Probing Algorithm and Functionality

• Probing steps for Samba.

  • Sleeps for Thorough_probe_interval.

  • Tries to connect to the Samba instance fault monitor share (scmondir), using smbclient, as user <fmuser>. If this fails, then the probe will restart the Samba resource.

  • The smbclient uses the NetBIOS name of the Samba server to verify that the NetBIOS name server works. If this fails, then the probe will restart the Samba resource.

  • If all Samba processes have died, pmf will interrupt the probe to immediately restart the Samba resource.

  • If the Samba resource is repeatedly restarted and exhausts the Retry_count within the Retry_interval, then a failover is initiated for the Resource Group onto another node.

• Probeing steps for Winbind.

  • Sleeps for Thorough_probe_interval.

  • Retrieves the defined <ntuser> by using getent passwd <ntuser>. If this fails, then the probe will restart the Winbind resource.

  • If the Winbind resource is repeatedly restarted and exhausts the Retry_count within the Retry_interval, then a failover is initiated for the Resource Group onto another node.

  ---

  Note –

  If nscd is on (default) and the defined <ntuser> is cached by the nscd daemon, then getent will return a success until the user has been removed from the cache, even if winbind itself isn't working.

  ---