Locking Conflicts panel
This panel shows existing Db2 locking contentions. It provides information about the owners and waiters currently involved in locking conflicts.
A locking conflict exists when a Db2 thread owning a resource causes another Db2 thread to wait. In general, when Db2 cannot satisfy a request for the resource, the requester will be suspended.
A suspended lock resource always has an owning thread. Multiple threads may be waiting for the same resource. It may also have one or more global owners that have global interest in the resource.
When S mode is used in data sharing environment, lock resources that are waited by a local thread will be displayed. All global owners and all local threads that are involved in this locking conflict are displayed.
When G mode is used in data sharing environment, lock resources that are waited by a local or remote thread will be displayed. All the global owners and all local and remote threads that are involved in the locking conflict are displayed.
The plan names of all threads for which the elapsed time exception has tripped are highlighted. For more information about current thread exceptions, type E.A on the top line of the panel to display the Exception Messages for Thread Exceptions panel.
________________ ZLOCKC VTM O2 V550.#P SAD G 11/05/13 14:00:19 2
> Help PF1 Back PF3 Up PF7 Down PF8 Zoom PF11
> L.
===============================================================================
> LOCKING CONFLICTS
XLOK
+Stat Plan Corrid Type Lvl Resource
+---- -------- ------------ ---- --- ------------------------------------------
+Owned by Subsystem SAC ROW DB=DSNDB04 PS=RLOCKTSG PG=00000002/01
+Owned by Subsystem SAD ROW DB=DSNDB04 PS=RLOCKTSG PG=00000002/01
+ OWN DSNESPCS MIS ROW X DB=DSNDB04 PS=RLOCKTSG PG=00000002/01
+WAIT DSNESPCS MIS ROW U DB=DSNDB04 PS=RLOCKTSG PG=00000002/01
===============================================================================
Highlighting
Field | Exception | Reason |
---|---|---|
Plan | ETIM | Elapsed time is high. |
Navigation
- A lock owner, move the cursor to the lock owner line and press F11 (Zoom).
- A lock waiter, move the cursor to the lock waiter line and press F11 (Zoom).
- Other topics, use the PF keys.
Fields
The output lines are displayed in descending order, with those reflecting the greatest degree of conflict displayed first.
- Stat
- The status of the thread can be one of the following:
- OWN indicates that the thread is a lock owner.
- WAIT indicates that the thread is suspended because the lock it requested is owned by another thread.
- Plan
- The name of the Db2 plan that is currently active in this thread.
- Corrid
- The Db2 correlation identifier for the thread involved in the locking conflict.
- Type
- The type of lock that is owned or requested and is unavailable.
- Lvl
- The level at which the lock is owned or was requested. The level of a lock requested is frequently the reason for lock suspensions.
- Resource
- Identifies the resource that is currently locked
by the lock owner, or the lock resource causing the lock to be suspended.
The resource varies depending upon the type of lock held. Note: DB=database name can be represented as a DBID=identifier, which is the decimal identifier of the database. PSID=identifier can be represented as PSID=identifier, which is the decimal identifier of the table space. HEX values are displayed for lock resources if the resource is owned by a thread on a remote LPAR, and no OMPE task is available from that LPAR, or XCF is not set up correctly. If the owner is a thread running on some other member of the data sharing group, the Db2 subsystem name is displayed.This field displays the following resources:
- Lock Type
- Resource
- ACSC
- Accelerator Services commands lock
- ALBP
- Alter buffer pool lock
- BIND
- Autobind/remote bind lock
- BMBA
- Buffer manager SCA MBA L-lock
- BPPS
- Buffer manager pageset RP P-lock
- CCAT
- CATMAINT convert catalog lock
- CDBL
- Compress dictionary build lock
- CDIR
- CATMAINT convert directory lock
- CDRN
- Cursor Stability drain lock
- CMDS
- Db2 Command Serialization lock
- CMIG
- CATMAINT migration lock
- COLL
- Collection lock
- DBEX
- Database exception LPL/GRECP lock
- DBXU
- Database exception update lock
- DGTT
- DGTT URID lock
- DPAG
- Db2 pageset page lock
- DPLK
- 32 K Desperation Pool lock
- DSET
- Partitioned lock
- DTBS
- Database lock
- GRBP
- Group buffer pool start/stop lock
- HASH
- Hash anchor lock
- IEOF
- Index end of file lock
- IPAG
- Index page lock
- LBLK
- LOB lock
- LPLR
- LPL Recovery lock
- MDEL
- Mass delete lock
- PALK
- Partition lock
- PBPC
- Group buffer pool castout P-lock
- PCDB
- DDF CDB P-lock
- PDBD
- DBD P-lock
- PDSO
- Pageset or partition pageset open lock
- PITR
- Index manager tree P-lock
- PPAG
- Page P-lock
- PPSC
- Pageset/partition level castout P-lock
- PPSP
- Pageset/partition P-lock
- PRLF
- RLF P-lock
- PSET
- Pageset lock
- PSPI
- Pageset piece lock
- RDBD
- Repair DBD Test/Diagnoes lock
- RDRN
- Repeatable Read drain lock
- RGDA
- Retry Getpg During Abort
- ROW
- Row lock
- RSTR
- SCA access for restart/redo lock
- SDBA
- Start/stop lock on DBA tabs
- SENV
- SYSENV serialization lock
- SKCT
- Skeleton cursor table lock
- SKPT
- Skeleton package table lock
- SPRC
- Sys Level PIT Recovery lock
- SREC
- Log range lock
- TABL
- Table lock
- UIDA
- Util I/O Damage Assessment
- UNDT
- Undetermined lock
- UTEX
- Utility exclusive execution lock
- UTID
- Utility UID lock
- UTSE
- Utility serialization lock
- WDRN
- Write drain lock
- XMLK
- XML lock
- UTOB
- Utility object lock
- DBDL
- DBD load lock
- IXKY
- Index key lock
- HPSP
- Header page bucket or Stored Proc Cmd lock