2 gigabytes of available RAM should be sufficient for most environments. More memory is helpful.
The Reload server only supports the Intel Platform x-86 Platform
The faster the better; SMP is not to be considered a requirement, however, testing of Reload on SMP has shown that Reload is benefited by SMP.
Disk space must be carefully considered for each GroupWise Disaster Recovery Server. When the Reload server reaches the error threshold for disk space, it will no longer create backups. Make sure the target system has plenty of space to accommodate the needs of your Reload backups.
Reload backs up GroupWise post offices and domains according to profiles that the administrator specifies, which tells Reload what to backup, where it is located, and where to store the data. The Profile can also keep track of disk space and when to expire the backups it creates. After the initial backup, each successive backup takes an average of 12% the full size of the live post office, also saving network and disk resources.
Disk speed directly impacts Reload performance. A disk that is tuned for fast writes to disk is highly recommended and will improve performance. When considering disk space requirements, also take into account future growth of GroupWise post offices, and the space that will be required to sustain that growth.
If Reload runs out of hard disk space, backups will no longer be created. A Reload profile has a threshold of days to keep in hot backups. This can help prevent systems from running out of disk space.
This backup method in with the Reload Server gets a client connection via a Linux mount point to the post office or domain to be backed up. Reload would then copy data from the live server to the Reload server via the mount point it had established to the live server housing GroupWise. Prior to Reload 5, this was the only method of performing backups. In this scenario, the Reload Server should be in close network proximity to the GroupWise servers that are being backed up with Reload. If a Reload Server and some GroupWise post offices are on a Gigabit switch, it would be best to have a network card in the Reload server that supports a Gigabit speed.
This backup method is new to Reload 5. A “Collector” runs on the server that houses a GroupWise post office or domain. The Collector then uses Rsync as the transport to get data from the GroupWise server to the Reload Server. This model requires the following disk space on the GroupWise server:
2 times the size of the post office OFUSER, OFMSG and GWDMS directories. But not including the OFUSER/INDEX directory.
So for example, if the OFUSER directory contents was 1.5 GB and the OFMSG directory was 2.5 GB and the GWDMS directory was 1 GB the total space needed to service the Collector would be 1.5+2.5+1 = 5 X 2 = 10 GB needed somewhere on the GroupWise server to sustain the Reload Collector.
The disk space may reside on the Reload server or on a SAN. The requirements of the disk solution are as follows:
The disk solution should always be mounted for use with Reload.
The disk solution should be formatted with a Linux based file system, specifically a file system that supports symbolic links.
If the disk is not a SAN, but a NAS solution for example, the mount should be an NFS mount. No other protocol other than NFS.
The rule of thumb is 2.5 times the post office size per profile.
The Reload server requires the following disk space per profile:
Initial backup: 100% of the size of the post office.
Then 14 days of backups (on average) will be retained: 150% of the size of the post office.
Remember to take Post Office growth and the host Operating system into account while calculating disk space needs.
For example: A 100 gigabyte post office, for example, would require approximately 250 GB of disk space to retain 14 days worth of backups.