GPFS daemon communication

In a cluster environment, the GPFS daemon depends on the correct operation of TCP/IP.

These dependencies exist because:
  • The communication path between nodes must be built at the first attempt to communicate.
  • Each node in the cluster is required to communicate with the cluster manager and the file system manager during startup and mount processing.
  • Once a connection is established, it must remain active until the GPFS daemon is shut down on the nodes.
Note: Establishing other communication paths depends upon application usage among nodes.

The daemon also uses sockets to communicate with other instances of the file system on other nodes. Specifically, the daemon on each node communicates with the file system manager for allocation of logs, allocation segments, and quotas, as well as for various recovery and configuration flows. GPFS requires an active internode communications path between all nodes in a cluster for locking, metadata coordination, administration commands, and other internal functions. The existence of this path is necessary for the correct operation of GPFS. The instance of the GPFS daemon on a node goes down if it senses that this communication is not available to it. If communication is not available to another node, then one of the two nodes exit GPFS.