PRIMECLUSTER Installation and Administration Guide 4.1 (for Solaris(TM) Operating System)

C.1 Collecting Troubleshooting Information

If an error occurs in the PRIMECLUSTER system, collect the information required for the error investigation from all nodes that construct the cluster and the cluster management servers. Then, contact your customer support representative.

1. PRIMECLUSTER investigation information
   • Use fjsnap to collect information.
   • Retrieve the system dump.

     If a hardware error, an operating system error, a panic, or a login failure occurs, collect the followings:

     • System dump
     • SCF* dump

       * SCF: System Control Facility

   • Collect the Java Console on the clients.

     See "Appendix B.2.2 Java console" in the "PRIMECLUSTER Web-Based Admin View Operation Guide."

   • Collect a hard copy of the client screens.

     See "Appendix B.2.3 Screen hard copy" in the "PRIMECLUSTER Web-Based Admin View Operation Guide."

   • If a client is using RC2000, collect RC2000 information for the client.

     See "D.2 Collecting Troubleshooting Information" in the "RC2000 User's Guide."

2. Investigation information for the failed application
3. Error reproduction procedure description if the error can be reproduced

• When reporting a problem, collect the information required for an error investigation. If you do not provide information for problem checking and error reproduction execution, it may take a long time to reproduce and diagnose the problem or it may become impossible to do so.
• Collect investigation material promptly from all nodes of the PRIMECLUSTER system. Necessary information may become lost if a long time elapses after the error occurs. This applies especially to information collected by fjsnap.
• If a node was forcibly terminated, execute sync in OBP mode, and collect a system dump.

---

C.1.1 Executing the fjsnap Command

C.1.2 System dump

C.1.3 SCF Dump