PRIMECLUSTER Global File Services Configuration and Administration Guide 4.1 (Solaris(TM) 10 Operating System)

C.2 Collecting information for troubleshooting

Collect following material for investigation when you request troubleshooting to local Customer Support and if your manual instructs you to do so.

1. Material acquired in /opt/FJSVsnap/bin/fjsnap -a filename (Need to be executed with root access privilege)
2. /var/crash/*/unix.*, /var/crash/*/vmcore.*
3. Output of following command: "fsck -F sfcfs -N -o nolog raw_device" * When file system restoration is necessary
4. Output of pkginfo

If the investigation material must urgently be sent, prepare the followings at least. This material is a part of material collected by fjsnap and required for the initial investigation of the GFS Shared File System.

1. Material collected by /etc/opt/FJSVgfs/bin/sfcsnap.sh -a -d dirname (Collect by executing with a root access privilege)
2. /var/adm/messages*
3. Output of pkginfo

• All material above should be collected from all nodes where the GFS Shared File System is shared.
• For above 3, collect troubleshooting information required to recover a file system. For raw_device, specify a character special file of the file system where the problem occurs.
• This material should be collected as soon as the symptom occurred. Some material might not be collected after rebooting the system.
• Specify the file name that does not exist for filename that is specified with fjsnap.
• For the directory dirname designated in sfcsnap.sh, specify the existing directory. Material is stored in the sub-directory.
• a and b information is of initialized investigation. Collect information from 1 to 3 after collecting it.

For information on how to collect troubleshooting information on PRIMECLUSTER systems, see the "PRIMECLUSTER Installation and Administration Guide (for Solaris Operating System)".

---

C.2.1 Collecting a crash dump file

C.2.2 Collecting a core image of the daemon