To control and monitor the cluster, the manager has several tools, three of which are illustrated as follows. Return to the simple cluster example with EAGLE as the boot server and file server and BEAVER, LOON, and WWW as satellites. To confirm the health of his or her system, the manager can use the following command (this is the primary command used to examine the status of a cluster):
$ SHOW CLUSTER View of Cluster from system ID 64518 node: LOON 4-AUG-2002 18:54:49 +-----------------------------+ ≠ SYSTEMS ≠ MEMBERS ≠ +-------------------+---------≠ ≠ NODE ≠ SOFTWARE ≠ STATUS ≠ +------- +----------+---------≠ ≠ LOON ≠ VMS V7.3 ≠ MEMBER ≠ ≠ BEAVER ≠ VMS V7.3 ≠ MEMBER ≠ ≠ EAGLE ≠ VMS V7.3 ≠ MEMBER ≠ ≠ WWW ≠ VMS V7.3 ≠ MEMBER ≠ +-----------------------------+
Notice that this display was created when logged into LOON, not EAGLE. SHOW CLUSTER is a rich command with several subcommands. Only three fields—NODE, SOFTWARE, and STATUS—are shown. There are nearly 100 fields available for display. SHOW CLUSTER can be run continuously or provide only a cluster snapshot. Both status and performance data can be displayed.
Another way to examine the performance of the cluster members is with the MONITOR CLUSTER command. The following display shows four performance components of the top six nodes in the cluster:
CPU utilization expressed in busy percentage
Memory utilization expressed in percentage
Disk I/O rate in operations per second
Resource lock rate in operations per second
$ MONITOR CLUSTER Statistic: CURRENT OpenVMS Monitor Utility 4-AUG-2002 18:55:37 CLUSTER STATISTICS CPU | MEMORY | CPU Busy 0 25 50 75 100|%Memory In Use 0 25 50 75 100 +----+----+----+----+| +----+----+----+---+ LOON 1 | |WWW 47 |≠≠≠≠≠≠≠≠≠ WWW 1 | |LOON 44 |≠≠≠≠≠≠≠≠ EAGLE 1 | |BEAVER 19 |≠≠≠ BEAVER | |EAGLE 14 |≠≠ | | | | | | -----------------------------------------+------------------------------------ DISK | LOCK | I/O Operation Rate 0 25 50 75 100|Tot ENQ/DEQ Rate 0 125 250 375 500 +----+----+----+----+| +-----+----+----+----+ LOON$DKA0: | |LOON | BEAVER$DKA0: R | |BEAVER | EAGLE$DKA0: R | |EAGLE | EAGLE$DKB0: R | |WWW | EAGLE$DKB100: R | | | WWW$DKA0: R | | |
This real-time display is updated every six seconds to show the dynamics of the cluster. MONITOR can display more than 20 other resources. MONITOR can also record and/or average these statistics.
The following SHOW DEVICE display shows the devices (in this case, only disks) served from node BEAVER. Mass Storage Control Protocol (MSCP) is used to remotely access disk files and Tape Mass Storage Control Protocol (TMSCP) is used to remotely access tape files.
MANAGER> show device/served/host MSCP-Served Devices on BEAVER 24-AUG-2002 12:04:37.22 Queue Requests Device: Status Total Size Current Max Hosts DKA0 Online 4110480 0 2 3 DKA100 Avail 4110480 0 0 0 Queue Requests Host: Time of Connection Current Max Devices LOON 25-JUN-2002 10:07:18.41 0 5 1 WWW 25-JUN-2002 10:20:26.05 0 4 1 EAGLE 13-AUG-2002 13:31:40.32 0 2 1 TMSCP-Served Devices on BEAVER 24-AUG-2002 12:04:37.28 Queue Requests Device: Status Position Current Max Hosts TMSCP-Server code not loaded