Backup and restore of cluster setup
You can back up the current state of a NetScaler cluster node. Later, you can use the backed-up files to restore the node to the same cluster state. As a precautionary measure, you must use this feature before performing an upgrade on the cluster nodes.
Back up a cluster Setup
You can take a basic or full backup depending on the following:
- Type of data to be backed up.
-
Frequency at which you create a backup.
- Basic back up. Backs up only configuration files. You might want to perform this type of backup frequently, because the files it backs up change constantly. The files that are backed up are listed in the table.
Directory
Sub-Directory or Files
/nsconfig/
- ns.conf
- ZebOS.conf
- rc.netscaler
- snmpd.conf
- nsbefore.sh
- nsafter.sh
- inetd.conf
- ntp.conf
- syslog.conf
- newsyslog.conf
- crontab
- host.conf
- hosts
- ttys
- sshd_config
- httpd.conf
- monitrc
- rc.conf
- ssh_config
- local time
- issue
- issue.net
/var/
- download/*
- log/wicmd.log
- wi/tomcat/webapps/*
- wi/tomcat/logs/*
- wi/tomcat/conf/catalina/localhost/*
- nslw.bin/etc/krb.conf
- nslw.bin/etc/krb.keytab
- netscaler/locdb/*
- lib/likewise/db/*
- vpn/bookmark/*
- netscaler/crl
- nstemplates/*
- learnt_data/*
/netscaler/
- custom.html
-
vsr.html
- Full back up. Apart from the files that are backed up by a basic backup, a full backup backs up some less frequently updated files. The files that are backed up when using the full backup option are listed in the table.
Directory
Sub-Directory or Files
/nsconfig/
- ssl/*
- license/*
- fips/*
/var/
- netscaler/ssl/*
- wi/java_home/jre/lib/security/cacerts/*
- wi/java_home/lib/security/cacerts/*
Important
The backup and restore do not work if CLAG is configured on an SDX cluster setup.
The backup is stored as a compressed TAR file in the /var/ns_sys_backup/ directory. To avoid issues due to non-availability of disk space, you can store a maximum of 50 backup files in this directory. You can use the rm system backup command to delete existing backup files so that you can create more backups.
When you perform the back-up operation on a CLIP of a cluster setup, back up files are created on each of the cluster nodes.
How to back up a cluster setup
To back up the cluster setup on CLIP by using the NetScaler CLI.
At the command prompt, do the following:
-
Save the configuration.
save ns config<!--NeedCopy--> -
Create the backup file (basic or full).
```create system backup [
][–level (basic | full)][-comment ] **Example** ```create system backup cluster-backup-1 –level basic<!--NeedCopy-->
The preceding command creates a back-up TAR file on each of the cluster node with the specified file name. For example, the CLUSTER-backup-1.tgz file is created on each of the cluster node.
Note
If the file name is not specified, back up TAR files are created on each of the cluster nodes with the following naming convention:
backup_<level>_<nsip_address of the cluster node 0>_<date-timestamp>.tgz<!--NeedCopy-->backup_<level>_<nsip_address of the cluster node 1>_<date-timestamp>.tgz<!--NeedCopy-->For example, on a three node cluster setup,
backup_<level>_<nsip_address of the cluster node 0>_<date-timestamp>.tgz<!--NeedCopy-->is created on node0backup _<level>_<nsip_address of the cluster node 1>_<date-timestamp>.tgz<!--NeedCopy-->is created on node1backup_<level>_<nsip_address of the cluster node 2>_<date-timestamp>.tgz<!--NeedCopy-->is created on node2
-
Verify the created backup files on CLIP.
show system backup<!--NeedCopy-->
Restore a cluster setup
When a cluster node becomes faulty, you can replace this node with a new node. You can set the new node for a cluster, using a backup file of the faulty node.
For example, in a three-node cluster setup, if node1 becomes faulty, you can replace this faulty node with a new node as node1. Using the restore operation, you can restore one of the faulty node’s backup files on the new node.
Note
The restore operation does not succeed if the backup file is renamed or if the contents of the file are modified.
How to restore a cluster node
To restore a cluster node by using the CLI
At the command prompt, do the following:
-
Obtain a list of the backup files available on CLIP.
show system backup<!--NeedCopy--> -
Copy the backup tar file to the /var/ns_sys_backup directory of the cluster node, which is to be restored.
-
Add the backup tar file to the cluster node memory by running the following command on the cluster node.
```add system backup
**Example** ```add system backup CLUSTER-BACKUP-1.tgz<!--NeedCopy-->Note
The command must be run on the cluster node to be restored.
-
Restore the cluster node by specifying the backup file.
```restore system backup
**Example** ```restore system backup CLUSTER-BACKUP-1.tgz<!--NeedCopy-->Note
The command must be run on the cluster node to be restored.
-
Reboot the cluster node.
reboot
Note
The command must be run on the cluster node to be restored.