Source:How to create a Remote Device Server (RDS)
Overview
SEP sesam Remote Device Server (RDS) is a storage management component that controls the preparation of data, intended for backup on a SEP sesam Client and writes the backup data to the backup media. During a restore it locates the savesets and sends the data to a client.
RDS consists of three components: Sesam Transfer Protocol Server (STPD), Sesam Multiplex Stream Server (SMS), and SEP sesam Client (SBC) including remote access. The control of the tasks is maintained by the SEP sesam Server. It is available for installation at SEP Download Center as a separate – RDS installation package.
Key features
If your network spans multiple locations, you can administer storage devices across locations using a SEP sesam Server (e.g., removed tape libraries or SAN devices.) However, if your infrastructure spans several sites that do not allow for fast data transfer to the central SEP sesam Server, you should use a Remote Device Server to back up data to locally attached storage at a remote location. RDS enables efficient data transfer, takes the workload off the primary SEP sesam Server and uses the storage resources available at the site.
Hence, at remote locations RDS acts as a backup server and can either serve as a backup proxy to deliver data to the main backup server or save data to locally attached storage. By using RDS you can easily and and conveniently manage many remote locations from one central console.
License
RDS requires a separate license. The RDS license includes one backup client and updates for 12 months. To learn more about SEP sesam licensing options, see Licensing.
Configuration and launch of SEP sesam RDS: Basic Scenario
This scenario describes how to set up an RDS on Linux. The configuration assumes two sites with the following situation:
- Site A (Munich): SEP sesam Server with NAS storage and virtual disk images
- Site B (Chemnitz): 5 clients are going to be backed up via RDS to NAS storage (virtual disk images) that is connected in Chemnitz
All clients and devices are managed centrally from the site in Munich.
In this scenario, it is assumed that both systems are running on Debian GNU\Linux (Lenny) and that you have already performed all installation-related preparations. If you need help with these requirements, refer to SEP sesam Basics Administrator (SBA).
Installing SEP sesam Server and RDS
In this scenario, you will install the following components at the respective sites:
- Munich: SEP sesam Server
- Chemnitz: SEP sesam RDS
Prerequisites
Make sure that the hardware and software requirements are met. For details, see What are the hardware and software requirements for a SEP sesam Server. In addition to these requirements, the following prerequisites must also be fulfilled:
- Before starting the SEP sesam installation, make sure that you are logged in as a local administrator or domain administrator.
- To install SEP sesam Server or Remote Device Server on Linux, you will require an installation file for Linux, which can be downloaded from SEP Download Center. The installation files are names as sesam_srv-<version> (SEP sesam Server) and sesam_rts-<version> (SEP sesam RDS). Make sure to download the correct file for your processor type.
To install SEP sesam on Windows, you will require an installation file for Windows, which can be downloaded from SEP Download Center. SEP sesam for Windows provides all SEP sesam components in one package. During installation, you have to specify which of the four SEP sesam components you want to install (SEP sesam Server, SEP sesam Remote Device, SEP sesam GUI or SEP sesam Client). You can install GUI together with RDS. For details on installation, see SEP sesam Quick Install Guide. - GUI requires Java Runtime Environment (JRE) to be installed on the system. For details on the required Java version, see Java Compatibility Matrix. For 64-bit operating systems, a PostgreSQL database must be installed.
- SEP sesam uses network resolution for server to client communication. Before installing, you should test DNS name resolution. For details, see How to check DNS configuration.
- Ensure that any used SCSI devices are recognized by the operating system to which you are installing SEP sesam. SEP sesam checks the SCSI bus attached storage devices during the installation and adds its data to the database. SEP sesam can only see devices recognized by the operating system.
- It is recommended to disable firewall to avoid problems during the SEP sesam installation. Once SEP sesam is installed, you can enable the firewall with exceptions made for the SEP sesam services.
- On Windows, the .Net Framework 4 is required for server installation and can be deselected for all other SEP sesam components during installation.
- On Windows, x86 operating systems with more than 3.25 GB RAM must either reduce the amount of RAM to below to 3.25 GB or migrate to an x64 operating system. This is necessary because SEP sesam requires 64 KB blocks for LTO (Linear Tape Open) whereas an x86 system can only write 32 KB blocks to tape drive because the PAE (Physical Address Extension) is automatically activated. Also, a loader cannot be accessed properly by SEP sesam.
