Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

Navigation
Guide That Contains This Content
[+] Expand All
[-] Collapse All

    compute-node

    Syntax

    compute-node compute-node-name {interfaces {bridge bridge-name {family inet {(address ip-address | dhcp);}interface bridge-interface-list;management;}ethernet ethernet-interface-name {enable-passthrough;family inet {(address ip-address | dhcp);}management;}}mac-address mac-address;monitor-cpu normal threshold;monitor-memory normal threshold;monitor-storage normal threshold;syslog {facility level;}}

    Hierarchy Level

    [edit services app-engine compute-cluster compute-cluster-name]

    Release Information

    Statement introduced in Junos OS Release 12.3.

    Option monitor-cpu normal threshold introduced in Junos OS Release 13.2R2.

    Option monitor-memory normal threshold introduced in Junos OS Release 13.2R2.

    Option monitor-storage normal threshold introduced in Junos OS Release 13.2R2.

    Description

    Create a compute node. A compute node is the interface with the VSE device that runs the host operating system (host OS). A compute node is a member of a compute cluster. Each compute node requires that you configure network interfaces and the media access control (MAC) address of the compute node that you want to create. Configuring the syslog statement is optional.

    When you commit the compute node configuration, the compute node is booted. Compute nodes bootstrap using network boot, or PXE. For instructions on how to enable PXE and complete setting up the compute node for applications, see Configuring the DHCP Server and TFTP Server for Compute Nodes.

    The values for monitor-cpu, monitor-memory, and monitor-storage can be set at any of these hierarchy levels:

    • [edit services app-engine]
    • [edit services app-engine compute-cluster compute-cluster-name]
    • [edit services app-engine compute-cluster compute-cluster-name compute-node compute-node-name]

    Settings at the [edit services app-engine compute-cluster compute-cluster-name] hierarchy level override those at the [edit services app-engine] hierarchy level and settings at the [edit services app-engine compute-cluster compute-cluster-name compute-node compute-node-name] hierarchy level override those at any other level.

    Options

    compute-node-name

    Name of the compute node. When you commit the compute-node name, the compute node is booted with that name as the operational node name. Any time you change the compute-node name in the CLI, the compute node needs to be rebooted.

    mac-address

    MAC address of the network boot interface of the compute node. The MAC address is essential so that the compute node can PXE-boot, or boot through the network interface, rather than from the local hard disk. The compute node identifies itself to the router with the network boot interface mac-address.

    monitor-cpu normal threshold

    Normal and maximum processing loads as percentages of the CPU. If the overall system CPU usage or per-CPU usage exceeds the threshold percentage, the system raises the jnxVECNProcessingLoadHigh or jnxVECNProcessorLoadHigh trap. When the system detects a jnxVECNProcessingLoadHigh or jnxVECNProcessorLoadHigh condition and then drops to or below the normal percentage, the jxnVECNProdessingLoadOk or jnxVECNProcessorLoadOk alarm is triggered.

    Default: normal default is 80; threshold default is 90

    monitor-memory normal threshold

    Normal and maximum RAM usages as percentages. If the RAM usage rises above the threshold percentage, the system raises the jnxVECNMemoryLow trap. When the RAM usage rises above the maximum and then drops below the normal percentage, the jnxVECNMemoryOk alarm is triggered.

    Default: normal default is 80; threshold default is 90

    monitor-storage normal threshold

    Normal and maximum storage usages as percentages. If the space usage under volume LFS (/volume/lfs) or volume SFS (/volume/sfs) rises above the threshold percentage, the system raises the jnxVECNStorageLow trap. When the storage usage recovers and drops below the normal percentage, the jnxVECNStorageOk alarm is triggered.

    Default: normal default is 80; threshold default is 90

    The remaining statements are explained separately.

    Required Privilege Level

    admin—To view this statement in the configuration.

    admin-control—To add this statement to the configuration.

    Published: 2014-03-31