IC69170: NEW DBSPACE IS NOT USED IMMEDIATELY

APAR status

• Closed as documentation error.

Error description

• Once a database location has been added to the Tivoli Storage
  Manager V6 Server, it may not be fully utilized immediately.
  Certain events with Tivoli Storage Manager will trigger the
  added database space to become increasingly used over time.
  For example, table reorganization, or certain temporary database
  transactions such as long running select statements may
  lead to start filling the added database space up.
  
  The DB space redistribution among all extends will need a few
  days or even weeks.
  If the existing DB extend is already near to full when the new
  extend is added, the Server may encounter instability due to
  lack of space reported in db2diag.log like filesystem (being)
  full.
  This is because DB2 still did not have the time and ability to
  spread the DB data, and some dynamic space allocation was not
  yet possible for temporary table space.
  
  It is possible to immediately spread the database on all
  containers by removing the DB and restoring it to the same or
  new containers.
  The method documented in the administration guide chapter
  "Moving only the database" can be used to do so.
  Infocenter link =>
  http://publib.boulder.ibm.com/infocenter/tsminfo/v6/topic/com.ib
  m.itsm.srv.doc/t_move_db_only.html
  
  This must be better documented in Tivoli Storage Manager
  manuals.
  
  
  Tivoli Storage Manager Versions Affected:
  6.1  , 6.2
  
  Customer/L2 Diagnostics (If Applicable)
  
  - Error from db2diag.log
  
  FUNCTION: DB2 UDB, buffer pool services, sqlbWritePageToDisk,
  probe:2
  MESSAGE : ADM6017E  The table space "LGTMPTSP" (ID "7") is full.
  Detected on  container
  "/tsm/server/db/tsm1/NODE0000/TSMDB1/T0000007/C0000000.TMP"
  (ID "0").  The underlying file system is full or the maximum
  allowed space usage for the file system has been reached. It
  is also possible that there are user limits in place with
  respect to maximum file size  and these limits have been
  reached.
  
  
  Initial Impact: Medium
  
  Additional Keywords:
  Container , tablespace, zz61, zz62 , extend dbspace
  

Local fix

• Use remove/restore method to spread db in all containers.
  Steps:
   - Take a full database backup
   - Stop  Tivoli Storage Manager server
   - Copy current volumehistory and devconfig files to a safe
     location
   - Issue Admin Command "Dsmserv removedb tsmdb1"
   - Verify that all db containers are empty
   - Create a file that list the db target containers, these can
     be the same or new locations.
       (in the following example it is file c:\db.txt)
   - Issue Admin Command "Dsmserv restore db on=c:\db.txt"
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All Tivoli Storage Manager server version    *
  *                 6.1 and 6.2 users.                           *
  ****************************************************************
  * PROBLEM DESCRIPTION: If you add database space to the        *
  *                      Tivoli Storage Manager server,          *
  *                      it might not be used immediately.       *
  *                      Depending on database activity,         *
  *                      it could take a long time to            *
  *                      distribute the database space.          *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  The following text will be added to the Administrator's Guides
  ("Increasing the size of the database" topic) and the
  Administrator's References ("EXTEND DBSPACE" topic)
  
  "After a  new directory is added to a Tivoli Storage Manager
  version 6 server, the directory might not be fully used
  immediately. Some Tivoli Storage Manager events can cause
  the added database space to become increasingly used over
  time. For example, table reorganizations or some temporary
  database transactions, such as long running select statements,
  can cause the added database space to begin filling up.
  The database space redistribution among all directories
  can require a few days or even weeks. If the existing
  database directories are nearly full when the new directory
  is added, the server might encounter an out of space condition
  (reported in db2diag.log).  If this condition occurs, halt and
  restart the server. If the restart does not correct the
  condition,  remove the database and then restore it to the
  same or new directories. This procedure is documented in the
  "Moving only the database" section  of the Administrator's
  Guide.
  Infocenter link => http://publib.boulder.ibm.com/infocenter/
  tsminfo/v6/topic/com.ibm.itsm.srv.doc/t_move_db_only.html."
  

Problem conclusion

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

Applicable component levels