Veritas NetBackup™ Flex Scale Administrator's Guide
- Product overview
- Viewing information about the NetBackup Flex Scale cluster environment
- NetBackup Flex Scale infrastructure management
- User management
- Considerations for managing NetBackup Flex Scale users
- Adding users
- Changing user password
- Removing users
- Modifying user roles
- Considerations for configuring AD/LDAP
- Configuring AD server for Universal shares and Instant Access
- Configuring AD/LDAP servers for NetBackup services
- Configuring additional AD/LDAP servers for managing NetBackup services/Universal Shares/Instant Access
- Configuring AD/LDAP servers on clusters deployed with only media servers
- Directory services and certificate management
- Region settings management
- About NetBackup Flex Scale storage
- About Universal Shares
- Cloud bucket support for NetBackup Flex Scale
- Backing up data to Data Domain storage
- Node and disk management
- NetBackup Flex Scale network cabling
- Adding a node to the cluster using the NetBackup Flex Scale web interface
- Adding a node using the REST APIs
- Replacing a node in a cluster
- Starting and stopping nodes
- Rebooting a node
- Adding an excluded node to the cluster
- Replacing a disk
- Adding an excluded disk to the cluster
- Viewing the disk sync status
- Viewing disk details
- Viewing node details
- Switching management console to another cluster node
- License management
- Stopping NetBackup service containers
- Starting NetBackup service containers
- Managing the Fibre Channel ports
- Requirements
- Enabling BOM (Bill of Materials) configuration for Fibre Channel
- Assigning Fibre Channel ports
- Discovering attached devices
- Rescanning Fibre Channel cards
- Cleaning Fibre Channel ports
- Unassigning Fibre Channel ports
- Viewing details about the Fibre Channel ports
- Disabling BOM (Bill of Materials) configuration for Fibre Channel
- Managing hardware vendor packages
- Updating credentials for HPE iLO administrator users
- User management
- NetBackup Flex Scale network management
- About network management
- Modifying DNS settings
- Configuring MTU on public interfaces
- Configuring the console FQDN
- About bonding Ethernet interfaces
- Bonding operations
- Configuring NetBackup Flex Scale in a non-DNS environment
- Data network configurations
- Choosing the correct input method for data network configuration
- Network configuration on plain device (eth5)
- Network configuration on VLAN (eth5)
- Network configuration on bonded interfaces (bond0 on eth5 and eth7)
- VLAN on bond of eth5 and eth7 (bond0)
- Network configuration on management interface (eth1)
- Network configurations for adding a partial data network
- Support for multiple VLAN when disaster recovery is configured
- Configuring static routes on a NetBackup Flex Scale cluster
- NetBackup Flex Scale infrastructure monitoring
- Resiliency in NetBackup Flex Scale
- EMS server configuration
- Site-based disaster recovery in NetBackup Flex Scale
- About site-based disaster recovery in NetBackup Flex Scale
- Configuring disaster recovery using GUI
- Clearing the host cache
- Automated NetBackup SLP management
- DNS key management
- Managing disaster recovery using GUI
- Performing disaster recovery using RESTful APIs
- Active-Active disaster recovery configuration
- NetBackup optimized duplication using Storage Lifecycle Policies
- NetBackup Flex Scale security
- About the security meter
- STIG overview for NetBackup Flex Scale
- FIPS overview for NetBackup Flex Scale
- Managing the login banner
- Changing the password policy
- Support for immutability in NetBackup Flex Scale
- Authenticating users using digital certificates or smart cards
- About system certificates on NetBackup Flex Scale
- Deploying external certificates on NetBackup Flex Scale
- Configuring isolated recovery environment (IRE)
- Configuring multifactor authentication
- About multifactor authentication
- Considerations before configuring multifactor authentication
- Configuring multifactor authentication for your user account
- Disabling multifactor authentication for your user account
- Enforcing multifactor authentication for all users
- Configuring multifactor authentication for your user account when it is enforced in the cluster
- Resetting multifactor authentication for a user
- Single Sign-On (SSO)
- Appendix A. Maintenance procedures for HPE servers
- Replacement procedure for a chassis fan
- Replacement procedure for power supply
