HP ProLiant DL380G5-WSS 3.7.0 HP StorageWorks HP Scalable NAS File Serving Sof - Page 250
The sentinel node, Cluster requirements, Change required for SLES10, Replication tuning
View all HP ProLiant DL380G5-WSS manuals
Add to My Manuals
Save this manual to your list of manuals |
Page 250 highlights
The sentinel node The nodes on the source cluster that are involved with replication are called the replication group. One node from the replication group coordinates replication activities such as performing the file transfers. This node is called the sentinel node. When you configure replication, you can specify the node that you prefer to use as the sentinel. Occasionally when the replication group is formed and replication starts, a different node may actually become the sentinel node. This can happen for a variety of reasons; for example, the preferred node may have been down at the time the replication group was formed. You can use replication commands to determine which node is currently acting as the sentinel node and to force the preferred sentinel to become the actual sentinel node. Cluster requirements The source and destination clusters must both be running the same operating system and HP Scalable NAS 3.7.0 (or higher). The source cluster must be able to reach the destination cluster via ssh. Change required for SLES10 The LSB package provided with the SLES10 operating system must be removed before using replication. This package conflicts with replication operations and causes replication to fail. Use the following command to remove the package: rpm -e lsb Replication tuning Replication substantially increases the work load for the system that is the replication source; not only does it have to create the data to be replicated, that data must be noted as needing to be replicated, and the system must read the data to be replicated from disk and transmit it to the destination cluster. This places a higher load on the systems participating in replication, especially the sentinel node, which is responsible for sending all of the filesystem changes for all nodes to the destination cluster. For best performance and to ensure the best possible replication success, systems must be tuned for replication. Replication impacts the system most heavily in the following areas: 250 Configure and manage replication