PRIMECLUSTER Global File Services Configuration and Administration Guide 4.5

C.2.2 Collecting a core image of the daemon

It may be necessary to collect a core image of each daemon according to the corrective action of the GFS Shared File System.

The following procedure shows how to collect a core image of the sfcfrmd daemon on all the nodes that share the GFS Shared File System using super user access privileges.

1. Determine the process ID

   Determine the process ID using ps(1). For non sfcfrmd daemons, change argument of grep(1).

   # /bin/ps -e | /bin/grep sfcfrmd <Enter>
     5639 ?        00:00:25 sfcfrmd

   The beginning of the above output shows a process ID of the sfcfrmd daemon. If the sfcfrmd daemon is not running, the process number will not be output. In such a case, collect troubleshooting information on the other node.

   See

   For details about ps(1) and grep(1), see the online manual page.

2. Acquire a core image

   Acquire the core image of sfcfrmd in the "/var/tmp/sfcfrmd_node1.5639" file using gcore(1). After that, compress the file using tar(1).

   # /usr/bin/gcore -o /var/tmp/sfcfrmd_node1 5639 <Enter>
     :
   Saved corefile /var/tmp/sfcfrmd_node1.5639
   # /bin/tar czvf /var/tmp/sfcfrmd_node1.5639.tar.gz /var/tmp/sfcfrmd_node1.5639 <Enter>
     :
   /var/tmp/sfcfrmd_node1.5639
   # /bin/ls -l /var/tmp/sfcfrmd_node1.5639.tar.gz <Enter>
   -rw-rw-r--   1 root     other    1075577  Jun 12  16:30 /var/tmp/sfcfrmd_node1.5639.tar.gz
   #

   See

   For details about tar(1), see the online manual page.