Cisco Headend System Release 2.7 Technical References

 

 

Operations Alert Bulletin 
Working with Cisco Services to 
Recover From a DBDS System 
Failure 

When a Digital Broadband Delivery System (DBDS) failure occurs, your Informix 

database may be at risk of becoming corrupted, even if your facility uses a 

standby Digital Network Control System (DNCS). This Operations Alert Bulletin 

explains the steps you should take after a DBDS system failure. 

 
 

If you experience a DBDS system failure involving a Solaris panic or an Informix 

assertion failure, you should contact Cisco Services for assistance. Cisco engineers 

will defragment your tables and rebuild your indexes using the dncsDbData and 

formatDbSpace.sh utilities. If these recovery steps are not performed, you may 

experience other database failures in the future. 
Solaris panics are logged in the /var/adm/messages file, and Informix assertion 

failures are logged in the export/home/informix/online.log. To determine 

whether you should contact Cisco Services, look for messages in the 

/export/home/informix/online.log stating that Informix performed work during 

fast recovery.  

The following example indicates that your tables do not need to be defragmented. 

Messages indicate that no work was needed or performed during the recovery. 

This example shows what you will see upon startup after a graceful shutdown. 

15:16:36  Physical Recovery Started. 
15:16:36  Physical Recovery Complete: 0 Pages Restored. 
 
15:16:36  Logical Recovery Started. 
15:16:36  20 recovery worker threads will be started. 
15:16:39  Logical Recovery Complete. 
          0 Committed, 0 Rolled Back, 0 Open, 0 Bad Locks 

In this case, you should restart the system and verify that there are no other 

problems, then check the log files once again for errors.