Description

This article explains how to reboot the backup Routing Engine in Junos OS Evolved.

Symptoms

An engineer needs to reboot the backup Routing Engine in the PTX10008 chassis that is running Junos OS Evolved (20.4R2-EVO).

Solution

In classical Junos OS, the solution would be: 

  1. request system power-off other-routing-engine
  2. request chassis cb slot ${BACKUP} offline
  3. request chassis cb slot ${BACKUP} online

However, the above shown commands do not work in Junos OS Evolved. This is the place where the "node" knob comes to play.

System versus Nodes:

  • In Junos OS, a system would mean the Routing Engine itself.

  • In EVO, the system is a collaboration of all nodes, such as the Routing Engine and the Flexible PIC Controller (FPC), in the cluster. Given this, in order to reboot/halt/offline/online a specific node (backup RE), you need to use the node command knob.

To see all active nodes on a PTX10008 device with two FPCs: 

labroot@jtac-re0-re0> show system nodes
Node: fpc0
  Node Id    : 2201170739216
  Node Nonce : 1203107635
  Status     : online, apps-ready
  Attributes : ASICS (Active), BT (Active), FABRIC_PFE (Active), FPC (Active), PIC (Active)

Node: fpc1
  Node Id    : 2201170739217
  Node Nonce : 1653087638
  Status     : online, apps-ready
  Attributes : ASICS (Active), BT (Active), FABRIC_PFE (Active), FPC (Active), PIC (Active), TIMINGD_FPC (Active)

Node: re0
  Node Id    : 2201170739204
  Node Nonce : 2663466812
  Status     : online, apps-ready
  Attributes : FABRIC_CONTROL (Active), FABRIC_FCHIP_PARALLEL (Active), RE (Active), TIMINGD_RE (Active), MasterRE (Active), GlobalIPOwner (Active)

Node: re1
  Node Id    : 2201170739205
  Node Nonce : 264169259
  Status     : online, apps-ready
  Attributes : FABRIC_CONTROL (Spare), FABRIC_FCHIP_PARALLEL (Spare), RE (Spare), TIMINGD_RE (Spare), BackupRE (Active)

To reboot a specific node, for example, the backup RE: 

labroot@jtac-re0-re0> request node reboot re1
                                                          
labroot@jtac-re0-re0> request node reboot ?
Possible completions:
  fpc0
  fpc1
  re0
  re1
  at                   Time at which to perform the operation in hh::mm 24 hrs format
  disk1                Boot off disk1(primary) device
  disk2                Boot off disk2(backup) device
  in                   Number of minutes to delay before operation
  message              Message to display to all users
  network1             Network boot via PXE over port1
  usb                  Boot off USB device

If you need to reboot all the nodes, the right command is request system reboot , which reboots the whole PTX10008 chassis.

Related Information