This article explains how to manually collect logs from the nodes or virtual machines (VMs) on QFabric and upload them to the Director.

You can use this method in the following scenarios:

• The /root/createpblogs.sh script does not collect logs from the backup VMs, such as the backup NW-NG VM and so on.

• If logs from any node in the problem state cannot be successfully collected via the script, then this manual procedure will help.

To manually collect logs from the nodes or VMs on QFabric and upload them to the Director:

1. Select the node or VMs from which the logs need to be collected:

   root@Qfabric-G-Static> show fabric administration inventory
   Item                    Identifier              Connection      Configuration
   Node group
     NW-NG-0                                       Connected       Configured
       D04-34              BBAK8063                Connected
       D04-35              BBAK3177                Connected
     RSNG-1                                        Connected       Configured
       D04-30              P7339-C                 Connected
       D04-31              BBAK1186                Connected
     RSNG-2                                        Connected       Configured
       D04-32              TA3716160105            Connected
       D04-33              VF3716090515            Connected
   Interconnect device
     IC-F1052                                      Connected       Configured
       F1052/RE0                                   Connected
       F1052/RE1                                   Connected
     IC-F3947                                      Connected       Configured
       F3947/RE0                                   Connected
       F3947/RE1                                   Connected
   Fabric manager
     FM-0                                          Connected       Configured
   Fabric control
     FC-0                                          Connected       Configured
     FC-1                                          Connected       Configured
   

2. In this example, logs are collected from RSNG-1. First, check the internal IP address of RSNG-1 by using dns.dump via the Director shell:

   [root@dg0 ~]# dns.dump | grep RSNG-1
   dcfnode-default-RSNG-1.pkg.dcbg.juniper.net. 45	IN A 169.254.193.11
   
   

3. Log in to RSNG-1 via SSH:

   [root@dg0 ~]# ssh 169.254.193.11
   Password:
   --- JUNOS 14.1X53-D17.1 built 2015-07-26 23:34:29 UTC
   
   
   
   root@RSNG-1%
   
   

4. Collect the logs by using the following command:

   root@RSNG-1> file archive compress source /var/log/* destination /var/tmp/RSNG-1_logs.tgz
   /usr/bin/tar: Removing leading `/' from member names

5. Log out from RSNG-1 and use SCP from DG shell:

   [root@dg0 ~]# scp [email protected]:/var/tmp/RSNG-1_logs.tgz /var/tmp/
   Password:
   RSNG-1_logs.tgz                                                                                                                                                           100%   10MB   3.4MB/s   00:03
   [root@dg0 ~]#
   
   [root@dg0 ~]# ls -lt /var/tmp
   total 15380
   -rw------- 1 root root 15726960 Dec 21 16:40 RSNG-1_logs.tgz
   [root@dg0 ~]#

6. Perform the SCP to the Director via any desktop or server on the management network and download the logs (by default, SCP is enabled on the Director).