DB2: When TSA is showing up in db2diag log

database cannot start because:


$ db2start

02/21/2025 15:31:05     0   0   SQL1042C  An unexpected system error occurred.

SQL1032N  No start database manager command was issued.  SQLSTATE=57019


In the db2diag log, there is:
(1)
2025-02-21-11.24.31.528593-360 I13219E400            LEVEL: Event
PID     : 1224844              TID : 140199623386880 PROC : db2sysc 0
INSTANCE: dev1ins1             NODE : 000
HOSTNAME: db2tst602
EDUID   : 1                    EDUNAME: db2sysc 0
FUNCTION: DB2 UDB, fast comm manager, sqkfNodeManager::initNodeManager, probe:9
DATA #1 : <preformatted>
Is Cluster Manager enabled? true or false = 1


(2)
2025-02-21-11.24.32.758682-360 E15369E777            LEVEL: Error
PID     : 1224844              TID : 140199623386880 PROC : db2sysc 0
INSTANCE: dev1ins1             NODE : 000
HOSTNAME: db2tst602
EDUID   : 1                    EDUNAME: db2sysc 0
FUNCTION: DB2 UDB, high avail services, sqlhaOnlineClusterObject, probe:6703
RETCODE : ECF=0x90000552=-1879046830=ECF_SQLHA_OBJECT_DOES_NOT_EXIST
          Cluster object does not exist
DATA #1 : Cluster object, PD_TYPE_SQLHA_CLUSTER_OBJECT_INFO, 136224 bytes
clusterObjectName: db2_dev1ins1_0-rs
objState         : Unknown
objType          : HA_CLUSTER_RESOURCE_DB2_T
option           : 0
generic.instanceName   : dev1ins1
generic.num            : 0
generic.controlResource:
generic.parentGroup    :
generic.hostname       :



(3)
2025-02-21-11.24.32.758959-360 I16147E606            LEVEL: Severe
PID     : 1224844              TID : 140199623386880 PROC : db2sysc 0
INSTANCE: dev1ins1             NODE : 000
HOSTNAME: db2tst602
EDUID   : 1                    EDUNAME: db2sysc 0
FUNCTION: DB2 UDB, base sys utilities, sqleCMResourceGroupUnlock, probe:19193
MESSAGE : ZRC=0x827300CF=-2106392369=HA_ZRC_OBJECT_DOES_NOT_EXIST
          "Cluster object does not exist"
DATA #1 : String, 53 bytes
db2nodes.cfg is inconsistent with the cluster manager
DATA #2 : String, 8 bytes
dev1ins1
DATA #3 : Member ID, PD_TYPE_SQLZ_MEMBER_ID, 2 bytes
0

2025-02-21-11.24.32.759108-360 I16754E465            LEVEL: Severe
PID     : 1224844              TID : 140199623386880 PROC : db2sysc 0
INSTANCE: dev1ins1             NODE : 000
HOSTNAME: db2tst602
EDUID   : 1                    EDUNAME: db2sysc 0
FUNCTION: DB2 UDB, base sys utilities, sqleSysCtlr, probe:4592
MESSAGE : ZRC=0xFFFFFABF=-1345
          SQL1345N  Execution failed due to a cluster manager error.

DATA #1 : <preformatted>
db2sysc abrupt termination


DATA #1 : <preformatted>
System Controller is exiting with RC =-1345, Erorr=-1345





$ db2 get dbm cfg | more

          Database Manager Configuration

     Node type = Enterprise Server Edition with local and remote clients

 Database manager configuration release level            = 0x1500

 CPU speed (millisec/instruction)             (CPUSPEED) = 9.446887e-08
 Communications bandwidth (MB/sec)      (COMM_BANDWIDTH) = 1.000000e+02

 Max number of concurrently active databases     (NUMDB) = 32
 Federated Database System Support           (FEDERATED) = NO
 Transaction processor monitor name        (TP_MON_NAME) =

 Default charge-back account           (DFT_ACCOUNT_STR) =

 Java Development Kit installation path       (JDK_PATH) = /usr/opt/app/db2/admin/dev1ins1/sqllib/java/jdk64