Note | |
On RHEL, the SEP sesam installation changes the permissions of /var/run/postgresql to grant SEP sesam users PostgreSQL access privileges. |
Sample Linux installation
Munich: Installing SEP sesam Server root@muenchen# apt-get install sesam-srv Reading package lists... Done Building dependency tree Reading state information... Done The following NEW packages will be installed: sesam-srv 0 upgraded, 1 newly installed, 0 to remove and 293 |
Chemnitz: Installing SEP sesam RDS root@chemnitz:~# apt-get install sesam-rts Reading package lists... Done Building dependency tree Reading state information... Done The following NEW packages will be installed: sesam-rts 0 upgraded, 1 newly installed, 0 to remove and 25 |
Configuring RDS
To configure a SEP sesam RDS, you have to create a location, used for logical grouping of SEP sesam clients, including RDS. Each client is always created within (and assigned to) a location, which can be defined as a group of clients and further specified as sub-locations. In our example, we will create a new location named Chemnitz and then assign the RDS with the same name Chemnitz to it. For details, see Configuring Clients.
Prerequisites
- Make sure that the DNS names are correctly resolved (forward and reverse DNS lookup). For details, see How to check DNS configuration.
- Adjust the sm_ctrld.auth on the RDS system. For details, see FAQ: What problems may occur when adding a new client?
Embedding the SEP sesam RDS
Create a new location as follows:
- In the Main selection -> Components -> Clients, go to the content pane and click New Location. Note that the location LOCAL is always created by default.
- In the New Location window, enter the name Chemnitz.
- Click OK. Your new location is displayed in the Clients content pane.
Now add the Chemnitz RDS as a client to your newly created location:
- In the Clients content pane, right-click Chemnitz and then click New Client. Enter the name of the RDS – Chemnitz. The name has to be resolvable either by DNS or hosts file. If there is a name server (DNS) in the network, SEP sesam automatically selects the TCP/IP address. Otherwise you have to enter it manually in the etc/hosts file.
- From the Platform drop-down list, select the relevant platform, in our example, LINUX. Note that by selecting the LINUX platform, the LINUX operating system is selected automatically.
- From the Access mode drop-down list, leave the default SMSSH (default) access mode.
- Click OK. A new dialog box appears, asking whether you want to create a new backup job for this client. Click No. Your new client Chemnitz is displayed in the Clients content pane.
Note | |
Once your RDS is added as a client, check the access to get a proper identification entry in the sesam database. |
Adding storage hardware to RDS
A Remote Device Server can be used with attached disk storage, tape libraries or single tape drives. In case of a disk storage, you can configure different data stores, e.g., default data store Path, SEP Si3 Deduplication Store, etc. Depending on the type of storage you want to configure, select one of the following:
Backup on the Remote Device Server
Generally speaking, all data of the clients in the Chemnitz location (backed up onto media in the media pool PoolChemnitz) is only moving on the net segment of that particular site. Therefore, no data is transported to the SEP sesam Server over WAN.
To test this, set up a test backup task in the SEP sesam GUI (Main Selection -> Tasks -> By Clients-> New Backup Task) that will perform a quick self-backup of the SEP sesam RDS (directory /etc) to the storage mounted on RDS. For details, see Creating a Backup Task.
Once you have configured a test backup, start it:
- In the Main Selection -> Tasks -> By Clients, right-click the test backup task and select Immediate Start.
- In the Immediate Start:Backup window, select the Media pool referring to the data store you have previously configured on RDS, in our example, PoolChemnitz and click Start.
Tip | |
You can view the status of your backup jobs in the SEP sesam GUI -> Main Selection -> Last Backup State. As of 4.4.3 Beefalo V2, you can also check the details of your backups online by using new Web UI. For details, see SEP sesam Web UI. |
You can also schedule your test backup. For details, see Creating a Schedule.
Now your RDS is configured and ready to use.
Note | |
Make sure that you follow the general recommendations and configure required exclusions with your antivirus product as well as ensure that the antivirus scans are not scheduled during backup operations. |