- Replacement procedure for a single OS disk
- Replacement procedure for both OS disks on a non- management console node
- Replacement procedure for NVMe disks (SSDs)
- Replacement procedure for RAID controller
- Replacement procedure for an Integrated Lights-Out (iLO) port
- Replacement procedure for quad-port NIC
- Procedure for memory expansion (DIMMs)
- Replacement procedure for memory (DIMMs)
- Replacement procedure for Mellanox port
- Replacement procedure for SFP port
- Replacement procedure for chassis
- Replacement procedure for a hard disk drive
- Replacement procedure for a Fibre Channel card for a cluster node
- Replacement procedure for a Fibre Channel card for a node that is not in a cluster
- Appendix B. Configuring NetBackup optimized duplication
- Appendix C. Disaster recovery terminologies
- Appendix D. Configuring Auto Image Replication
Deploying external certificates on NetBackup Flex Scale
You can generate and use external certificates instead of internal certificates. External Certificate Authority (ECA) certificates are the digital credentials that attest to the certificate owner's identity and affiliation. Once you deploy the external certificates, all the NetBackup Flex Scale components use them. These include the NetBackup primary server, media server, storage engine, management gateway, and the NetBackup Flex Scale web services. One certificate is deployed for all the components. The external certificates also deploy a certificate bundle and (optionally) certificate revocation list. To generate an external certificate, you have to create a certificate request with proper 'Subject Distinguished Name' and 'Subject Alternative Names.' You can generate a certificate request using the GUI. The necessary FQDNs are auto-populated to generate the correct request. You can add additional information as needed. Based on the certificate request, you can create an external certificate. When deploying external certificate for the first time, you have to provide a CA certificate bundle. This is used to validate the incoming and deployed external certificate. You can also optionally provide a certification revocation list. NetBackup components use the CRL.
Some important terminologies:
A certificate authority, also known as a certification authority, is a trusted organization that verifies websites (and other entities) so that you know who you are communicating with online. Their objective is to make the internet a more secure place for both organizations and users. Becoming a Certificate Authority (CA) means that you (or your customers) oversee the issuing process of cryptographic pairs of private keys and public certificates.
Certificate bundle (CA bundle) is a file that contains root and intermediate certificates. The end-entity certificate along with a CA bundle constitutes the certificate chain.
Certificate Revocation List (CRL) is a list of digital certificates that have been revoked by the issuing Certificate Authority (CA) before their scheduled expiration date and should no longer be trusted. CRL is optional. It may be provided as a file or embedded in certificate as a URL.
Subject Alternative Name: This field lets you specify additional host names (such as sites, IP addresses, common names) to be protected by a single SSL certificate. They are added to generate certificates for new nodes or additional VLAN IPs to be added in the future.
Considerations while deploying ECA:
All certificates for communication should be obtained from a common trusted CA. Auto Image Replication (AIR) between MDSPs that uses different external CAs is not supported but you can concatenate the individual root CA certificates into one file and upload them as a CA bundle.
After ECA is deployed on the cluster, you can renew or update the ECA.
It is recommended to pause backup/restore operations before starting ECA deployment/renewal.
The CA bundle and CRL file independent of other security artifacts.
When you deploy security artifacts, they are validated and if inconsistencies are found, you are notified, and deployment does not proceed. If you provide an external certificate and CA certificate bundle, the EC certificate is validated against the user provided CA certificate bundle. If only one of the items is provided, it is validated against deployed artifacts.
Only NetBackup Certificate Authority (NBCA) + ECA deployment is supported in this release.
You cannot revert to NBCA deployment once NBCA + ECA deployment is done.
