VPLEX: How to collect director logs manually

           

   Article Number:     537206                                   Article Version: 2     Article Type:    How To 
   

 


Product:

 

VPLEX VS2,VPLEX VS6,VPLEX GeoSynchrony 5.5 Patch 1,VPLEX GeoSynchrony 5.5 Service Pack 1,VPLEX GeoSynchrony 5.5 Service Pack 1 Patch 1,VPLEX GeoSynchrony 5.5 Service Pack 1 Patch 2,VPLEX GeoSynchrony 5.5 Service Pack 2

 

Instructions:

 

 

   

      Issue:     
      Collect Diagnostics were collected from the VPLEX, yet the director logs are missing from the logs or     
      only the director logs are needed.     
          

   

      Purpose:   

   
         
  1.         How to collect the director logs when they are not collected with the VPLEX Collect-Diagnostics or when only the director logs are needed.     
  2.      
  3.         How to tar the director logs once they have been collected     
  4.    
   

     
      Instructions:   

   

      1] Access the VPLEX management server for the cluster you will be collecting the director logs for.     
     
      2] ssh as root to the director(s) you need to collect the logs from.     
              
           Example:     
           service@ManagementServer:~> ssh root@128.221.252.35     
           Last login: Fri Sep 27 22:15:48 2019 from 128.221.252.33     
           VPLEX     
           service@director-1-1-a:~ #     
     
      3] Change directory (cd) to the /var/log directory of the director you are on.     
     
           Example:     
           service@director-1-1-a:~ # cd /var/log     
           service@director-1-1-a:/var/log #     
     
      4] Next "tar" all the files under the /var/log directory by running the command shown below,     
          Note: for the purpose of the example in this article we will use dir1-1-a as the director name.   

   

           tar cvzfp <directorname.tar.gz> *     

   

           (the asterisk is a wildcard to copy everything)     
     
           Example:       
             
director-1-1-a:/var/log # tar cvzfp dir1-1-a.tar.gz *     
           acpid     
           boot.msg     
           boot.omsg     
           cimomlog.txt     
           cimomlog.txt.0     
           cimomlog.txt.1     
           ...     
           director-1-1-a:/var/log #     
     
      5] To check that the tar file is there run the long list, ' ll ', command as follows,     
            ll *.tar*     
     
            Example:     
            director-1-1-a:/var/log # ll *.tar*     
            -rw-r--r-- 1 root root 87375791 Oct 10 15:41 dir1-1-a.tar.gz     
     
      6] Next you will need to SCP the tar.gz file up to the management server's  /tmp directory from the     
          director as follows, be sure to use the internal IP (eth2 ) address for the mgmt server you are on,     
          cluster-1, 128.221.252.33 or cluster-2, 128.221.252.65. You will be prompted for the service     
          account password before the file can be transferred up.     
     
          run scp <name.tar.gz> service@128.221.252.65:/tmp/.     
     
          Example:     
          director-1-1-a:/var/log # scp dir1-1-a.tar.gz service@128.221.252.65:/tmp/.     
          Password:     
          dir1-1-a.tar.gz                                                                         100%   83MB  41.7MB/s   00:02     
          director-1-1-a:/var/log #     
     
      7] Once the director /var/logs files have been tar'd and scp'd to the management server, on the same     
          director, cd to the /core directory     
     
           Example:     
           director-1-1-a:/var/log # cd /cores     
           director-1-1-a:/cores #     
     
      8] Run the ' ll ' command to see what files are in this directory     
     
          Example:     
          director-1-1-a:/cores # ll     
          total 65428     
          -rw------- 1 root root  1222336 May  2 14:37 fast_trace     
          drwx------ 2 root root    16384 Jun  1  2015 lost+found     
          drwxr-xr-x 2 root root     4096 Feb 28  2018 nsfw-trace     
          -rw------- 1 root root 64909936 Sep 27 01:54 tracebufferDumpFast     
          -rw------- 1 root root    96216 Sep 27 01:55 tracebufferDumpSlow     
          -rw------- 1 root root   655416 Sep 27 01:56 tracebufferDumpUdcom     
     
      9] Repeat the steps 4, 5 and 6, only name the tar.gz file <directorname>_cores.     
     
      10] Once the steps above have been done for all directors the files were needed from, open     
             WinSCP or FileZilla and copy the files from the management server to your computer, then     
             upload to the logserver to the customer folder.   

   

      11] Once the logs are on the logserver, in the customer folder, get the URLs by using     
            'vplex_scrip_menu' option 6 then copy the URLs and paste them into the attachment upload     
             field in the opened JIRA and update the JIRA to let engineering know the requested director     
             logs have been uploaded to the JIRA ticket.     
     
      12] Once all the needed tar.gz files from each director they were requested for have been attached to     
             the appropriate JIRA go back to each director(s) you tar'd files under the /var/log and /cores     
             directories and remove them to conserve space on t he partitions on each director.     
     
             Example:     
             for the /var/logs directory:     
             rm dir1-1-a.tar.gz   

   

             for the /cores directory:     
             rm dir1-1-a_cores.tar.gz