 Diagnostic error capture level              (DIAGLEVEL) = 3
 Notify Level                              (NOTIFYLEVEL) = 3
 Diagnostic data directory path               (DIAGPATH) = /usr/opt/app/db2/admin/dev1ins1/sqllib/db2dump/ $m
 Current member resolved DIAGPATH                        = /usr/opt/app/db2/admin/dev1ins1/sqllib/db2dump/DIAG0000/
 Alternate diagnostic data directory path (ALT_DIAGPATH) =
 Current member resolved ALT_DIAGPATH                    =
 Size of rotating db2diag & notify logs (MB)  (DIAGSIZE) = 0

 Default database monitor switches
   Buffer pool                         (DFT_MON_BUFPOOL) = OFF
   Lock                                   (DFT_MON_LOCK) = OFF
   Sort                                   (DFT_MON_SORT) = OFF
   Statement                              (DFT_MON_STMT) = OFF
   Table                                 (DFT_MON_TABLE) = OFF
   Timestamp                         (DFT_MON_TIMESTAMP) = ON
   Unit of work                            (DFT_MON_UOW) = OFF
 Monitor health of instance and databases   (HEALTH_MON) = OFF

 SYSADM group name                        (SYSADM_GROUP) = DB2IADM1
 SYSCTRL group name                      (SYSCTRL_GROUP) =
 SYSMAINT group name                    (SYSMAINT_GROUP) =
 SYSMON group name                        (SYSMON_GROUP) =

 Client Userid-Password Plugin          (CLNT_PW_PLUGIN) =
 Client Kerberos Plugin                (CLNT_KRB_PLUGIN) =
 Group Plugin                             (GROUP_PLUGIN) =
 GSS Plugin for Local Authorization    (LOCAL_GSSPLUGIN) =
 Server Plugin Mode                    (SRV_PLUGIN_MODE) = UNFENCED
 Server List of GSS Plugins      (SRVCON_GSSPLUGIN_LIST) =
 Server Userid-Password Plugin        (SRVCON_PW_PLUGIN) =
 Server Connection Authentication          (SRVCON_AUTH) = NOT_SPECIFIED
 Cluster manager                                         = TSA



Run this command to clean up:
$ db2 update dbm cfg using cluster_mgr NULL
DB20000I  The UPDATE DATABASE MANAGER CONFIGURATION command completed
successfully.



$ db2 get dbm cfg | more
[...]
should look like htis:
 Server Connection Authentication          (SRVCON_AUTH) = NOT_SPECIFIED
 Cluster manager                                         =



Now database can start successfully:


$ db2start

02/21/2025 15:48:51     0   0   SQL1063N  DB2START processing was successful.

SQL1063N  DB2START processing was successful.


Comments

Post a Comment

Popular posts from this blog

Oracle: To clean up WRI$_ADV_OBJECTS

 set linesize 200; select segment_name, segment_type, bytes from dba_segments where tablespace_name='SYSAUX' order by bytes asc; COL OCCUPANT_NAME FORMAT A30 SELECT OCCUPANT_NAME,SPACE_USAGE_KBYTES FROM V$SYSAUX_OCCUPANTS ORDER BY SPACE_USAGE_KBYTES DESC; COL SEGMENT_NAME FORMAT A30 COL OWNER FORMAT A10 COL TABLESPACE_NAME FORMAT A10 COL SEGMENT_TYPE FORMAT A15 SELECT * FROM (SELECT SEGMENT_NAME,OWNER,TABLESPACE_NAME,BYTES/1024/1024 "SIZE(MB)",SEGMENT_TYPE FROM DBA_SEGMENTS WHERE TABLESPACE_NAME='SYSAUX' ORDER BY BYTES DESC) WHERE ROWNUM<=10; 1.  automatic clean-up default 31 days           SELECT dbms_stats.get_stats_history_availability from Dual; 2. modify retention policy        exec  dbms_stats.alter_stats_history_retention (14); 3. a manual cleaning  exec dbms_stats.purge_stats (sysdate-14);   ALTER TABLE WRI$_ADV_OBJECTS MOVE; ALTER INDEX WRI$_ADV_OBJECTS_IDX_01 REBUILD; ALTER INDEX WRI$_ADV_OBJECTS_IDX_02 R...
Read more