Debugging Oracle Solaris Cluster HA for Alliance Gateway

How to Set the Debug Flag for Oracle Solaris Cluster HA for Alliance Gateway

Oracle Solaris Cluster HA for Alliance Gateway has an option file in /opt/SUNWscsag/etc directory that enables you to set a debug flag.

Normally, the debug information for Oracle Solaris Cluster software does not reside in the log files. You must edit the syslog.conf file to enable logging of messages of level debug. You can edit the syslog.conf to log those messages in another file.

1. Edit the /etc/syslog.conf file.
   1. Change daemon.notice to daemon.debug.

      The following output from the grep daemon /etc/syslog.conf command shows that daemon.debug has not been set.

      grep daemon /etc/syslog.conf
      *.err;kern.debug;daemon.notice;mail.crit
      /var/adm/messages
      *.alert;kern.err;daemon.err
      operator

   2. Restart syslogd.

      # pkill -1 syslogd

      The syslogd command forces syslog to reread its configuration file and account for changes.

      The following output from the grep daemon /etc/syslog.conf command shows that daemon.debug has been set.

      grep daemon /etc/syslog.conf
      *.err;kern.debug;daemon.debug;mail.crit
      /var/adm/messages
      *.alert;kern.err;daemon.err
      operator

2. Edit the /opt/SUNWscsag/etc/config file.

   Change the line DEBUG= to DEBUG=ALL or DEBUG=resource.

How to Debug a Failed Restart for the Alliance Gateway Resource

If the Alliance Gateway instance failed to restart on a cluster node or zone, but the instance did successfully fail over to another cluster node or zone, perform the following steps on the node or zone where the restart failed.

1. Log in as the Alliance Gateway UNIX user swnet.
2. Use the ipcs(1) command to determine if there are active inter-process communication (IPC) facilities for the swnet user.

   $ ipcs -opqs | grep swnet

3. If you see output for the command you ran in Step 2, verify that there is no Alliance Gateway instance running on this node or zone.

   To clean up the IPC resource from the swnet user, either use the ipcrm(1) command for each facility for the swnet user, or reboot the node or zone.

   You can switch the resource group with the Alliance Gateway instance back to this node or zone only if there are no IPC resources left for the swnet user.