You do not get any alert for NBCA expiry or renewal. An event is raised when NBCA is about to expire and renewed in the background.
NBCA is auto renewed 60 days before expiration.
If NBCA renewal fails, a failed task can be seen on NetBackup Flex Scale GUI.
You are notified 60 days before the expiration of the ECA certificates. An alert appears on the appliance GUI and an email is also sent.
You can deploy external certificate only if all NetBackup Flex Scale components are up and running. These include NetBackup primary and media services, storage engines, management gateway, and NetBackup Flex Scale management web services.
You cannot deploy security artifacts, if upgrade, add node or VLAN operation is in progress and vice versa.
If the ECA's subject alternative names have information on new nodes (FQDNs) to be added, add node operation succeeds seamlessly and all services come up after the add node operation. If subject alternative names are not updated, add node operation fails.
For Nutanix, HBase workloads using SSL certificates, append the respective SSL certificates to the CA bundle after ECA certificates are renewed. If you do not append the SSL certificates to the CA bundle during ECA renewal, backup and restore operations for the workloads may fail.
If you want to deploy ECA on a cluster on which disaster recovery is already configured, ensure that you configure ECA on the primary cluster.
If ECA is deployed on the primary cluster before adding the secondary cluster, then you must redeploy ECA from the primary cluster after disaster recovery configuration is complete. This is to ensure proper connectivity between the primary server, media server, and storage services.
If CRL mode is selected as CRL URL during ECA deployment, ensure that the CRL URL host name is resolvable by the existing DNS servers. If there are no DNS servers or if the DNS server cannot resolve the CRL URL host name, you must add the CRL URL as a custom host entry for the NetBackup container and the cluster nodes. This is also applicable if a DNS server is present during ECA deployment but is removed later.
If you do not want to generate CSR from the GUI, then you can use your own certificate for ECA deployment. In such a scenario, you must upload your own unencrypted private key.
If ECA is configured with the CRL as an URL, and if the CRL server becomes unreachable or unavailable for more than 24 hours for any reason, the NetBackup services on the NetBackup Flex Scale cluster appears as degraded. Once the connectivity to the CRL server is established again, the NetBackup services appear as healthy.
Considerations while deploying ECA on a cluster on which only media server is deployed:
There are some additional considerations that you need to keep in mind when you deploy ECA on a media server only cluster.
If you have deployed media server only clusters with external NetBackup primary server on BYO:
If ECA deployment is done after media server only configuration:
The primary server should be configured in ECA + NBCA mode before starting ECA deployment on the cluster.
The CA chain (Root + Intermediate ) used should be same trusted certificate chain for both primary and media server only cluster.
If media server only deployment is done after ECA configuration on NetBackup BYO:
Pure ECA mode is not supported.
If the primary server is deployed in NBCA + ECA mode then media server can be deployed using it and ECA can be configured on media server only cluster.
The CA chain (Root + Intermediate ) used should be same trusted certificate chain for both primary and media server only cluster.
If you have deployed media server only cluster with external NetBackup primary server in a NetBackup Flex Scale cluster:
If ECA deployment is done after media server only configuration:
Primary server should be configured in ECA + NBCA mode before starting ECA deployment on the media server only cluster.
This can be done using the NetBackup Flex Scale ECA deployment workflow.
The CA chain (Root + Intermediate ) used should be same trusted certificate chain for the cluster on which both primary and media servers are deployed and media server only cluster.
If media server only deployment is done after ECA configuration on a NetBackup Flex Scale cluster on which both primary and media server are deployed
Pure ECA mode is not supported a NetBackup Flex Scale cluster on which both primary and media server are deployed.
If the cluster is deployed in NBCA +ECA mode, then media server only cluster can be deployed using it and ECA can be configured on media server only cluster.
The CA chain (Root + Intermediate ) used should be same trusted certificate chain for the cluster on which both primary and media servers are deployed and media server only cluster.