Description

This article provides troubleshooting steps if device discovery fails for a supported Junos Space device.

Symptoms

In Junos Space, the following error is reported:

Status: 'Add Device failed'.
Description: Device is not running Junos or is not a supported Junos device

alt
 

In the /var/log/jboss/servers/server1/server.log , the following error messages are reported:

WARN  [net.juniper.jmp.cmp.deviceIOMgr.DeviceInboundConnectionFactory] (Thread-122611 (HornetQ-client-global-threads-515124559)) Password authentication success for 10.x.x.13

WARN  [net.juniper.jmp.cmp.deviceIOMgr.DeviceInboundConnectionFactory] (Thread-122611 (HornetQ-client-global-threads-515124559)) Failed to start netconf subsystem. Trying to set netconf over ssh ......

ERROR [net.juniper.jmp.cmp.deviceIOMgr.DeviceInboundConnectionFactory] (Thread-122611 (HornetQ-client-global-threads-515124559)) Device inbound cmd 10.x.x.13 failed to request netconf subsystem

ERROR [net.juniper.jmp.cmp.deviceIOMgr.DeviceInboundCmd] (Thread-122611 (HornetQ-client-global-threads-515124559)) Send failure message for add reachable device with ip 10.x.x.13 failed
--- Device is not running Junos or is not a supported Junos device

Solution

Possible Causes:

  1. The user trying to discover the device does not have read-write permission on the device.
  2. The device has uncommitted changes.
  3. Device user does not have permission to configure netconf channel or the system services are protected.
  4. The device user performs a rollback every time the Junos Space device tries to commit the netconf services on the device.

Troubleshooting

  • Check the device user permission by which device has been discovered to Junos Space.​                                                                                                                  
    root@210-2# show system login

    It should have read-write permission or it should belong to “super-user”, as Junos Space sends some set of commands to the device to commit .  If the user does not have read-write permission, then the command will fail. This leads to device discovery failure.

  • Check if there are any uncommitted changes on the device. If there are, remove them and try to discover the device again.

  • Check the device configuration, if system service is protected:
    root@240-POE-2# run show configuration | match service | display set
    set system services ftp
    set system services ssh
    protect system services ssh

    If it is protected, then unprotect it with the following command:
    unprotect system services

    Now if you try to discover the device, it should get discovered successfully.

  • Under the device logs, you may see the following messages:

         sshd[28811]: subsystem request for netconf by user Junospace
         sshd[28811]: subsystem request for netconf failed, subsystem not found
         mgd[28818]: UI_DBASE_LOGIN_EVENT: User 'Junospace' entering configuration mode
         mgd[28818]: UI_COMMIT: User 'Junospace' requested 'commit' operation (comment: none)
         mgd[28769]: UI_DBASE_LOGOUT_EVENT: User 'Junospace' exiting configuration mode
         mgd[28818]: UI_LOAD_EVENT: User 'Junospace' is performing a 'rollback'

    along with the 'subsystem request failed on channel' message reported when a SSH command from the Space server to the device is performed:

         [root@space-005056a9cff7 ~]# ssh [email protected] -s netconf
         Password:
         subsystem request failed on channel 0

    If so, it means the user for whom you are trying to add the device to Junos Space performs a rollback when Junos Space tries to commit the netconf service on the device and hence the device discovery fails.

           To resolve this, set the netconf ssh on the device by running the command:
                     #set system services netconf ssh

             Then try to discover the device again.

If none of the above suggestions help, please contact your technical support representative for assistance.

Modification History

2019-09-19: Changed the path of the log file to /var/log/jboss/servers/server1/server.log which was incorrect previously. Also added the screenshots from latest version.