.1, obtain an updated backuprestore script. If you installed NetQ 2.4.1 as a fresh install, you can skip this step. Replace <version> in these commands with 2.4.1 or later release version.

    cumulus@switch:~$ tar -xvzf  /mnt/installables/NetQ-<version>.tgz  -C /tmp/ ./netq-deploy-<version>.tgz

cumulus@switch:~$ tar -xvzf /tmp/netq-deploy-<version>.tgz   -C /usr/sbin/ --strip-components 1 --wildcards backuprestore/*.sh

  2. Run the backup script to create a backup file in /opt/<backup-directory> being sure to replace the backup-directory option with the name of the directory you want to use for the backup file.

    cumulus@switch:~$ ./backuprestore.sh --backup --localdir /opt/<backup-directory>

    You can abbreviate the backup and localdir options of this command to -b and -l to reduce typing. If the backup directory identified does not already exist, the script creates the directory during the backup process.

    This is a sample of what you see as the script is running:

    [Fri 26 Jul 2019 02:35:35 PM UTC] - Received Inputs for backup ...
[Fri 26 Jul 2019 02:35:36 PM UTC] - Able to find cassandra pod: cassandra-0
[Fri 26 Jul 2019 02:35:36 PM UTC] - Continuing with the procedure ...
[Fri 26 Jul 2019 02:35:36 PM UTC] - Removing the stale backup directory from cassandra pod...
[Fri 26 Jul 2019 02:35:36 PM UTC] - Able to successfully cleanup up /opt/backuprestore from cassandra pod ...
[Fri 26 Jul 2019 02:35:36 PM UTC] - Copying the backup script to cassandra pod ....
/opt/backuprestore/createbackup.sh: line 1: cript: command not found
[Fri 26 Jul 2019 02:35:48 PM UTC] - Able to exeute /opt/backuprestore/createbackup.sh script on cassandra pod
[Fri 26 Jul 2019 02:35:48 PM UTC] - Creating local directory:/tmp/backuprestore/ ...  
Directory /tmp/backuprestore/ already exists..cleaning up
[Fri 26 Jul 2019 02:35:48 PM UTC] - Able to copy backup from cassandra pod  to local directory:/tmp/backuprestore/ ...
[Fri 26 Jul 2019 02:35:48 PM UTC] - Validate the presence of backup file in directory:/tmp/backuprestore/
[Fri 26 Jul 2019 02:35:48 PM UTC] - Able to find backup file:netq_master_snapshot_2019-07-26_14_35_37_UTC.tar.gz
[Fri 26 Jul 2019 02:35:48 PM UTC] - Backup finished successfully!

  3. Verify the backup file has been created.

    cumulus@switch:~$ cd /opt/<backup-directory>
cumulus@switch:~/opt/<backup-directory># ls
netq_master_snapshot_2019-06-04_07_24_50_UTC.tar.gz

To create a scheduled backup, add ./backuprestore.sh --backup --localdir /opt/<backup-directory> to an existing cron job, or create a new one.

    Copyright © 2024  NVIDIA Corporation